Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning? 9783836624053, 9783836674058

Citation preview

Alice Klever

Copyright © 2009. Diplomica Verlag. All rights reserved.

Behavioural Targeting: An Online Analysis for Efficient Media Planning?

Diplomica Verlag

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Alice Klever Behavioural Targeting: An Online Analysis for Efficient Media Planning? ISBN: 978-3-8366-2405-3 Herstellung: Diplomica® Verlag GmbH, Hamburg, 2009

Copyright © 2009. Diplomica Verlag. All rights reserved.

Dieses Werk ist urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funksendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen und der Speicherung in Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Eine Vervielfältigung dieses Werkes oder von Teilen dieses Werkes ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland in der jeweils geltenden Fassung zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtes. Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in diesem Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme, dass solche Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei zu betrachten wären und daher von jedermann benutzt werden dürften. Die Informationen in diesem Werk wurden mit Sorgfalt erarbeitet. Dennoch können Fehler nicht vollständig ausgeschlossen werden und der Verlag, die Autoren oder Übersetzer übernehmen keine juristische Verantwortung oder irgendeine Haftung für evtl. verbliebene fehlerhafte Angaben und deren Folgen. © Diplomica Verlag GmbH http://www.diplomica-verlag.de, Hamburg 2009

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Table of Contents 1.

Introduction.................................................................................................................5

2.

Media planning ...........................................................................................................7

3.

4.

5.

2.1.

Definition ...............................................................................................................7

2.2.

Classification of media planning into marketing planning......................................7

2.3.

Objectives..............................................................................................................8

2.4.

Target group segmentation .................................................................................11

2.4.1.

Segmentation criteria ...................................................................................11

2.4.2.

Target group segmentation online ...............................................................13

2.5.

Media strategy.....................................................................................................15

2.6.

Media selection programmes ..............................................................................17

2.7.

Requirements for optimal media planning: Elements and process .....................20

Online as advertising medium ................................................................................22 3.1.

The importance of online advertising ..................................................................22

3.2.

Brief overview of the history of online advertising ...............................................24

3.3.

The current state of the industry..........................................................................25

3.4.

Online advertising formats and performance measurement ...............................27

Targeting Online .......................................................................................................30 4.1.

Geo Targeting .....................................................................................................31

4.2.

Prime-Time Targeting..........................................................................................31

4.3.

Technical Targeting.............................................................................................32

4.4.

Profile Targeting ..................................................................................................32

4.5.

Contextual Targeting ...........................................................................................32

4.6.

Keyword Targeting ..............................................................................................33

Behavioural Targeting..............................................................................................34

Copyright © 2009. Diplomica Verlag. All rights reserved.

5.1.

Three different types of Behavioural Targeting ...................................................35

5.1.1.

Affinity-based Behavioural Targeting ...........................................................35

5.1.2.

Re-targeting .................................................................................................36

5.1.3.

Predictive Behavioural Targeting .................................................................37

5.2.

Behavioural Targeting in practice........................................................................38

5.3.

Requirements for successful Behavioural Targeting...........................................41

5.4.

The process of BT ...............................................................................................42

5.5.

A Behavioural Targeting SWOT-Analysis ...........................................................44

5.5.1.

Strengths......................................................................................................44

5.5.2.

Weaknesses.................................................................................................46 3

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Copyright © 2009. Diplomica Verlag. All rights reserved.

6.

5.5.3.

Threats .........................................................................................................48

5.5.4.

Opportunities................................................................................................50

Current state of affairs in BT ...................................................................................53 6.1.

The international development of Behavioural Targeting....................................53

6.2.

Demands on Behavioural Targeting and recommended course of action for media planners....................................................................................................55

6.3.

Case Study..........................................................................................................57

6.4.

Future outlook .....................................................................................................61

7.

Summary and final remark ......................................................................................66

8.

Bibliography..............................................................................................................69 8.1.

Books ..................................................................................................................69

8.2.

Articles in collected editions ................................................................................71

8.3.

Journal articles ....................................................................................................71

8.4.

Internet ................................................................................................................72

8.5.

Miscellaneous......................................................................................................76

8.6.

List of figures .......................................................................................................77

4

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

1. Introduction Consumers today are in the focus of a wide range of companies in a more and more global world, all of which implies an increase of marketing activities and hence an increase of advertising effort. Hundreds or even thousands of advertising messages are put in front of consumers every day through various media channels. However, every targeted customer has a very limited ability of perceiving advertising messages and even a smaller percentage of what has been recognized is considered as being relevant. 1 Therefore consumers quite often feel harassed and overloaded by the mass of information, the quality of advertising and the context in which it is shown, and react protective towards advertising, especially when they do not belong to the target group of the advertised product or service. 2 This implies a great amount of budget wastage and a potential damage to brand images, since too many consumers are targeted with the wrong advertising message. Furthermore media consumption has changed over the years, while the consumption of TV for instance remains stable, the internet consumption increases dramatically, in the younger target groups it has already outgrown the TV consumption, and therefore an attenuation of this trend can be expected. 3 This means more and more prospective consumers will spend more and more time online and are therefore important potential targets for advertising campaigns. Keeping in mind the negative consequences of the irrelevant and too intense advertising the question arises if there is a way to avoid the drawbacks of offline media and use the opportunities offered by online media. The necessity of defining the right target group for an advertising campaign is as undoubted as the difficulties implied in that process, the end product being able to target the right persons online. Behavioural targeting might offer a way to analyse online media and therefore to contribute to the requirements for optimal media planning in order to achieve a high degree of efficiency.

Copyright © 2009. Diplomica Verlag. All rights reserved.

Against the background of the increasing interest in behavioural targeting an examination of its potential importance for marketing will be conducted in this paper in order to asses if behavioural targeting can increase the efficiency of media planning in online media.

1

Cp. Hamm, Ingo: Internet-Werbung – Von der strategischen Konzeption zum erfolgreichen Auftritt, Stuttgart 2000, Schäffer-Poeschel Verlag, p. 45 2 Cp. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=6281 / 10.09.2008 3 Cp. EIAA, in: http://www.eiaa.net/news/eiaa-articles-details.asp?id=154&lang=1 / 10.09.2008

5

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

The foundation for this paper is provided by the description of media planning in general, its objectives including the target group segmentation, media strategy and selection programmes and concluding by summarizing the requirements for optimal media planning. In order to systematize the subject area of behavioural targeting an insight into online advertising is provided. This study aims to preserve an international scope, however due to the fact that the actual maturity of the online media industry is at very diverse stages worldwide, focus will be on the U.S. and selected countries in Europe such as Germany, France, the UK and Scandinavia. The focus of this study is online display advertising which comprises both branding and performance advertising but excludes search, meaning the use of text links on web pages. Behavioural targeting is one of several online targeting methods; therefore the different principles of targeting are reviewed in order to frame behavioural targeting in this context. Three major approaches to behavioural targeting exist: ‘Affinity-based’, ‘Predictive’ and ‘Re-targeting’. These will be examined along with a model process of applying behavioural targeting. A behavioural targeting SWOT- Analysis frames the current merits and shortcomings of the methods, and a presentation of the current state of affairs gives an overview of the current maturity state of the technology in practice. Based on the behavioural targeting SWOT-Analysis and concluding an future outlook shall be dared and furthermore a summary with a final remark on the importance of behavioural

Copyright © 2009. Diplomica Verlag. All rights reserved.

targeting regarding media planning.

6

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

2. Media planning In order to evaluate if behavioural targeting can form the basis for efficient media planning, the theoretical foundations of this discipline need to be explained. This includes the objectives and the requirements for optimal media planning as well as a description of existing media selection programmes. Furthermore an evaluation of the importance of the medium Internet compared to traditional media channels is provided since it constitutes the basis for behavioural targeting itself and its relevance for media planning.

2.1. Definition “Within the communication-mix media planning deals with the decisions concerning the applied media vehicles, with the selection of media which best meet the aimed marketing objectives and the identification of the optimal media-mix (…).” 4 The prime objective of media planning is a highly efficient way of conveying the advertising message to the right target group with optimal timing. Therefore channels of media need to be drafted, analyzed and selected in order to meet the defined communication objectives.

5 6

2.2. Classification of media planning into marketing planning As a result media planning is part of the communication-mix. The table below shows the different media channels, which is one element of marketing planning next to the product-, distribution- and price-mix. Those elements add up to the marketing-mix, which is “a combined and coordinated assignment of the marketing-instruments with the objective of reaching the enterprise and marketing targets as efficient as possible in a harmonious

Copyright © 2009. Diplomica Verlag. All rights reserved.

adaptation of the instrument’s specifications.” 7

4

Koschnik, Wolfgang J.: Standard-Lexikon für Mediaplanung und Mediaforschung in Deutschland, Band 2, LZ, 2. edition, München 1995, Saur Verlag, p. 1230 (Translation by the author) 5 Cp. Unger, Fritz/Durante, Nadja/Gabrys, Enrico/Koch, Rüdiger/Wailersbacher, Rainer: Mediaplanung – Methodische Grundlagen und praktische Anwendungen, 5. edition, Berlin Heidelberg 2007, Springer Verlag, p. 2 6 Cp. Arens, William/Weigold, Michael F./Arens, Christian: Contemporary advertising, 11. edition, Boston 2008, McGraw-Hill/Irwin, p. 271 7 Bruhn, Manfred: Kommunikationspolitik – Systematischer Einsatz der Kommunikation für Unternehmen, 4. edition, München 2007, Verlag Franz Vahlen, p. 11 (Translation by the author)

7

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Marketing planning contains various sub-mix category groups, which add-up to the marketing-mix and the table below demonstrates the challenge to find the appropriate coordination. 8

Product

Price

Communication

• innovation • improvement • differentiation • marketing • naming • service • selection • packaging

• price • rebates • incentives and cash discounts • conditions of delivery • terms of sale

Product-mix

Price-mix

• media advertisements • sales promotion • direct marketing • public relations • sponsoring • personal communication • exhibitions • event marketing • multimedia communication • employee communication

Communication-mix

Distribution • distribution system • elements of sale • logistics system

Distribution-mix

Marketing-mix

Figure 1: Based on Bruhn (2007): Classic marketing instruments (4Ps) in the marketing-mix.

9

2.3. Objectives Media objectives are derived from the advertising strategy and should define what the media planning has to accomplish. Media objectives have two components, the audience objectives and the message-distribution objectives 10 , meaning to “reach a defined percentage of the target group (therewith the reach of the media plan is addressed) with a certain advertising impact (therewith the frequency is addressed with which persons within the target group should be reached)” 11 Reach and advertising impact or frequency respectively are competing objectives owing mainly to budget reasons. A high reach Copyright © 2009. Diplomica Verlag. All rights reserved.

normally does not allow a high frequency and vice versa, so either the reach or the frequency is too low which symbolizes one dilemma in media planning. 12 This dilemma is addressed with the Gross Rating Points (GRP) which means a multiplication of reach with

8

Cp. Koschnik, 1995, p. 1230-1231 Cp. Bruhn, 2007, p. 11 10 Cp. Arens/Weigold, 2008, p. 280 11 Unger, 2007, p. 5 (Translation by the author) 12 Unger, 2007, p. 9

8

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

frequency. Next to other objectives, these three are the most common in media planning and should be explained further. 13 Audience objectives These address the definition of the target group by the media planner and will be discussed in depth under section 2.4. However, it can be said that the customer types an advertiser wants to reach are not congruent with the target groups a marketing manager wants to reach, since the advertising targeting group is usually rather more focused than the more comprehensive marketing target group, therefore a conversion from the marketing target group to the media target group is necessary. 14

15

Message distribution objectives 1. Reach “The percentage of the target audience exposed to an advertisement in a given time period” 16 Reach refers to the total number of people or households covered by the campaign, meaning they are exposed to the advertisement, but it does not necessarily imply that they also consume it 17 . They simply have the

18

“Opportunities to see

(OTS):The cumulative exposures achieved in a given time period, usually used in media planning in place of frequency.” 19 A certain percentage of reach or maximized reach represent one of the most important objectives of media planning. 2. Frequency “The number of times an individual member of the target group is exposed to an advertisement in a given time period” 20 Frequency goes hand in hand with reach and is very important because the more often a target person is exposed to an advertisement the better this person remembers it. A specific degree of repetition or frequency respectively is another important objective. 3. Message weight Through application of reach and frequency the message weight can be determined and

Copyright © 2009. Diplomica Verlag. All rights reserved.

therefore the optimal level for the campaign spending can be identified. The message

13

Cp. Tellis, Gerard J./Ambler, Tim: The SAGE Handbook of Advertising, Los Angeles 2007, SAGE Publications, p. 308 14 Cp. Löffler, Jens: Media – Planung für Märkte, 7. Edition, Hamburg 2004, Axel Springer Verlag, p. 19 15 Cp. Arens/Weigold, 2008, p. 280 16 Percy, Larry/Elliott, Richard: Strategic Advertising Management, 2. Edition, New York 2005, Oxford Press, p. 165 17 Cp. Tellis/Ambler, 2007, p. 301-302 18 Cp. Arens/Weigold/Arens, 2008, p. 283 19 Percy/Elliott, 2005, p. 165 20 Percy/Elliott, 2005, p. 165

9

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

weight is expressed by the “Gross Rating points: A summary statistic for the achievement of a media plan, calculated by multiplying reach by frequency” 21 . Reach X Frequency = GRP, meaning the multiplication of the percentage rating of all households who had been exposed to the campaign with the number of the times the households had been exposed results in the message weight in percentage or the gross impressions. 22 Since media campaigns are normally on a tight budget it is of high importance to find the optimal message weight in order to avoid wastage. 4. Continuity Within a given time period of a campaign, continuity refers to the media schedule. There are three patterns of media continuity, 1. Continuous advertising: A constant schedule which is very rare, 2. Flighting: The media effort is varied, but there are also times where there are none at all and 3. Pulsing: A variation of media efforts with no interruption, typically with a “heavy up” before prime selling times and less spending in off-seasons or anti-cyclical to the sales curve. 23

24

As already mentioned frequency is important to create recall, but the pattern of continuity has to manage to sustain that recall. All three objectives reach, frequency and continuity are important and reciprocal. A limited budget makes it impossible to pursue all objectives to the same extent and every media planning situation is different. Therefore a great disunity in the advertising industry exists since there is no panacea for the determination of reach, frequency and continuity objectives. 25

26

Every campaign is planned in a unique situation, which creates uncertainty as how to set the media objectives; the only solution to that problem is a profound research phase and thoroughly balancing the results with the experience of the media planner. 5. Optimization Considering the definitions of the terms reach, frequency and continuity it becomes clear that they do not state anything about the quality of the exposure. This is described by the effective reach, which “measures the number or percentage of the audience who receive

Copyright © 2009. Diplomica Verlag. All rights reserved.

enough exposures to truly receive the message.” 27 The same concept accounts for the

21

Percy/Elliott, 2005, p. 165 Cp. Tellis/Ambler, 2007, p. 284 23 Cp. Arens/Weigold/Arens, 2008, p. 284-285 24 Cp. Barban, Arnold M./Cristol, Steven M./Kopec, Frank J.: Essentials of media planning, 2. edition, Chicago 1987, NTC Business books, p. 57-58 25 Cp. Arens/Weigold/Arens, 2008, p. 285 26 Cp. Barban/Cristol/Kopec, 1987, p. 61 27 Arens/Weigold/Arens, 2008, p. 285 22

10

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

effective frequency: “A hypothetical construct that attempts to estimate the number of OTS required to have an effect in the target audience” 28 Frequency has been stated as being the most important media objective, but the question remains how to analyze what effective frequency is, how often people have to be exposed to a campaign until it creates awareness and recall. Therefore a suggested solution is to evaluate the specific situation and set an individual optimum level of reach, frequency, message weight and continuity accordingly.

29

It becomes clear that there are no set standards in the advertising industry defining media objectives, therefore a strategy to achieve those objectives can only be created under uncertainty.

2.4. Target group segmentation In order to provide a foundation for the chapter, the following definition shall be the basis: “A target group is a circle of decision maker, purchaser and consumer to which marketing efforts are aimed at. In advertising these are groups of persons and/or institutions to which the advertising effort is aimed at in the form of physical approach (distribution) and sales approach and accordingly argumentation (presentation) in order to reach the advertising objectives most efficiently.”

30

Target group segmentation results from market segmentation, which has been conducted within the strategic marketing planning. Those segments represent the foundation of the target group segmentation but cannot be adopted without adaptation based on various segmentation criteria which will be discussed under point 3.2. 31

2.4.1. Segmentation criteria The objective of target group segmentation is a more appropriate adjustment of the marketing activities to the needs and the behaviour of the consumer. Therefore there are

Copyright © 2009. Diplomica Verlag. All rights reserved.

two groups of consumer characteristics: Some criteria allow an interpretation of future consumer behaviour (nr.1-4) and some are based on actual decision and information behaviour (nr. 5).

32

28

Percy/Elliott, 2005, p. 165 Arens/Weigold/Arens, 2008, p. 286 30 Pepels, Werner: Marktsegmentierung: Erfolgsnischen finden und besetzen, 2. Edition, 2007 Düsseldorf, Symposion Verlag, p. 264 (Translation by the author) 31 Cp. Bruhn, 2007, p. 192 32 Cp. Pepels, 2007, p. 268 29

11

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

1. Demographic segmentation Demographic segmentation is the best known criteria and covers aspects such as age, sex, family status and size of household. 2. Geographic segmentation Another simple way of approaching the target group definition is the geographic segmentation, which refers to the fact that purchasing behaviour varies from region to region. The combination of demographic and geographic segmentation is described as geodemographic segmentation. 3. Socioeconomic segmentation The socioeconomic criteria are characteristics such as education, occupation, income, purchase power, resources and its combination into social class. A clear separation between the criteria within the demographic, geographic and socioeconomic segmentation is hardly possible, therefore a reasonable combination needs to be found. The advantage of these three segmentation approaches is straightforwardness, since information is most often available from secondary sources. However, their disadvantage is the fact that the gathered information is too simple and does not allow any conclusion regarding the purchasing behaviour of the target group. This makes them misleading in the sense that they pretend the target persons within one segmentation criteria are alike in their way of purchasing. 33

34 35

4. Psychographic segmentation The criteria refer to general personality traits such as activities, interests and attitudes, furthermore expectations regarding the benefits, motives and purchase intentions and so forth. 36 Compared to the first three criteria the psychographic approach has a greater control mechanism regarding the creation of the advertising message since it refers to the feelings and disposition of the consumer and vice versa the psychographic criteria can be influenced by the advertising message. 37

38

5. Behavioural segmentation Examples of behavioural criteria are point of purchase as well as quantity and frequency, Copyright © 2009. Diplomica Verlag. All rights reserved.

mouth-to-mouth-communication and use of media behaviour, therefore these criteria are the result of advertising effort. The behaviour can be interpreted for future advertising campaigns and target group segmentation respectively and hence give an indication of how the target group might behave, of course without any validation. Some information for 33

Cp. Percy/Elliott, 2005, p. 70-72 Cp. Bruhn, 2007, p. 196-197 35 Cp. Arens/Weigold/Arens, 2008, p. 173-174 36 Cp. Arens/Weigold/Arens, 2008, p. 196 37 Cp. Bruhn, 2007, p. 197 38 Cp. Arens/Weigold/Arens, 2008, p. 177 34

12

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

the behavioural segmentation such as sales figures and so forth is easily available but data concerning the communicational behaviour of the consumer is rather difficult to obtain.

39 40

Considering the advantages and disadvantages of the individual segmentation approaches an isolated selection of segmentation criteria would lead to a misinterpretation and therefore false definition of the target group. Therefore a balanced combination of the described criteria is necessary in order to cover all consumer and their characteristics. This is done by so called typologies, which contain target group descriptions based on the various criteria and segmentation dimensions leading to better target group definitions and higher marketing efficiency. 41 The dilemma of the definition of the target group becomes obvious considering the wide range of criteria and sub-categories, especially when several target groups should be addressed.

2.4.2. Target group segmentation online The research regarding Internet segmentation is still in its infancy. Segmentation online was of no interest to media planners and advertisers before 1997, at that stage 76 million people worldwide used the internet 42 and were an homogenous group of middle- and upper-income men with a high education and interested in gaining information online. 43 This has drastically changed; in 2008 an estimated number of 1.5 billion people worldwide are internet users 44 as well as striking changes in the structure of the internet user occurred, “the typical online user does not exist (anymore)” 45 . The fast global growth of internet usage exacerbates the need to identify those users with similar characteristics and segment them for the purpose of maximizing advertising effectiveness and efficiency. In order reach an efficient spending of the budget, there are several objectives of the

Copyright © 2009. Diplomica Verlag. All rights reserved.

segmentation effort online:

39

Cp. Pepels, 2007, p. 268 Cp. Bruhn, 2007, p. 197 41 Cp. Deyle, Hanno-G.:Mit neuen Segmentierungsverfahren zu besseren Zielgruppendefinitionen, in: Diller, H.(ed.): Zielgruppen finden und überzeugen, 2006, Wiss. Ges. für Innovatives Marketing, p. 109 42 Cp. Barnett, Alex: http://blogs.msdn.com/alexbarn/archive/2005/10/08/478598.aspx/ 24.7.2008 43 Cp. Schumann, David W./Thorson, Esther: Internet advertising – Theory and reserach, New Jersey 2007, Erlbaum Associates Inc., p. 150 44 Cp. Internet Usage Statistics; http://www.internetworldstats.com/stats.htm/ 204.7. 2008 45 Fittkau, Susanne: Nutzer und Nutzung des Internets in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketing-Börse GmbH, p. 139 40

13

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

1. Identifying and characterizing heavy Internet users 2. Identifying likely targets for new technologies 3. Improving existing web site design and navigation 4. Focusing Internet advertising efforts for greater impact.

46

However, the approach for segmentation online is based on traditional segmentation variables such as socio-demographics, geographics and psychographics 47 as well as behavioural criteria, which are in the sense of Internet segmentation also divided into product/service-based variables (Internet usage and motives) and consumer-based variables (demographics and product-related attitudes). 48 Schumann/Thorson (2007) have carried out a survey in order to analyze the extent to which these variables commonly used in traditional media planning work within the context of the Internet. The overall finding was that it is strongly recommendable to combine traditional segmentation criteria in order to define target groups online.

49

Concluding from the survey it can be said that both product/service-based variables and consumer-based variables are useful for segmenting online and predicting online behaviour. Within predicting online behaviour the product/service-based variables, meaning the Internet usage and usage motives, are better for predicting the frequency for online shopping, while the consumer-based variables, meaning the product related attitudes are better to predict the attention to Internet advertising. Therefore a combination of variables is appropriate for online segmentation as well and a match of segmenting with outcome variables useful in order to increase the probability of the prediction. Since those findings are related only to one survey more research is necessary in order to identify useful ways for target group segmentation online as not all possible variables, for instance demographics, have been examined. Furthermore the relationship between Internet attitudes and behaviour has not been investigated very well yet.

50

Copyright © 2009. Diplomica Verlag. All rights reserved.

So although the current state of research is not very sophisticated yet, the approach of using traditional segmentation criteria in a combination seems to be common and useful in order to define target groups online.

46

Schumann/Thorson, 2007, p. 152 Cp. Bogner, Thomas: Strategisches Online Marketing, Wiesbaden 2006, Deutscher Universitäts-Verlag, p. 60 48 Cp. Schumann/Thorson, 2007, p. 153-156 49 Cp. Schumann/Thorson, 2007, p. 163-164 50 Cp. Schumann/Thorson, 2007, p. 170-179 47

14

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

The question arises how behavioural targeting identifies the relevant target groups and what kind of criteria is used in order to find a meaningful segmentation. In order to give an insight to the possibilities and limitations of behavioural targeting, it should be seen in the context of the other targeting methods since a combination of those is executed in the online advertising industry and quite common. The other targeting methods are of importance because they influence the efficiency of behavioural targeting and should therefore be explained in section 4.

2.5. Media strategy Once the media objectives have been set, a media strategy must be formulated in order to achieve the objectives. Therefore is the media strategy “the description of the ways and

Copyright © 2009. Diplomica Verlag. All rights reserved.

means with which the media objectives of an advertiser should be achieved.” 51

Figure 2: Based on Koschnik (1995): Criteria for media selection in the media planning process

51 52

Koschnik, 1995, p. 1236 (Translation by the author) Koschnik, 1995, p. 1236

15

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

52

Important decisions to be made in the media strategy are the determination of which media are most suitable in order to convey the advertising message most efficiently to the target group, and furthermore the balance of media meaning the determination of the media mix. As described in Figure 2 four criteria are critical in order to select the media types in the media mix: 1. Technical criteria determine the feasibility of a campaign from a technical point of view, which is of particular importance in online-advertising. Production costs for online formats can be disproportionally high compared to the budget and are therefore another criterion. 53 2. Qualitative criteria refer to the quality of the media contact, which can influence the advertising message in a positive or negative way and is therefore the “impression of the media”. 54 3. Quantitative criteria cover the operating figures or objectives of media planning respectively, which have been described under 2.3. 4. Administrative criteria “comprise for instance the degree of flexibility and lead time of a campaign booking, the allowed design options of the advertising media regarding interaction and individualisation options as well as binding character of the placement etc.” 55 Furthermore in order to weight the media channels, five categories influence the strategy: 1. The target market Research is necessary in order to analyze the prospect consumer's buying behaviour and decision criteria, leading to s decision as to which media effort is necessary to induce the purchase. 2. The budget Using intuition and experience and analyzing the competitor's spending the budget needs

Copyright © 2009. Diplomica Verlag. All rights reserved.

to be determined and distributed across media type and geographic areas. 3. Media All communication vehicles may be used and it needs to be decided which ones will best achieve the set objectives. This is different in every situation and market but a mixedmedia approach has several advantages: 53

Cp. Gärtner, Sandra: Die Währung für Online-Werbung, Köln, 2007, EUL Verlag, p. 16 Cp. Gärtner, Sandra: Die Währung für Online-Werbung, Köln, 2007, EUL Verlag, p. 17 quoting Abel, Bodo: Medienmarketing im Werbemarkt (powerpoint presentation), 2005, p. 45 55 Cp. Gärtner, p. 17 quoting Abel, Bodo: Medienmarketing im Werbemarkt (powerpoint presentation), 2005, p. 48-53 (Translation by the author) 54

16

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

a) Target persons, who cannot be reached with only one medium, can be reached with a media-mix. b) Saturation and wear-out effects through a single-media strategy can be avoided c) The communication impact can be strengthened via the usage of interaction and transfer effects and enable synergy effects. d) The possibility of both addressing persons with highly emotional involvement as well as highly cognitive involvement. e) Repetition of exposure or frequency can be obtained in a less expensive secondary medium after an optimum reach with a first medium has been achieved. f)

The creative effectiveness of a campaign can be increased by using a second medium 56

57

4. Mechanics A central element of the media mix or the strategy respectively is the timing of the campaign, meaning whether the campaign should be placed continuously, pulsing or flighting. Furthermore decisions on spot length and sizes of print advertisements need to be made. 5. Methodology The methodology represents the overall strategy of selecting and scheduling media in order to achieve the target message weight, reach, frequency and continuity. In order to determine the optimal media mix media selection programmes are applied, which will be explained in section 2.6. 58

59 60

2.6. Media selection programmes Media selection has the intention of distributing the given advertising budget with the objective of reaching maximum advertising impact. Media selection can be divided into Copyright © 2009. Diplomica Verlag. All rights reserved.

intra-media selection and inter-media selection. 61 The inter-media selection describes "the budget allocation to the respective forms of appearance of one instrument" 62 (compare to Figure 1: communication mix). In order to compare the different forms, heuristic methods

56

Cp. Koschnik, 1995, p. 1217-1221 Cp. Arens/Weigold/Arens, 2008, p. 297-298 58 Cp. Barban/Cristol/Kopec, 1987,p. 71-76 59 Cp. Koschnik, 1995, p. 1216-1225 60 Cp. Arens/Weigold/Arens, 2008, p. 286-287 61 Cp. Koschnik, 1995, p.855+1236 62 Bruhn, 2007, p. 298 57

17

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

such as scoring-models, equation/checklists and portfolio-analysis are used. The intramedia selection means "the division of the budget of the respective forms to the several media" 63 in order to reach the objectives discussed under section 2.3. The overall objective is creating a media-mix with a division of the advertising budget to several media in such a way that with no other allocation a greater advertising impact could have been realized. There are three model-supported methods to reach this objective: 1. Ranking methods 2. Evaluation methods 3. Optimization methods 64

1. Ranking methods Ranking methods are computer-based methods, where the potential media types are listed by the media planner, after which a ranking is calculated by the computer. This ranking depends on performance criteria such as reach, GRP or cost-per-thousand ad impressions (CPM), which are related to the resulting costs. The product is a list of media, which are ranked by their return-on-investment (RoI). CPM is the most common key performance indicator (KPI); it indicates the costs incurred to reach 1000 contacts in the target group. Through value-based assesment of the target group in relation to the used media, the Return-on-investment is in the focus of this method. 65

66 67

Criticisms of ranking methods point out the isolated approach, the disregard of qualitative differences and non-distinction of primary and repeated contacts. Therefore ranking methods are mostly used for a pre-selection of the eligible media types. 68

69

2. Evaluation methods After a pre-selection of different media types a range of media-mix plan alternatives are created by the media planner. Within those plans the different media types and placement intensity vary and are then evaluated by a computer program on the basis of different Copyright © 2009. Diplomica Verlag. All rights reserved.

criteria, e.g. reach, CPM and frequency. The program therefore locates the best solution out of the media plans created by the media planner. A great disadvantage is that 63

Bruhn, 2007, p. 301 (Translation by the author) Cp. Bruhn, 2007, p. 323-324 65 Cp. Schweiger, G/Schrattenecker, G.: Werbung, 6. Edition, Stuttgart 2005, Lucius & Lucius Verlag, p. 303 66 Cp. Bruhn, 2007, p. 324 67 Cp. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=4682/ 18.7.2008 68 Cp. Bruhn, 2007, p. 325 69 Cp. Koschnik, Wolfgang J: FOCUS-Lexikon, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=4682/ 18.7.2008 64

18

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

selection only takes place within a narrow set of alternatives, and the best solution might lie outside of this set owing to the limited knowledge and experience of the media planner. 70

71 72

3. Optimization method Optimization methods are also computer-based programs, which analyze the optimal media plan, relying on the defined target such as maximized target group reach, efficient cost-performance-relationship and so forth, as well as constraints such as budget, obligatory media etc. Optimization methods are based on mathematical methods such as linear programming and do not consider contact-overlapping, -allocation or frequency and therefore do not live up to the expectations of media planning and are for that reason hardly used anymore in practice. The most common method despite its disadvantages is the evaluation method, because it involves a multitude of criteria and is therefore best suited to find a solution for a successful media plan. 73

74

The situation in online media planning differs from the initial situation of traditional media since in contrast to offline media, where the campaign can be optimized on the basis of several possibilities before it is placed and then evaluated after it has been run, an online campaign can be optimized in the course of its placement based on the permanent measurement of media performance. Starting with a ranking method to pre-select different options is very common in online media planning; reach and frequency are as important index numbers as in traditional media planning. A big difference here is the ability – via cookies – to completely control the actual frequency. Next to those numbers the quality of the contacts is of high interest, meaning whether the target group has been reached with the advertisement Here the cookie-based approach is an advantage, as you can objectively demonstrate exactly who got the ad how many times. Starting with the ranking method tends to result in coarse media planning in the first few weeks, which means a waste of time and money and

Copyright © 2009. Diplomica Verlag. All rights reserved.

therefore less efficiency before the target group online is actually reached. 75

70

Cp. Schweiger/Scharttenecker, 2005. p. 304 Cp. Bruhn, 2007, p.325-326 72 Cp. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=1704/ 18.7.2008 73 Cp. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?stichwort=Optimierung&x=0&y=0/ 18.7.2008 74 Cp. Bruhn, 2007, p. 326-328 75 Cp. Johannsen, Nicolai: Banner-@advertising – Klickraten als Möglichkeit einer Mediaselektion im Internet?, Wiesbaden 2002, Deutscher Universitätsverlag, p. 12-20 71

19

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

In this paper it will discussed under section 5 if behavioural targeting can deliver a solution to that drawback and enables media planners to set up an efficient media plan before the campaign, and continuously improving it while minimizing waste.

2.7. Requirements for optimal media planning: Elements and process This chapter has examined the requirements for optimal media planning and can be summarized in the following elements: 1. The foundation are clearly defined and set marketing objectives such as for example market share and sales increase as well as communication objectives which intend to influence the attitudes and motivations of consumer provide the foundation for meaningful media objectives. 2. As a consequence of the marketing and advertising objectives the media objectives need to be equally well defined and include the reach, the frequency, the message weight and the continuity as well as the definition of the target group. 3. Optimal media planning means reaching the right customers, therefore the definition of the target group is at the centre of attention.. 4. In order to be able to define the target group and the market segment a solid information basis is necessary. Market research therefore provides the foundation of media planning and represents next to the setting of objectives the beginning of the process. It comprises a market analysis in order to get an overview of market forces and their various dynamics, as well as consumer research and a competitive analysis. Market analysis is done by a variety of market research firms such as AC Nielsen, publishing houses such as Gruner+Jahr as well as small specialized agencies, which use a number of analysis methods.

76

5. After the media objectives are set, the market research is done and the target group is defined the budget has to be determined on that basis. In practice this is Copyright © 2009. Diplomica Verlag. All rights reserved.

the most difficult part, since there is a deep disagreement in the industry over which objective is most important, and therefore a variety of budgeting methods exist. 77

76

Cp. Unger, 2007, p. 43 Cp. Unger, Fritz: Mediaplanung – Vorraussetzungen, Auswahlkriterien und Entscheidungslogik, in: Scholz, Christian(ed.): Handbuch Medienmanagement, Berlin Heidelberg 2006, Springer Verlag, p. 739-745 77

20

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

6. Based on the media objectives and the budget, media channels need to be identified and selected. It includes the determination of the media scheduling (booking, timing, placement, and dimensions of an advertisement). This step represents the media strategy in order to meet the target group and the objectives which have been set. 7. Since the budget determines the media strategy, it also determines the media plan, meaning the choice of channels and concrete media in each channel. So once the media strategy has been established, several alternative media plans are created, e.g. via the media selection programmes as explained in section 2.7., in order to find the best one. Once a final media plan exists a briefing of the media and creative agencies can be carried out. 8. The final the step is the supervision and control of the media plan execution, which includes an evaluation of the campaign effectiveness after it has run and an

Copyright © 2009. Diplomica Verlag. All rights reserved.

assessment of the extent to which the objectives have been reached.

78

Cp. Unger, 2007, p. 26-33 Cp. Koschnik, 1995, p. 1232-1233 80 Cp. Tellis/Ambler, 2007, p. 299-300 81 Cp. Barban/Cristol/Kopec,,1987, p. 4-6 79

21

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

78 79 80 81

3. Online as advertising medium There is a wide range of media types such as TV, magazines, news papers, outdoor advertising, radio, cinema and the internet. 82 For the purpose of this study a short comparison of offline and online media is appropriate. In order to define online advertising, the term ‘advertising’ is defined first: “Advertising is a paid, mediated form of communication from an identifiable source, designed to persuade the receiver to take some action now or in the future.” 83 Therefore “all paid space on a Web site or in an email is considered advertising. Internet advertising parallels traditional media advertising, in which companies create content and then sell space to outside advertisers.” 84 “Under Online advertising in its original sense the placement of advertising material, primarily banner, on web pages in order to support marketing and communication objectives is understood”. 85 Under traditional media all types which are mentioned above besides the internet are understood.

3.1. The importance of online advertising The amount of media effort has drastically increased in the last years within all types of media. This has resulted in a situation where the audience is not just limited in the capacity to capture advertising messages, it even refuses to do so on a massive scale. Therefore the new keywords are individualisation and personalisation. 86 While traditional offline media work through the ‘push-principle’, meaning that a mass communication and marketing is only directed at the prospect consumer through broadcasting with little or no interaction, the new media internet works with the ‘pull-principle’, meaning the individual user pulls the content and related advertising into attention through Copyright © 2009. Diplomica Verlag. All rights reserved.

his own selection. On the Internet the consumer has the possibility to interact and influence actively, opening the possibility of one-to-one communication, but he can also

82

Cp. Unger, 2007, p. 185-364 Richard, J.I./Curran, C.M.: Oracles of „advertising“: Searching for a definition, Journal of Advertising, 31(2), 2002, p. 74 84 Strauss, Judy/El-Ansary, Adel/Frost, Raymond: E-Marketing, 4. Edition, 2006 USA, Pearson Prentice Hall, p.322 85 Lammenett, Erwin: Praxiswissen Online-Marketing, Wiesbaden 2006, Gabler Verlag, p. 122 (Translation by the Author) 86 Cp. Schmahl, Diana: Moderne Online-Marketing-Methoden, Saarbrücken 2007, VDM Verlag Dr. Müller, p. 3-4 83

22

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

be targeted with mass communication, the one-to-many communication. The greatest improvement gained by online advertising is therefore the possibility to create a personal and individual communication between consumer and advertiser, which functions in both directions.

87 88

The ‘Digital Influence Index Study’ 89 from June 2008 which has addressed the issue of understanding the role of the internet in the lives of consumers in the UK, Germany and France is based on an online survey among a representative sample of nearly 5000 internet users across the mentioned countries between December 2007 and January 2008 and shows very well the changed picture within the media landscape. The most interesting finding is: “The internet is by far the most important medium in the lives of European consumers – but companies are underinvesting in that influence.” 90 The Internet is twice as influential as TV and has ten times the influence of print media, however the global online advertising spending accounts only for 7-8% of the total advertising market. Reich (2006) arrives at the same conclusion stating that the advertising impact is higher within online advertising than traditional media. 91 The worldwide online advertising spending added up to 32.867.900 in 2007, compared to 9.451.300 in 2002 – an increase of 350% in five years. 92 So online advertising is experiencing strong growth but as mentioned a big discrepancy still exists between online ad spend and online media consumption and advertising effectiveness. The online advertising market can sensibly be divided into the U.S. and Europe, since the rest of the world is currently not nearly as advanced as these markets and can therefore not provide a foundation for the examination of the actual state of affairs. Europe in itself is also very diverse in its development. Therefore the U.S. market has grown much faster than the European one with total internet advertising revenues of $12.5 billion in 2005 forecast to grow by nearly 500% to $61.300 billion in 2012. 93 In contrast Europe can be seen as a patchwork with its countries at different stages of development and divided by language barriers. The ‘digital giants’ 94

Copyright © 2009. Diplomica Verlag. All rights reserved.

in Europe are the UK, France and Germany, which account for more than 85% of online

87

Cp. Schumann/Thorson, 2007, p. 17-18 Cp. Schmahl, 2007, p. 4-5 89 McRoberts, Brian/Terhanian, George H., Fleishman Hillard/Harris Interactive: Digital Influence Index Study, June 2008 90 McRoberts, Brian/Terhanian, George H., Fleishman Hillard/Harris Interactive: Digital Influence Index Study, June 2008 91 Cp. Reich, Sandra: Die persuasive Werbewirkung von Online-Medien – Eine empirische Studie zu Wirkung von Werbung, Aachen 2006, Shaker Verlag, p. 190-194 92 Euromonitor International, data exported 09/04/2008 93 VanBoskirk, Shar/Forrester: US Interactive Marketing Forecast 2007 to 2012, Cambridge USA 2007, Forrester Research Incorporation, p.9 94 Interactive Advertising Bureau Europe: Digital Europe – Tracking the growth of online marketing spend, Digital Insight Report, 2006, p. 4 88

23

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

advertising spending, furthermore the Scandinavian countries grow rapidly, while other countries still have to cross the tipping point in order to realize a substantial growth within their digital markets. This circumstance causes some heterogeneity in the European market but is forecasted to balance out within the next few years.

95

So although online advertising spending is rapidly growing, from the researchers point of view advertisers are still ‘underspending’ – but this is very likely to change: “Advertising spend eventually synchronize with shifts in consumer attention from television to other media formats.” 96

% Online Ad Spend of Total Ad Spend

40%

% Time Spent Online of Total Time Spent by Media 31%

30%

29% 27% 25% 23% 21%

20%

19% 17%

10%

2.8%

Figure 3: % Time Spent Online vs. % Share of Online Adspend across Europe

20

10

09 20

08 20

07 20

06 20

05

0%

20

04 20

20

03

1.8%

9.4%

8.3%

7.1%

5.9%

5.1%

3.7%

97

The future development of online advertising will be one issue examined in section 6.4., since it is of great importance for the growth of behavioural targeting and the implications for media planner in a changing media environment.

Copyright © 2009. Diplomica Verlag. All rights reserved.

3.2. Brief overview of the history of online advertising The dawn of online advertising is dated to the 24th of October 1994 when the US-based online magazine hot-wired.com published the first advertising-financed issue. The

95

Cp. Interactive Advertising Bureau Europe: Digital Europe – Tracking the growth of online marketing spend, Digital Insight Report, 2006, p. 4-21 96 Berman, S.J./Battino, B./Shipnuck, L./Neus, A.: IBM Global Business Service – The end of advertising as we know it, 2007, IBM Global Business Services, p. 7 97 EIAA Media Scope Europe: Media Consumption Study 2006 (Hours – 2003, 2004, 2005 actual, 2006-2010 prediction, Adspend – 2003, 2004, 2005 actual and 2006-2010 prediction)

24

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

company AT&T placed the first banner in a specific format, which still exists today and will be explained under section 3.4. 98

99

100

While traditional media mostly target a broad audience the dispersion losses could be reduced online. Since 1994 different banner formats have been introduced and gradually changed from static images to dynamic ‘mini-movies’, often with interactive elements. The online advertising, while still quite immature, is very innovative and under constant development.

3.3. The current state of the industry The value chain in the advertising industry does not deviate much from the traditional advertising pattern of the advertiser-creative agency-media agency constellation on the demand side. However, in online advertising the supply-side is much more heterogeneous creating a complete new structure or rather non-structure, “the ecosystem is in chaos” 101 and “the industry is extremely immature” 102 . Ownership consolidating

Advertiser

Media Agency Ad Agency

Sales House/Portals Network

Large Publisher

User

Small Publisher

Figure 4: Value-chain in online advertising (Based on Skou) 103

Sales houses (sales companies owning or owned by significant parts the sold media),

Copyright © 2009. Diplomica Verlag. All rights reserved.

Portals and ad networks (sales companies with no ownership of the sole media) consolidate several online properties into single points of sale, thereby facilitating planning, booking, invoicing and reporting to some extent. However, the business 98

Cp. Wolf, Volkhard: E-Marketing, 2007 München, Oldnebourg Wissenschaftsverlag, p. 244 Cp. Sander, Carsten/Schott, Alexander: Bannerwerbung – Formate, Erfahrungen, Trends, in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketingBörse GmbH, p. 277 100 http://www.mediensprache.net/de/werbesprache/internet/formen/banner.asp/ 30.7.2008 101 Rasmussen, Andrea/Ude, Carolyn/ Landry, Edward: Interactive Advertising Bureau: HD marketing 2010 – Sharpening the conversation, New York 2007, p. 10 102 Skou, Kasper, Senior Vice President Product and Business Development, wunderloop, Interview 28.7.2008 103 Skou, Kasper, Senior Vice President Product and Business Development, wunderloop, Interview 28.7.2008 99

25

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

processes of the industry are still very manual, relying on phone calls, faxes and manual distribution and inspection of ad material distribution. This creates high process costs for media agencies in the handling media – despite the aggregation function performed by portals, sales houses and ad networks – and results in a situation where online media planners can only deal with 10-20 different sales points. Needless to say this is not optimal for the advertiser, who is interested in getting the best RoI on ad spend independently of the particular relationships an online planner is able to cultivate with the sales side. Looking at the power distribution across the value chain the geographical markets are quite different. The German and Scandinavia markets are distinctive from the U.S., UK and the French markets, since sales houses and portals are of the highest importance in order to sell and place advertisements. Sales houses in Germany (Vermarktungsgemeinschaften) are for instance GWP online marketing and G+J Electronic Media Sales GmbH, both owned by large traditional publishing houses such as Gruner + Jahr. Portals include Yahoo!, AOL and T-Online (represented by Interactive Media, a wholly-owned subsidiary). The biggest sales houses and some portals are organized in the so-called ‘Online-Vermarkterkreis (OVK)’, and the power of this institution can be estimated by looking at the extreme concentration in the German market: The 23 top players account for over 80% of the monthly page impressions measured by the IVW. 104 The less powerful players in Germany are the sales networks, who have less control over the sold inventory as there is no cross-holding, and who control a much smaller percentage of the monthly available page impressions. Scandinavia shows a similar pattern of traditional media houses dominating and the ad networks being in a much weaker position. This is different in the U.S., UK and France where the networks have the greatest power. In these markets networks and portals have been able to aggregate substantially more reach than any traditional media houses, making them much more interesting to media buyers. 105

Copyright © 2009. Diplomica Verlag. All rights reserved.

Over the last 2 years an interesting new position has been carved out in the industry value chain by the so-called ad exchanges. The raison-d’être of ad exchange is the aggregation of inventory into one point of sales on the basis of technology – not expensive human resources. Large publishers put their remnant 106 inventory into the exchange and smaller sites not represented by ad networks put their entire inventory into the exchange. The 104

Cp. IVW, http://www.ivwonline.de/ausweisung2/search/ausweisung.php / 08.09.2008 Cp. Skou, Kasper, Interview 28.07.2008 106 The inventory not sold by the publisher or its sales representative and therefore either monetized through 3rd parties such as advertising.com, Google AdSense, affiliate networks or ad exchanges, or simply not monetized. 105

26

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

media planner can log into the exchange and find the desired inventory through a graphical user interface, replacing the expensive and manual process of involving the media sales person. Many ad exchanges attempt to add different types of targeting to the offering in an effort to raise the value of the inventory, e.g.: •

Adify enables contextual targeting by forcing the publisher to categorize his own content (making the quality of this targeting dubious)

•

wunderloop’s exchange Connect has full-blown behavioural and sociodemographic targeting

•

Yahoo!s Right Media Exchange allows 3rd-party targeting via cookies

Ad Exchanges have not achieved great traction in the market place yet, but they manifest a trend in the industry of which Google is also an example: The position of the media agencies as aggregators of buying power leading to volume discounts is coming under pressure from tools, which threaten to automate the function of the media agencies with technology thus eliminating their raison-d’être. Likewise, the aggregation function on the supply-side is under similar pressure by the same technologies. The benefit for the advertisers and publishers is a vastly more streamlined value chain with attendant savings through disintermediation. 107

108

3.4. Online advertising formats and performance measurement The main online advertising format is the classic banner advertisement; an example of the

Copyright © 2009. Diplomica Verlag. All rights reserved.

different types is shown below.

107

Cp. Skou, Kasper, Interview 28.7.2008 Cp. Dannenberg, M./Wildschütz, F.H.: Erfolgreiche Online-Werbung, 2. edition, 2006 Göttingen, Business Village Verlag, p. 70-73 108

27

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Figure 5: Classification of banner Dannenberg/Wildschütz (2006)

The formats as shown above are described by their dimensions in pixel. The Internet Advertising Bureau constantly checks banner formats concerning their efficiency and sets non-binding standards, which are accepted by the industry. Through those standards banners can be placed in standardized ‘containers’ on web pages, which reduces costs. 109

110

Within the functionality three types of banner can be identified: 1. Static banner: They can be compared to a classic advertisement and refer to another web page through a hyperlink, which is activated if the banner is clicked. These banners can only attract attention through a static picture so to convey an advertising message is rather difficult and this is the reason why those banners are rapidly going out of fashion. 2. Animated banner: These banners represent the second stage of development and are the enhancements of static banners. Through animations more attention should be attracted from the users and the design surface can be multiplied. These banners

Copyright © 2009. Diplomica Verlag. All rights reserved.

usually rely on a click-link as well. 3. Interactive banner: Whereas the only interaction possible with static and animated banners is the click, interactive banner represent the third and highest stage in development of online advertising formats. These banners allow the user to interact with a banner, meaning

109 110

Cp. Kollmann, Tobias: Online-Marketing, 2. Edition, 2006 Göttingen, Business Village Verlag, p. 179-180 Cp. Dannenberg/Wildschütz, 2006, p. 24-25

28

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

the banner might expand to reveal a micro-site or movie player integrating control buttons, hyperlinks, mouse-over-functions and pull-down-menus. Other forms of online display advertising include: 1. Pop-ups These can also be described as banner but they are opened in a separate browser window and allow better attention attraction and greater design surface. However, this form of advertising usually creates greater irritation with the user and is loosing importance because of pop-up blockers. 2. Interstitials: These usually take the form of a layer ‘hovering’ on top of the page the user wishes to access and prevents access to this page until the user either closes the layer or clicks the ad. Interstitials are very eye-catching, but also tend to cause major irritation with the user. 3. Streaming ads: These are interactive commercials, where the content is streamed from a 3rd-party server. This makes it possible to integrate full movie trailers without slowing down

Copyright © 2009. Diplomica Verlag. All rights reserved.

page load times (will often be in the form of an interactive banner).

111 112

111 112

Cp. Dannenberg/Wildschütz, 2006, p. 24-37 Cp. Kollmann, 2006, p. 180-183

29

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

4. Targeting Online Targeting methods, as they are described in the following, only refer to online advertising practices. The term ‘targeting’ is derived from ‘target’ and therefore “Targeting is the process through which marketing communicators deliver messages more accurately and prevent wasted coverage to prospects not included in the intended audience.” 113 Targeting referring to online advertising derives from banner advertising 114 as it has been described in section 3.4. “and characterizes targeted advertising effort on the Internet.” 115 The different forms of targeting, which will be described below, manage the distribution of online advertising by different criteria and “enable advertisers and media planners to identify and target customers in that way that waste coverage in online advertising can be reduced significantly.” 116 The central goal of targeting is therefore the decrease of dispersion wastage thus increasing efficiency. The main advantage of online-targeting compared to offline-targeting is the utilization of the real-time capability of the internet, which allows a dynamic reaction to how a target group is reacting, the immediate implementation of data which enables a prompt optimization of placements and the delivery of a substantial amount of high quality data which would not be available or only with large costs in offline media. 117 So compared to offline media, the ability in online advertising to target single individuals is better. In general online display advertising relies on distribution via a dedicated system called an ad server. The web page on which display advertising is to be shown one or more containers for advertising, called placements, are included in the HTML-code of the page. Once an internet user opens the web page, the ad server is notified and sends an ad to the placement(s) in real-time, creating an ad impression per placement on the page. As this is done in real-time, a decision can be made – at the exact time when the user consumes the page – which advertisement will be shown. If the user clicks on the banner,

Copyright © 2009. Diplomica Verlag. All rights reserved.

the ad server is notified of this and the action is recorded as an ad click. At the same time

113

Micu, Anca C.: How did this ad get in my browser? A theoretical examination of online targeting and segmentation practices as theyrelate to media planning on the internet, American Academy of Advertising Conference Proceedings, 2005, p. 208 114 Cp. Mühling, Johannes: Targeting – Zielgruppen exakt online erreichen, München 2007, Verlag reinhard Fischer, p. 64 115 Bürlimann, Martin (2004): Web Promotion – Professionelle Werbung im Internet, 3rd Edition, Zürich 2004, Midas Management Verlag, p. 109 (Translated by the author) 116 http://www.bvdw.org/presse/news.html?tx_ttnews[tt_news]=1367&tx_ttnews[backPid]= 4&cHash=55c5427390/ 07.08.2008 (Translation by the author) 117 Cp. Hegge, Ulrich: Targeted Advertising, in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketing-Börse GmbH, p. 287

30

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

the user is taken to the landing page specified in the click-URL embedded in the banner advertisement. The foundation of targeting online is user data, which is connected to the user via different methods, e.g. the IP-address of the user or a cookie in the user’s browser. 118 The first attempts to introduce targeting was by the US-based company Engage technologies in the mid-nineties, but failed because it was too complicated, too costly and too inflexible for advertisers. In 2000 and 2001 two of the major targeting companies Revenue Science and Tacoda implemented their technology which can be seen as the start of the targeting business. 119

120

Similar to the target group segmentation with its varying criteria, targeting can be performed in different forms and based on different criteria.

4.1. Geo Targeting “Geo targeting can be described as the classic targeting.” 121 It means the optimization of the placement depending on the geographic location of the user consuming the page. Consumer can be targeted in a specific area via their IP-address with online advertisement for instance referring to local products or services. 122

123 124

A potent

application of this technology is e.g. the exclusion of South American users on Spanish sites, and foreign users on UK sites.

4.2. Prime-Time Targeting Prime-Time targeting is time-based meaning that on a fixed location on a web page varying advertisement can be shown at different times of the day or week. Based on the assumption that different target groups are using the web at different times and detailed ad server reporting proving which advertisement performs best at what time the targeting

Copyright © 2009. Diplomica Verlag. All rights reserved.

can be more effective. 125

118

Cp. Wolf, 2007, p. 251-252 Cp. Wood, Jeff: Behavioral Targeting: 2004 in Review, in: http://www.imediaconnection.com/content/4858.asp / 07.08.2008 120 Cp. Mühling, 2007, p. 65 121 Denker, Helge: Targeting – Sein verhalten macht ihn zum Ziel, March 2007, in: http://www.adzine.de/de/site/contentfcmsv445f073499f7dc/artikel.xml /04.06.2008 122 Cp. Lammenett, 2006, p. 119 123 Cp. Dannenber/Wildschütz, 2006, p. 87 124 Cp. Lemonnier, Jonathan: Contextual targeting boosts loyal following, Advertising Age, 4/14/2008 ad Network & Exchange Gui, Vol. 79, p.7 125 Cp. Dannenber/Wildschütz, 2006, p. 86-87 119

31

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

4.3. Technical Targeting The ad server is able to identify the type of browser, operating system, language and bandwidth of the individual user in order to determine which advertisement or indeed version of the same advertisement can be displayed. This may also be used to limit distribution, as the operating system can give a hint who is using the computer; a Macintosh user is quite often a freelancer and/or working in advertising or the creative industry while a Linux user is IT-affine, for example. 126

127 128

4.4. Profile Targeting Profile targeting is based on behavioural and socio-demographic criteria as explained in 2.4.1. If privately identifiable information is used in the profile, permission of the user is needed (opt-in). Normally profiles exist only for a small percentage of the target group and it can be costly to maintain the actuality of the data. However, those profiles can be used as a foundation for media planning for standard target groups. 129

130

4.5. Contextual Targeting Contextual targeting refers to the placement of advertisements, which are contextually relevant to the content of the web page. The aim is to use the interest of the user in a specific topic. Google AdSense is the most well-known provider for contextual targeting offering a technology, which “understands” the web page through technology-based, automatic categorization and then placing advertisements, which fit this category. The system is not immune to failure, since not all content provides a basis for planning; both long-term interests and socio-demographic characteristics are not taken into account and

Copyright © 2009. Diplomica Verlag. All rights reserved.

there is no guarantee for a certain booking period or number of contacts. 131

132 133

126

Cp. Hegge, 2007, p. 288 Cp. Dannenber/Wildschütz, 2006, p. 87 128 Cp. Mühling, 2007, p. 68-69 129 Cp. Mühling, 2007, p. 72 130 Cp. Hegge, 2007, p. 288 131 Cp. Dannenber/Wildschütz, 2006, p. 72-73 132 Cp. Hegge, 2007, p. 289 133 Cp. Gay, Richard/Charlesworth, Alan/Esen, Rita: Online-Marketing – a customer-led approach, New York 2007, Oxford University Press Inc., p. 191 127

32

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

4.6. Keyword Targeting Keyword targeting is based on the search request a user types into a search engine. Google provides the most famous tool to target user by their search words: Google AdWords. Next to the search result, a list of textual advertisements with links is displayed, which is evoked by the search word and ranked according to an auction mechanism: The more you pay for the search term, the higher you will be ranked in the list of ads. Since the entry of search words indicates behaviour of the user, keyword targeting can be seen as a form of behavioural targeting, which will be explained and discussed in detail in the

Copyright © 2009. Diplomica Verlag. All rights reserved.

next chapter. 134

134 135

135

Cp. Dannenber/Wildschütz, 2006, p. 69-71 Cp. Hegge, 2007, p. 289

33

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

5. Behavioural Targeting .

Currently behavioural targeting (BT) is neither scientifically well-defined nor standardized, though there are several approaches defining and describing BT. The following statements should provide the background for the discussion on BT’s importance for media planning: “Behavioural Targeting (BT) is a method of advertising seeking out consumers based on where they go and what they do as indicators of what their interests are most likely to be. BT is mostly being applied in online ad targeting, where consumers’ behaviour on a relevant website is tracked to determine their interests, then ads are served to that person relevant to their interests.” 136 Forrester, a US-based technology and market research company, defines BT as follows: “Aligning ads with users whose online behaviour implies interest in a particular product or service” 137 The concept of BT is not the latest trend; it is rather a victim of the “dot-com bomb”

138

and

therefore disappeared after the year 2000. It is unclear who invented BT, the US-based company ValueClick holds two patents on the general principles of behavioural targeting, listed as inventor is Thomas Gerace who filed the patent in 1990. Currently ValueClick sues Tacoda over behavioural targeting patents; Tacoda was founded in 2001 and is regarded as the pioneer in the industry. 139 BT is experiencing a comeback, which was manifested in 2007, where multi-billion dollar acquisitions took place. Google bought DoubleClick for $3.1 billion, Yahoo! bought Bluelithium for $300 million and AOL bought Tacoda for approximately the same price. 140 “These moves and others have led observers to suggest that the next golden age of behavioural targeting is around the corner.” 141 Forrester forecast already in 2006 that BT “will be the most important way of doing interactive marketing in 2008.” 142 This has turned out not to be the case, however the industry is changing and still under development. (see section 6: The current state of affairs). The massive investments by the leading industry Copyright © 2009. Diplomica Verlag. All rights reserved.

players do manifest a very real interest and belief in BT, begging a central question: What

136

The 2007 Entertainment, Media & Advertising Market Research Handbook, USA 2007, Richard K. Miller & Associates, p. 117 137 VanBoskirk, Shar/Forrester: The Reality of Behavioral Ad Targeting, Cambridge USA 2006, Forrester Research Inc., p. 2 138 Tsai, Jessica: Oh Behave!, CRM Magazine, Jan 2008, Vol. 12, Issue 1, p. 24 139 Cp. Davis, Wendy: ValueClick Sues Tacoda Over Behavioral Targeting Patents, August 2008, in: http://www.mediapost.com/publications/?fa=Articles.san&s=87909&Nid=45708&p=944842 / 21.08.2008 140 Cp. Gengler, Barbara: Marktforscher erwarten starkes Wachstum für verhaltensbasierte Werbeformen im Internet, Computer Zeitung, Heft 33, 2007 141 Tsai, Jessica, 2008, p. 26 142 VanBoskirk, The Reality, 2006, p. 7

34

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

can BT accomplish, where are its advantages and disadvantages? These questions should be analyzed in the following with regards to the importance concerning media planning.

5.1. Three different types of Behavioural Targeting There are several approaches and technologies within the BT industry. A set nomenclature does not exist yet, and every company with a different approach tries to differentiate itself by coming up with a new label. The typology below should therefore be seen as a rough categorization only.

5.1.1. Affinity-based Behavioural Targeting Affinity-based or traditional BT is based on tracking the information consumption of the user. Through some form of categorization this content is assigned to clusters with marketing relevance. Consistent consumption of content in a given category is taken to mean an above-average interest in this category, which is assumed to lead to better advertising response. In this way advertising can be based on the unique profile of the user irrespective of the context in which the user currently is. 143 As an example, if a user shows interest in golf by typing in specific words into a search engine – keyword targeting is also being considered as BT since it allows a conclusion to the user’s interest 144 - or clicks on golf-related web pages and then moves on to a newspaper’s web page, he can still be targeted with golf-specific advertisements, although he is no longer in a golf-related environment. The user therefore does not have to be in a specific subject area online in order to be hit by the right advertisement; instead the right advertisement will find the right target person online. Furthermore affinity-based BT can be distinguished by situational and long-term BT. In the case of situational BT, according to the AIDA-formula 145

146

, the user is at a progressed

point in his decision process, the ‘Desire’, meaning that he has to be served with Copyright © 2009. Diplomica Verlag. All rights reserved.

advertisements which lead him directly to ‘Action’, the completion of a purchase. This could be the case in the situation of a user searching for information about cars and is willing to buy one in the near future.

143

Cp. Interactive Media/ENIGMA/GfK: Market Report 2006: Erfolgsfaktor Targeting – Funktionsweise, Wirkung und Marktbedeutung, p. 18 144 Cp. Bryant, Melchior D.: Neue Trends im Behavioural Targeting, in: Absatzwirtschaft, Issue 4, April 2008, p.42 145 Cp. http://www.werbe-lexikon.info/fachbegriffeaz/aidaformel.html / 22.08.2008 146 Cp. Bryant, 2008, p.42

35

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

In contrast to this long-term BT focuses on the circumstance where a user has a permanent interest in a subject, which can be classified as a basic interest, for instance golf, which according to the AIDA-formula describes the stage of ‘Attention’. Attention ⇒ Interest ⇒ Desire ⇒ Action 147

148

Therefore it can be said that it is important to catch a user by situational BT just at the right moment before he executes the purchase of for instance a new car and provide the right advertisement for him to gather information, because that is the topic he is currently the most interested in and therefore considers relevant advertisement. Long-term BT is as important but for other reasons. A long-term interest such as golf needs to be used by advertisement because it is connected with several purchases over time. There could be the need for new equipment, a week-end trip to a golf resort or the subscription to a subject-related magazine, all of which needs to be offered to the prospect customer repeatedly although he might not have visited golf related web pages for a while.

5.1.2. Re-targeting Re-targeting is simpler than affinity-based BT. It refers to the situation when a user has already visited a web page and was at the stage of making a desired action, such as a purchase, but then left the web page before the desired action was completed. The user is tagged – probably via a cookie – and can now be targeted on other topics for retention marketing. If the desired action was indeed completed, the user can be re-targeted with add-on or cross-selling propositions. For instance, if the user has already typed in all data for a flight connection to London including the date and the price range he is willing to pay (first class, business, economy etc.) but then did not book the flight and instead moves on to another web page, this user, in the period of ‘Desire’ can be targeted with a specific advertisement from the travel web page he has been visiting stating exactly what he was looking for can trigger him to execute the purchase. 149

150

Copyright © 2009. Diplomica Verlag. All rights reserved.

A tailor-made advertisement to the user’s most recent interest is possible via re-targeting, which is therefore described as the “hidden gem of behavioural targeting” 151 because it offers a high probability of turning a desire into a purchase.

147

Cp. http://www.werbe-lexikon.info/fachbegriffeaz/aidaformel.html / 22.08.2008 Cp. Bryant, 2008, p.42 149 Cp. Mühling, 2007, p. 82-83 150 Cp. Croft, Martin: Behavioral Targeting – it could be the holy grail of online advertisement, in: Marketing Week: Interactive, 29.03.2007, p. 36 151 Smith, Jack: Retargeting will feature in your future, January 2006, in: http://www.imediaconnection.com/content/7907.asp / 22.08.2008 148

36

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

For both affinity-based BT and Re-targeting large, heterogeneous reach is crucial, because the behaviourally targeted advertisements can only be placed in the right moment when the user is moving to web pages that are connected to the BT technology in use. Networks as large numbers of web sites hosting advertisements are suitable partners and superior to site-by-site implementations. 152

153

5.1.3. Predictive Behavioural Targeting Predictive targeting is based on different types of data, which can be collected online and offline. The first step is to gather data online, which means to analyze the behaviour via clicks and surfing the internet in real time. The user is monitored online, therefore it can be reconstructed which pages he has visited, how far he has gone through a site and where he has clicked. This gives an indication what the user is interested in and a first profile is created. The data is then completed by survey data and/or other sources of sociodemographic data, most commonly via questionnaires sent via ad placements on the page to selected users by the BT company itself or through data from market research institutions such as ACNielsen Netratings. The surveys provide information for a typical socio-demographic profile; data concerning demographics, interests and lifestyle complete the gathered date from the online behaviour or can be used to correct it.

154 155 156

The next step, after the survey data has been merged with the performance data in order to complete the profile, is the use of predictive technology. This allows the extrapolation of socio-demographic data from a sample of users to the greater population via similarities in the exhibited behavioural pattern of sample users and other users in the population. 157

158

159

The specific disadvantages of this approach are a lower precision since it is relying on predictions gathered from various data as well as slow reaction times to changes in behaviour. In order to collect that data a rather high effort has to be made, therefore the costs in relation to the efficiency need to be calculated. 160 Also, changes in behaviour will not be reflected very quickly, because the system relies on Copyright © 2009. Diplomica Verlag. All rights reserved.

answers to questionnaires and extrapolations from this data. For instance, if a user has been showing a great deal of interest in remortgaging products for months, this can be a 152

Cp. Tsai, Jessica, 2008, p. 28 Cp. Mühling, 2007, p. 82-83 154 Cp. Hegge, 2007, p. 291 155 Cp. Bryant, 2008, p. 42 156 Cp. http://www.nugg.ad/de/produkte/technik.html/ 22.08.2008 157 Cp. Irwin, Tanya: ValueClick To Launch Predictive Behavioral Targeting, July 2008, http://www.mediapost.com/publications/?fa=Articles.san&s=86905&Nid=45235&p=118697 / 22.8.2008 158 Cp. Hegge, 2007, p. 291 159 Cp. http://www.nugg.ad/de/produkte/technik.html/ 22.08.2008 160 Cp. Hegge, 2007, p. 291 153

37

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

part of the predicted profile. Suddenly this user changes behaviour and starts looking at expensive cars, arresting all interest in remortgaging. This user has remortgaged and is now looking to put the new-found wealth into an expensive car, but a predictive BT system will not be able to react quickly to this change. For that reason predictive BT is better suited to long-term interests than short-term. A fourth form constitutes the Integrated Targeting, which as the name says, integrates the different forms of targeting into one product. The idea behind this approach is that the specific needs of a campaign selects the forms of targeting relevant to the it, meaning all possibilities and combinations can be used, for instance geo-targeting joined with affinitybased BT. The goal is to find the right mix for the campaign, and there is no right or wrong. However, such a wealth of possibilities combined with a great amount of data might overburden planners and analysts, leading to information paralysis or the choice of a sub-optimal targeting strategy. 161 Referring to section 2.4.2. the combination of several segmentation criteria is very common in online advertising and according to the study by Schumann/Thorson very reasonable. The BT industry takes advantage of those findings and combines approaches with other targeting methods in order to increase efficiency.

5.2. Behavioural Targeting in practice Apart from the IP-address or cookie-based approach outlined above, the latest developments in user profiling come from Internet Service Providers (ISPs). By providing the user access to the internet, the ISP is able to follow the user wherever he goes on the internet and can therefore draw a much more complete picture of online behaviour than most BT providers, who will only see the part of the user’s behaviour that happens on sites connected to the provider’s technology. This approach is facilitated by companies such as Phorm, FrontPorch and NebuAd, and has spurred a lot of resistance from consumer rights groups and government agencies, landing these providers in hot

Copyright © 2009. Diplomica Verlag. All rights reserved.

waters. 162 In order to exemplify the operating mode of BT, wunderloop’s connect marketplace for behaviourally targeted advertising shall provide an insight via the diagrams below.

161 162

Cp. Hegge, 2007, p. 292 Cp. Hallerman, David (eMarketer): Behavioural Targeting: Marketing Trends, New York 2008, p. 11

38

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

1. Step

2. Step

Copyright © 2009. Diplomica Verlag. All rights reserved.

3. Step

39

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

4. Step

5. Step

Copyright © 2009. Diplomica Verlag. All rights reserved.

Figure 6: wunderloop’s connect marketplace 163

163

Skou, Kasper, Interview 28.7.2008

40

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

5.3. Requirements for successful Behavioural Targeting As already mentioned under section 5.1. marketers need to choose a BT provider with access to a great and heterogeneous network in order to achieve comprehensive reach and diverse profiling. 164 To get started with BT marketers should be aware of what their campaign goals really imply, meaning what BT should deliver in order to achieve those goals. For instance, what kind of behaviour are they looking for and based on that how can a message be spread in order to reach the right customer – and indeed who is the right customer? The target group needs to be defined in terms of the actual segmentation methods and attributes on offer by the BT provider in order to target these users. However, marketers should be flexible and be willing to learn from the technology. It might happen that the specified target group turns out not be the most important customer group, and certain BT systems can help marketers to discover the actual target group. This happens by profiling the users of the marketer’s goal page, e.g. the ‘Thank you’-page after signing up for a test drive of the new Golf 6. In this case BT is able to provide empirical evidence of who the actual customers of the marketer are and in that way even inform offline advertising efforts. Furthermore the combination of contextual targeting in order to increase awareness, and BT to target customers once they’ve left the content of the product or service can increase the success of BT overall. Highly important is the awareness that only a constant optimization can bring the intended results and with this goes expectation management. Marketers must be aware of the fact that BT is a long-term strategy and the technology needs to ‘be trained’, which means it takes a while to test and optimize since the data needs to be collected over time until it can indicate the right strategy. This requires patience and the will to accept failures in the short run, but in order to avoid failure in the long-run, the right campaign goals need to be set. Some goals are too ambitious and cannot be achieved no matter how much money is Copyright © 2009. Diplomica Verlag. All rights reserved.

spent and how sophisticated the technology is. Furthermore the metrics to measure need to be set right in order to being able to optimize since a complete change of the strategy might let the campaign collapse completely. 165

166 167

164

Cp. Fösken, Sandra: Die Qualität der Zielgruppe ist ein wichtiger Aspekt, in: Absatzwirtschaft, Nr. 4, April 2007, p. 76 165 Cp. Tsai, Jessica, 2008, p.28 166 Cp. Schirnke, Marla: BT discipline for newbies, May 2008, http://www.imediaconnection.com/content/19242.asp / 23.08.2008 167 Cp. VanBoskirk, The Reality, 2006, p. 5

41

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Publishers have to consider if BT is the right choice for their web site or web sites. If the web site is very low in traffic because it attracts a very specific target group, e.g. guinea pig owners, but not varied types of customers, they should rather look for contextual targeting such as Google’s AdSense to target those ones who are specifically looking for guinea pig related issues. Once there is enough traffic on the webpage publisher should sell what types of customers visit their site not only how much traffic they get in order to become interesting for BT. This can of course only work when the publishers join a network in order to increase their reach and optimize frequency by combining their data with data of other web pages and therefore being able to create user profiles by monitoring their behaviour across the sites of the network. 168

5.4. The process of BT Referring to the requirements for successful BT in section 5.3, a brief description of an ideal process will follow, representing one of the latest approaches already widely accepted in the USA. This represents a summary of Matthewson (2005). Phase 1: Planning for online advertising optimization BT is similar to other marketing activities and therefore the first stage implies the setting of objectives. These objectives in online advertising vary from click maximization to conversion optimization, such as sign-ups, purchases and downloads. Different business rules such as different products within a campaign with different profitability must be taken into consideration. So in order simplify the decision-process and to take into account that different users have different behavioural characteristics and different business rules exist an ‘optimisation tree’ is recommended, a form of decision tree which maps the different possibilities. 169 Phase 2: Data acquisition The goal in this phase is to create user profiles based on behaviour. In order to achieve that a benchmark needs to be established which measure the performance of an online

Copyright © 2009. Diplomica Verlag. All rights reserved.

system without any segmentation. This base, where every user is treated equally, means that the same advertisements are shown to any user, allows the forming of clusters of users, which are most or least attracted by an advertisement and a ranking of the different groups can be made. Each cluster has a set of varying characteristics. After data collection a ‘media yield curve’ is generated, which shows the advertiser where a waste in

168 Cp. VanBoskirk, The Reality, 2006, p. 6 169 Cp. Matthewson, James A.: Behavioural Targeting: can online advertising deliver in 2006?, in: Journal of Direct, Data and Digital Marketing Practice, Vol. 7, No. 4, 2005, p. 333-334

42

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

non-responsive media emerges, meaning how much advertisement was lavished and which clusters are the most responsive while the data collection period. 170 Phase 3: Optimisation Once the results from the data collection period have been analyzed an optimization model needs to be established, which contains the decision-making rules in order to meet the marketing objectives. “This phase is often referred to as ‘placement optimisation’ and ensures that the right message, offer or creative is delivered to the right type of customer, at the right time and in the right placement.” 171 The model will provide information as to which advertisement should be shown to which cluster to achieve the highest RoI or absolute goal actions (e.g. purchases or newsletter sign-ups), depending on the campaign goals. The optimization model can react independently to the circumstance where a user profile fits to several offers of the advertiser and will pick to best solution according to the goal. A control group of 5-10 per cent of the overall impressions delivered is non-optimised and functions therefore as a control group in order to measure the performance of the optimisation model. 172 Phase 4: Result analysis The following metrics are used in online advertising to measure the effectiveness of a campaign: 1. Impressions delivered in relation to numbers purchased 2. Clicks achieved by offer/creative combinations (click-through rate or CTR) 3. Cost per desired action taken as a result of the campaign (cost per action or CPA) 4. Best-performing publisher/placement/creative combinations This data is however not rich enough since it does not provide enough information about a consumer’s behaviour online, therefore analysing the well-performing clusters enable a further dimension and allow intelligent inference with respect to the users.

Copyright © 2009. Diplomica Verlag. All rights reserved.

In order to avoid wastage, a so called ‘passback’ process comes into effect: Once a user’s profile does not fit to a cluster, the advertisement is not shown to him and no costs occur. Another option is the ‘pass-sideways’ where instead of not showing an advertisement at all to the viewer whose profile does not fit in the first place another advertisement which is more appropriate to the user’s interests is displayed. 173

170

Cp. Matthewson, 2005, p. 335-337 Matthewson, 2005, p. 337 172 Cp. Matthewson, 2005, p. 337-338 173 Cp. Matthewson, 2005, p. 338-339 171

43

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Referring to section 2.7. (Requirements for optimal media planning: Elements and process) the requirements and the process for BT fit very well into the general media planning since it implies a similar procedure and can therefore be easily implemented: 1. Setting objectives 2. Defining the target group 3. Research 4. Optimization 5. Evaluation of the campaign results. Compared to the requirements for media planning, two stages in the process are missing, firstly the determination of the budget and secondly the definition of the media strategy, the media mix. The budget cannot be determined within the BT process, it has to result from the media planning process and based on this the decision to which extent BT can be applied needs to be made. An indication how prices within the BT industry are calculated will be presented in section 6.2. Regarding the media strategy or the media mix respectively BT cannot meet the requirements for optimal media planning yet since it only comprises one medium, the internet. BT is therefore seen isolated, outside the media strategy with an integrated approach. So far BT cannot contribute to the realisation of the advantages of a mixed media approach as explained in section 2.5. which demonstrates a weakness of BT and will be further discussed in section 5.5. and also an attempt to overcome that obstacle will be presented in section 6.4..

5.5. A Behavioural Targeting SWOT-Analysis 5.5.1. Strengths

Copyright © 2009. Diplomica Verlag. All rights reserved.

The greatest benefit of BT is increased effectiveness of campaigns compared to nontargeted campaigns. Surveys carried out with online marketers both in the U.S. and Germany demonstrated the general acknowledgment of “(…) Behavioural Targeting as

44

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

one of the best-performing online advertising tactics (…)” 174 and they are very satisfied with the results. 175

176 177

The increased effectiveness is obtained by the full exploitation of the maximum reach within one target group while minimizing wastage. Focusing on users who already have a high affinity towards the advertised product or service can furthermore improve contact quality and increase advertising impact. At this point BT can address one of the greatest obstacles of marketing, which is identifying the right segments or the right target group respectively because the segments are created by the users themselves through their behaviour online and are not imposed by defining them based on market research results. The major problem for marketing specialists are the shifting human needs and therefore customers’ behaviour and their demands towards products constantly changes which means target groups are never static. BT provides an answer to that problem since it can adjust the placement of advertisement in real time and furthermore equip marketers with the necessary tools in order to learn more about their consumer and even correct false assumptions about who the target group is. 178 The target groups, which have been assumed by the strategic planner of the advertisement agency, might not be a 100% congruent with the real target groups, and here BT makes it possible to react according to those lessons learned in realtime and optimize the campaign while it is running, constituting a huge strategic advantage. 179

180 181

Reach and identifying the right target group are not the only important objectives of media planning, furthermore the frequency and the continuity is of great interest. Compared to traditional media the control of frequency is much more effective. The advertiser himself can decide how often the individual user should see the advertisement and therefore he can determine the frequency himself via a technological feature called frequency capping. For instance within the TV medium it cannot be said how often a consumer actually sees an advertisement, ad servers can control how often a user sees an advertisement based on frequency capping. Along with that goes the continuity, which can also be defined

Copyright © 2009. Diplomica Verlag. All rights reserved.

through the delivery logic of the ad server (e.g. continuous, flighting or pulsing). As both

174

Hallerman, 2008, p. 5 Cp. (JupiterResearch) http://findarticles.com/p/articles/mi_m0EIN/is_2006_June_14/ai_n26896408 / 24.08.2008 176 Cp. Market Report 2006, p. 25 177 Cp. Hallerman, 2008, p. 5-6 178 Cp. Gossmann, 2008, p. 1-2 179 Cp. Tsai, Jessica, 2008, p.26 180 Cp. Market Report 2006, p. 19 181 Cp. Hallerman, 2008, p. 10 175

45

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

frequency capping and delivery logic are features of normal ad servers, these advantages are properties of online advertising in general and not specific to BT. 182 Importantly, BT addresses one of the major issues of media planning: Efficiency, the relation between obtained impact and budget. Various practical examples demonstrate that BT leads to an increase of efficiency and precision. 24/7 Real Media for instance has analyzed the campaign results of 800 campaigns where BT has been used and concluded that BT drastically increased cost efficiency. 183 Furthermore, BT can achieve higher conversion and click rates since the campaigns are highly relevant to the user and it is much more probable that attention is attracted, interest aroused and desire created. This causes a distinct increase in branding effects and a higher user involvement compared to non-targeted campaigns. 184

185

This represents “a win-win for marketers and publishers” 186 , as marketers can reach the right target audience with a highly relevant message, which will attract more attention and increase efficiency at the same time and the publishers can sell ad inventory, which formerly has been unmarketable or hard to sell because the web page is not compatible with contextual targeting (e.g. weather forecasts, TV programming, webmail clients etc.) Furthermore advertisement which is behaviourally targeted can be sold at a higher CPM compared to non-targeted campaigns. Another assumed advantage is that users consider behaviourally targeted advertisement more tolerable, because it addresses their interests, minimizing irritation. 187

5.5.2. Weaknesses The major problem concerning BT is the achievement of great reach and broad profiles. While the reach in certain target groups can obtain a very high level of coverage within the total population of the target group online, the affinity-based classification filters do not easily lend themselves to mass advertising. The combined criteria can narrow down the reach and consequential frequency so much that any acceptable reach cannot be

Copyright © 2009. Diplomica Verlag. All rights reserved.

achieved. This happens because the BT-enabled networks contain only a specific number

182

Cp. Skou, Kasper, Interview 02.9.2008 Cp. Shields, Mike: Study: Behavioural Targeting Succeeds on Web, February 2006, in: http://www.mediaweek.com/mw/news/interactive/article_display.jsp?vnu_content_id=1002074845 / 24.08.2008 184 Cp. Längin, Stefan: Mehr Effizienz durch Behavioural Targeting?, January 2006, in: http://www.adzine.de/de/site/contentfcmsv443d929d6473ee/artikel.xml / 24.08.08 185 Cp. Market Report 2006, p. 25 186 Cp. VanBoskirk, The Reality, 2006, p. 3 187 Cp. Hallerman, 2008, p. 7 183

46

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

of web pages and a behaviourally targeted campaign does not reach beyond the borders of the chosen network. 188

189 190

BT requires a great amount of data and a very sophisticated technology, which still undergoes fundamental development. The outcome of this is that behaviourally targeted campaigns can only be as good as the data and the technology executing it, raising quality issues as a further concern about BT. 191

192

BT, being still at the developing stage, is lacking standards and transparency about the different options and technologies within the industry. Marketers do not realistically have the possibility to compare providers and their options and decide on the best solution for them. The validity of the behavioural classifications is questioned, since the underlying assumptions are not transparent and therefore the BT tools are called a “black box” 193 . The question remains how the parameters are defined, and therefore the target group segmentation remains unclear. 194 Another weakness lies in the structure of the industry’s value chain, as explained under section 3.3. As discussed one great advantage of BT is that it is a learning system meaning the target group definition and description can be adjusted while the campaign is running. Unfortunately there is no connection between the BT provider and the strategic planners in the advertising agency 195 who are responsible for setting up a campaign strategy based on the target groups attitudes and interests. Being able to adjust the target group during a running campaign would also imply that an adjustment of the strategy is possible and resulting from at least a slight modification of the campaign in order to achieve better results. Currently this is not happening, not even knowledge from former campaigns based on BT is passed on to strategic planners. This is unfortunate, as the extra data would very well complement the application of market research studies, because data gathered online is quite often more topical and closer to reality. Furthermore with reference to section 2.5., the advantages of a mixed media strategy cannot take effect in the case of BT with its current approach. Offline and online advertising are carried out too isolated from each other instead of planning them in conjunction in order to avoid

Copyright © 2009. Diplomica Verlag. All rights reserved.

saturation and wear-out effects in one medium, enable synergy effects, save costs by achieving frequency in a second, less expensive medium and increasing creative

188

Cp. Bannan, Karen J.: Behavioural targeting, in: B to B, Vol. 92, Issue 7, 2007, p. 18 Cp. Market Report 2006, p. 26 190 Cp. Bryant, 2008, p. 42 191 Cp. Market Report 2006, p. 26 192 Cp. Hallerman, 2008, p. 12-13 193 Fösken, Sandra, 2007, p. 74 + Welch, Sarah: 3 themes shaping behavioural targeting, 2008, in: http://www.imediaconnection.com/content/18549.asp / 24.08.2008 194 Cp. Fösken, Sandra, 2007, p. 74 + Welch, Sarah: 3 themes shaping behavioural targeting, 2008, in: http://www.imediaconnection.com/content/18549.asp / 24.08.2008 195 Cp. Skou, Kasper, Interview 02.9.2008 189

47

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

effectiveness by mixing different media channels. In addition some target persons might not be reachable with online advertising, for instance elderly people, and need to be targeted through another medium, but since the knowledge gained through BT about the target group online is not exchanged with the offline planner, the advantages through a media mixed strategy are not supported by the BT technology. Therefore the lack of connections between the individual industry players is currently a great weakness. Facing the fact that there are no standards yet, the marketers have to accept higher costs since behaviourally targeted advertisements are sold for a higher CPM than non-targeted campaigns as already explained, but do not have the full proof yet that the advantage in effect can compensate the higher costs. 196 One technological problem for cookie-based BT is the fact that the cookies can be erased by the user, rendering all behavioural history lost, decreasing the reach of a given BT network. Furthermore there are different circumstances where BT can not be effective, for instance when a user simply accesses another computer, or when the user is completely new to the BT tool and no data concerning his behaviour is available in order to target him with the right advertisement. Relatedly, if more than one person uses a computer, for example in a family, all those user belong to different target groups, but since they share a cookie, BT technology can only react sensibly to their behaviour in real-time but not according to former surfing behaviour. 197 Additionally, the BT tool does not know what the intention of the user is: Is he searching for something work-related or for private use, for himself or someone else? The circumstance determines if a behaviourally targeted advertisement is relevant to him or not. Finally, there is a lack of proof regarding impact, control of success and independent scientific studies in order to give hard evidence of the efficiency of BT. Providers of the technology tend to publish only their successful results, meaning there are hardly any reports about failures with BT. For this reason the success stories of some campaigns

Copyright © 2009. Diplomica Verlag. All rights reserved.

have to be examined critically. 198

199

5.5.3. Threats There are several threats which currently slow down the growth of BT. One major concern for marketers is the undefined term of BT and its various meanings and implications, since

196

Cp. Market Report 2006, p. 31 Cp. Tsai, Jessica, 2008, p.26 198 Cp. Mühling, 2007, p. 113 199 Cp. Market Report 2006, p. 30 197

48

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

there is a wealth of approaches and technologies within the industry. The problem “(…) about behavioural targeting right now is that it’s really ill-defined,(…)” 200 , therefore many marketers do not understand what exactly BT is although many know the expression. It is connected to consumer segmentation and their interests, but also quite often mixed up with other targeting methods such as contextual targeting. This confusion causes scepticism with marketers since they are not sure how to adopt BT within their campaigns and what exactly it can do in order to increase the success. Also, a central problem is whether their organisation can actually handle the great amount of data required in order to behaviourally target, so “many marketers still see behavioural targeting as experimental” 201 and are currently still too reserved to consider it for their campaigns. 202 203

A further threat and another reason for reluctance by the marketers is the way BT places ads in alien contexts. Especially when a company is focused on brand management they fear that advertisements shown in alien contexts might harm the brand image if the context is not appropriate. 204 This implies that BT is currently only able to track where a user is going on the net and based on that behaviour place the ads but the technology cannot ‘understand’ at what point which advertisement might be counterproductive according to the achievement of making a positive impression. Relatedly, some publishers feel that BT is downgrading the brand of their pages. As BT is context-independent, it focuses on the users to the detriment of the sites. Taken in combination with the fact that most BT is traded through networks, the concern is that premium publisher’s brands might become commoditized and lose their ability to command high CPMs, exacerbating the reluctance to join the BT bandwagon. 205 This is a threat to BT, as fewer web pages imply fewer possibilities to target consumers and make BT less efficient, particularly in the case of special content web pages. A different type of issue, which hold advertisers and publishers back, are concerns about invasion of consumer privacy. This is based mainly on insecurity, but it makes advertisers and publishers think twice about using BT. This is based on current developments since

Copyright © 2009. Diplomica Verlag. All rights reserved.

data protection advocates initiated an investigation on European level, an EU-panel will examine if target methods including BT violate consumers privacy. If found to comply, the provider will be eligible for the European Privacy Seal (see www.european-privacy-

200

Tsai, Jessica, 2008, p.26 Cp. Hallerman, 2008, p. 4 202 Cp. Blythe, Alex: Predict a click, in: Revolution, April 2008, p. 53 203 Cp. Market Report 2006, p. 23 204 Cp. Hallerman, 2008, p. 4 205 Cp. Hallerman, 2008, p. 4 201

49

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

seal.eu). Laws concerning targeting in the U.S. will restrict BT in the future when a proposed legislation will pass through the Federal Trade Commission. 206

207 208

The BT industry denies violation of privacy since BT can only track where a consumer has been online; the person itself is not identified by personally identifiable information (PII). It is currently debated in the EU whether the IP-address of a user is PII. If found to be so, this will render all approaches based on the IP-address of the user illegal. 209

210

However, advertiser and publisher are sceptical since the discussion is ongoing and not settled. They are afraid that they might anger their customers or the public in general, harming their brand and singling them out for sanctions. 211 So what do consumers really think? Are they positive about advertisement, which is more relevant to them, or rather disturbed and feeling pressured? A survey was carried out in order to examine consumers’ reaction towards online behavioural tracking and targeting, revealing that consumers are concerned about their privacy but not alarmed. Furthermore, although the examined persons were familiar with the technology about tracking and targeting and knew they could prevent it if they delete cookies, they were not so concerned about their privacy that they actually took advantage of that possibility. 212 In the final analysis consumer attitudes towards targeting does not seem to be the greatest problem. However, most online users are not conscious of the fact that they are targeted and therefore the question arises if not only more explanation for marketers is required but also for the consumer.

5.5.4. Opportunities BT can use several opportunities in order to realize further growth. First of all compared to its early years BT has vastly improved, especially concerning the management of large data bodies in order to define target groups precisely. 213 This development will, embedded in the improvement of online advertising in general, further progress and address both concerns of advertisers and publishers about the efficiency of BT and difficulties in data

Copyright © 2009. Diplomica Verlag. All rights reserved.

management.

206

Cp. More scrutiny for online ad targeting, March 2008, in: http://www.mrweb.com/drno/printable/pn8116.htm /11.06.2008 207 Cp. Eu-Gremium prüft Targeting-Methoden, in: Der Kontakter, Nr. 9, February 2008, p. 32 208 Cp. Hallerman, 2008, p. 4 209 Cp. Welch, 2008, p. 2 210 Cp. Denker, Helge: Werbung, die (ver)folgt), March 2006, in: http://www.adzine.de/de/site/contentfcmsv44419c077c4ed4/artikel.xml / 25.8.2008 211 Cp. Welch, 2008, p. 2 212 Cp. Alreck, Pamela L./Settle, Robert B.: Consumer reactions to online behavioural tracking and targeting, in: Database Marketing & Customer Strategy Management, 2007, Vol. 15, 1, p. 11-23 213 Cp. VanBoskirk, The Reality, 2006, p. 3

50

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Major web players such as Google, Microsoft and Yahoo! are investing large sums in order to develop their advertisement networks to be able to address deficiencies regarding amount and qualities of data. 214 This might also solve the problem of too limited reach, as greater advertisement networks with BT mean greater reach. More and more web pages will be included in behaviourally targeted advertisement campaigns, which imply that those campaigns will then be able to reach a mass audience with the advantages of BT which were explained in section 5.5.1. Another way to a better utilise BT is the combination with other targeting options such as socio-demographics and psychographics as explained in section 4., in order to refine the target group since the behaviour cannot give information about all aspects which are necessary to characterize a person’s interests. One such approach is called Target Group Planning, which is developed and marketed by United Internet Media in Germany 217

215 216

and/or contextual targeting. “In fact, behaviour (which targets people) and context

(which targets pages) are two buckets that necessarily leak into each other.” 218 These blended campaigns address the concern of brand managers, who fear that their brand can be damaged if advertisements are placed in contexts, which contradict the perceived quality of the brand too much. By layering BT on top of contextual targeting, which is possible through a vertical network, meaning a range of websites, which cover one field of interest, e.g. travel, very widely including very specific websites 219 , advertisers can target their audience more efficiently, especially for niche products and services. However, by combining two targeting methods once again the reach could be reduced drastically – a problem, which could be ameliorated by greater advertisement networks. 220 As a remark, even though a combination of both BT and contextual targeting might remove certain constraints, it has been proven that behaviourally targeted advertisements are more likely to receive attention than contextual targeted advertisements, “in fact, behavioural outperformed contextual by as much as 22 percent.” 221 One key driver for the growth of BT might be the ambition of advertisers and marketing executives to personalize the relationship with their audience. Along with that goes the Copyright © 2009. Diplomica Verlag. All rights reserved.

rising influence of social networks online, meaning the possibility for consumers to interact on the internet, sharing ideas, contributing personal experiences and searching for 214

Cp. Hallerman, 2008, p. 13 Cp. Market Report 2006, p. 19 216 Cp. Bryant, 2008, p. 42 217 Cp. Längin, 2006, p. 1 218 Cp. Hallerman, 2008, p. 9 219 Cp. Lawson, Cree: Why to choose a vertical ad network, in: http://www.imediaconnection.com/content/12998.asp / 07.09.2008 220 Cp. Hallerman, 2008, p. 9 221 Cp. Jupiter Research: Behavioural Targeting & The Purchase Funnel Opportunity - Consumer survey prepared for AOL + RevenueScience, April 2007 215

51

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

people, events and topics, which are personally important. Web pages such as Facebook, YouTube and MySpace have grown explosively over the last couple of years, and the advertisement industry has realized their role in establishing a personal connection to the consumer. A predicted growth from $1.2 billion in 2007 to $2.2 billion advertisement spending in 2008 on social networks (with the majority in the U.S.) demonstrates the importance of these new media. In order to increase the efficiency of the rapidly growing advertisement spending BT might play a major role, since social networks provide a huge amount of data in order to target customers precisely based on their shared interests and behaviours. 222 Furthermore developments in media technology will be other key drivers for increased importance of BT. One is the mainstream implementation of online video advertising, which is more expensive than other online advertisement formats. Therefore BT is very interesting to marketers, since they will try to operate their campaign as efficiently as possible and target the right audience with a minimum of waste. Another advertising channel is connected to mobile devices, such as cell-phones and blackberries, since they are very personal artefact connected strongly to the individual. Therefore mobile marketing will be the next important issue for BT, although it is currently at a very early stage and there are concerns about what the user is willing to accept concerning advertisement on his mobile phone, reflecting the highly personal nature of the device. It is clear though that mobile technology enables the user to spend less time in front of a normal computer and therefore will detract media time spend from traditional online, decreasing the reach and frequency obtainable in this channel. Since the user will be very sensitive towards advertisement on his mobile phone, he might be less sensitive towards behaviourally targeted advertisements, which could help spur mobile marketing growth in

Copyright © 2009. Diplomica Verlag. All rights reserved.

the future. 223

224

222

Cp. Gossmann, Bill: Behavioral Targeting – what we’ve learned so far, January 2008, in: http://www.imediaconnection.com/content/17905.asp / 25.8.2008, p. 1 223 Cp. Hallerman, 2008, p. 1 224 Cp. Gossmann, 2008, p. 1-2

52

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

6. Current state of affairs in BT 6.1. The international development of Behavioural Targeting Section 3.3. of this paper provides the foundation for the assessment of the international situation concerning the diffusion of BT worldwide. As well as in online advertising in general the U.S.A. is leading the way regarding BT, while Europe’s development is very diverse. In the U.S. spending for behaviourally targeted advertising reached just over $500 million in 2007 and will reach $775 million in 2008, which implies a contribution to the overall online advertisement spending of only three percent. Although those numbers are the highest in the international comparison, they are still considered as being relatively low, with an even smaller growth potential in 2009. The given explanation are the objections against BT discussed in section 5.5.3, the concerns about the commoditization of online inventory and the privacy of users as well as the confusion about the options and effects of BT. However, it is forecasted that BT will overcome those concerns and further develop the technology. Additionally the influx of branded advertising spend to online will support a much greater increase of advertisement spending in the years 2010 until 2012 to up to $4,4 billion. 225 The greatest providers for BT technology in the U.S. are Revenue Science, Tacoda and 24/7 Real Media. 226 Interestingly, only Revenue Science remains independent today, as

Copyright © 2009. Diplomica Verlag. All rights reserved.

Tacoda has been acquired by AOL and 24/7 by WPP.

225 226

Cp. Hallerman, 2008, p. 4-5 Cp. VanBoskirk, The Reality, 2006, p. 4

53

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Figure 7: US behaviourally targeted online advertising spending 2006-2012

227

In Germany and the UK the development is not that progressed yet as in the U.S. but it is at the highest level across Europe. However, the spending for behaviourally targeted advertising is currently so modest that there are no measurements as to the actual spending 228 . In Germany a heterogeneous supplier structure characterizes the market with every BT provider offering its own technology and therefore concept, but it is mostly dominated by a few large companies such as Deutsche Telekom’s sales house Interactive Media labelling their BT approach User Centric Advertising, and United Internet Media with Target Group Planning™. This is in contrast to the US and UK, where large advertising networks dominate. This is caused by the extreme industry concentration in Germany combined

Copyright © 2009. Diplomica Verlag. All rights reserved.

with the fact that large media companies have developed proprietary targeting offerings early (often in collaboration with specialist BT vendors). Although the structure within the market is very heterogeneous it is considered interesting concerning its future development since the different approaches of BT are already quite sophisticated and

227 228

Cp. Hallerman, 2008, p. 4-5 Cp. Skou, Kasper, Interview 02.9.2008

54

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

therefore further innovations and improvements of the existing technologies can be expected as well as further market participants. 229 According to the Interactive Media/Enigma GfK Market Report 2006, 82 percent of the polled experts have already used targeting in general for their online media planning, while behaviourally targeted advertising has the highest usage percentage of 69% compared to other targeting methods. The experts using targeting for their media planning estimated an average share of 21% targeting spending of total online media spending.

230

While the German market is characterized by BT providing companies with German origin such as wunderloop and nugg.ad, the market in the UK is dominated by the U.S. based company Revenue Science, which already stated in 2006 that it reaches 84% of the population in the UK enabled by a partnership with Advertising.com. 231

232

The market for

BT in the UK is rapidly growing and both Germany and the UK are seen as early movers in the overall European market. BT is slowly proliferating through central Europe and Scandinavia, and Turkey and Poland are making headway entering the BT business. Statements about the state of affairs concerning the rest of the world are currently not made by experts; therefore a much slower development can be expected. 233

234

6.2. Demands on Behavioural Targeting and recommended course of action for media planners Although 86% percent of the experts in the Interactive Media/Enigma GfK Market Report 2006 state that the importance of BT will increase in the future and even 40% are convinced that it will be one of the most important factors within online advertising, the media planners are not very well educated or completely convinced by the advantages of the technology. The Market Report 2006 highlighted their demands of the BT industry: Very important to the media planners are valid data quality of the targeting criteria, monitoring and adjustments of running campaigns, the reach within the target segments and the optimization of the most effective frequency. Furthermore the flexible combination

Copyright © 2009. Diplomica Verlag. All rights reserved.

of targeting criteria and the placement across several web pages, the protection of the

229

Cp. Mühling, 2007, p. 115 Cp. Market Report 2006, p. 24-29 231 Cp. Business Wire: http://www.businesswire.com/portal/site/google/index.jsp?ndmViewId=news_view&newsId=20060926005456 &newsLang=en / 30.08.2008 232 Cp. BNet: http://findarticles.com/p/articles/mi_m0EIN/is_2007_May_31/ai_n27258672 / 30.08.2008 233 Cp. Mason, Jeremy: The past, present and future of behavioural targeting, October 2007, in: http://www.imediaconnection.com/content/16891.asp / 30.08.2008 234 Cp. Skou, Kasper, 02.09.2008 230

55

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

user’s privacy, transparency about determination of targeting criteria as well as standardization of the process. 235 While the BT industry is already able to meet some of these requirements, others will be addressed in the future depending on the development of technology as already discussed in previous chapters. Generally, although those concerns still exist, BT is recommended as an option, but in order to make it applicable both for advertisers and publishers a well-planned procedure is advisable. The following recommendations are embedded in the requirements for successful BT and its process (see section 5.3. and 5.4.). For advertisers, and consequentially for media planners, a well-defined campaign goal is essential, this implies the success parameters for quantitative and qualitative goals and a clearly verbalized brand message. Referring to the campaign goals the definition of the target group is of very high importance. Along with that details concerning the dimensions of the target group need to be clarified. The smaller the target group is and the harder it is to find this target group on specialised, vertical sites (e.g. women interested in mobile phones and lifestyle), the better it is suited for a behaviourally targeted campaign compared to a mass campaign from a financial point of view. The next step then is the choice of the BT provider. First of all it needs to be clarified if a great reach is of high importance since some provider can achieve more than others. This can be researched via internet surveys, which are published on a yearly basis. Furthermore the decision as to which forms of targeting offer the best way to reach the set campaign goals needs to me made. Since all providers vary in their options and often combine targeting forms, for instance BT with contextual targeting, a comparison is necessary. The reachable target group has to meet the publisher’s minimum volume and the advertiser’s target group definition has to be realisable in the BT technology chosen for the campaign. Once providers are shortlisted as being suitable, prices need to be negotiated in order to maximize the yield from the campaign’s budget. Some providers have a fixed CPM per Copyright © 2009. Diplomica Verlag. All rights reserved.

target group, while others (e.g. Interactive Media) increase prices proportionally with the number of attributes used in the target group definition. Lastly, an operating scenario needs to be prepared for optimization, meaning how the data, targeting criteria and goals can be adjusted in a running campaign. The publishers, acting as a counterbalance to the advertisers, need to assure that the campaign goal is clear to them and the data is available or can be sourced short-term. 235

Cp. Market Report 2006, p. 28-30

56

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Furthermore they need to clarify if the price-performance-reach-relation is appropriate, if they or the BT provider executes the optimization and how the procedure will be carried out. Expectation management of the advertisers is a task for both publishers and the BT provider relying on empirical data from other campaigns. 236

237

6.3. Case Study In order to illustrate the operating mode and execution of a behaviourally targeted campaign as well as its results, two case studies are presented and explained in the next paragraph. These studies are of special interest because in both cases the study has been carried out with FDM travel as the advertiser, one of the leading travel agencies in Denmark, working with two different BT providers successively In the first case Ekstra Bladet, one of Denmark’s largest online newspapers decided to collaborate with nugg.ad and their predictive BT technology in order to target more efficiently. Before then they had used contextual targeting and were not satisfied with the outcome. The nugg.ad technology collects user data online in real time and enhances it by using blind gifs, invisible graphic files in order to count the number of page views per cookie and to register the page code. In order to enhance the collected data online surveys are carried out with selected persons to gather more information concerning sociodemographics, lifestyle and interests. Those results are then combined with the online measurements and profiles are generated, which can now be reached with

Copyright © 2009. Diplomica Verlag. All rights reserved.

advertisements on a individual level. 238

236

Cp. Mühling, 2007, p. 118-120 Cp. Hegge, 2007, p. 295 238 Cp. nugg.ad: http://www.nugg.ad/en/products/technology.html / 29.08.2008 237

57

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Figure 8: nugg.ad’s technology system

239

Nugg.ad’s case study using the site ekstrabladet.dk in cooperation with FDM travel was performed in January 2008. The products advertised by FDM travel were getaways and city tours both in the behaviourally targeted campaign and the untargeted control campaign, and the campaigns were identical in format, content and advertisement impressions in order to be comparable. The test period comprised two days The first stage in the study was the presentation of the campaign to users which have been predicted by the nugg.ad technology as being ‘highly or very highly interested’ in short holiday trips. Compared to the untargeted campaign a 36% percent higher click-rate could be achieved. The second step was the combination of the criteria ‘high interest’ and ‘high household income’, which enabled an increase of 56% in the click-rate. As soon as the criterion ‘high household income’ was used by itself, the click-rate only increased by 6%. The next step therefore was the combination of three criteria, ‘high interest’ and ‘high household income’ as well as ‘female user’ based on the assumption that women are more interested in getaways than men. The result was a very high increase of 180% in click-rates compared to the untargeted group. Using female gender as the only criterion Copyright © 2009. Diplomica Verlag. All rights reserved.

achieved an increase of 83%. Based on those results, nugg.ad argues that “the intelligent combination of relevant targeting criteria proved to be the decisive factor.” 240 Criteria such as demographics and

239

Cp. nugg.ad: http://www.nugg.ad/en/products/technology.html / 29.08.2008 Nugg.ad: http://www.nugg.ad/fileadmin/redakteure/downloads/Case_Studies/ 080505_Case_Study_Ekstrabladet_de.pdf / 29.08.2008 240

58

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

interests are the result of predictions, which are possible through a combination of online measurements and online surveys. 241

Figure 9: Evaluation of the campaign results 242

The results prove on the one hand the advantages of BT, the higher flexibility of online advertising since the technology allows reacting in real time to measurement results and adjusting the campaign parameters, and furthermore to learn more about the target group. The fact that women are more interested in short trips were initially an assumption, and could then be confirmed by the behaviourally targeted advertisement in a relatively short timeframe. Overall it can be said that the behaviourally targeted campaign was successful

Copyright © 2009. Diplomica Verlag. All rights reserved.

and showed an effect since an increase of the click-rate of up to 180% percent could be demonstrated. The case study also showed a disadvantage of predictive BT: The target group definition is driven by hypotheses. nugg.ad hypothesised that women are more interested in short holidays than men, and afterwards had to test this hypothesis empirically. The more unclear the actual target group is (just think of marketing laser eye surgery), the more 241

Cp. Nugg.ad: http://www.nugg.ad/fileadmin/redakteure/downloads/Case_Studies/ 080505_Case_Study_Ekstrabladet_de.pdf / 29.08.2008 242 Ibid.

59

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

cumbersome and costly this hypothetical approach will be. Also, the predictions of nugg.ad rely on their system and are therefore only as good as their technology. It is unclear why their predictions are the right ones and why other combinations of criteria might not be more efficient than the ones chosen by the system. However, although the click-rates could be increased by a very high percentage, FDM Travel decided to leave nugg.ad and their technology simply for the reason that it was far too expensive compared to the sales results. Therefore FDM travel decided to test wunderloops ‘Intelligent Re-targeting’ or micro-targeting approach instead. In this case the publisher was krak.dk, the major Danish online yellow pages. The goal of the campaign was to re-target those customers, who have already shown interest in a specific travel destination but then left the FDM Travel site without buying. Krak.dk sees around 85% of the Danish internet population and is therefore very suitable for retargeting. wunderloop’s technology was implemented on the FDM Travel site and profiled users’ behaviour on the site. Thus if a user searched for hotels in New York and then for flights to and hotels in Mallorca, the wunderloop system would determine that this user had a prevailing interest in Mallorca and send this information to the ad server. As the user surfed to krak.dk he was recognised and his profile looked up in the wunderloop system, which would return the keyword ‘mallorca’. A dynamic banner generation engine created a Mallorca banner out of digital assets provided by FDM

Copyright © 2009. Diplomica Verlag. All rights reserved.

Travel, and this micro-targeted banner was shown to the user.

60

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Figure 10: wunderloop’s intelligent re-targeting or micro-targeting

243

The results of this case study also relied on a control group, and the comparison showed a slight increase in click-trough-rate, so lower compared to the percentage nugg.ad achieved. However, what was of real interest was measured further down to conversion funnel: wunderloop also measured the click-to-conversion rate, and this was increased by a factor of 20 or 2000%. 244 The wunderloop case demonstrates that the criteria used and the implementation of BT can be transparent and very straightforward and compared to the nugg.ad case the measured numbers are much more significant to the advertiser FDM Travel, since they express what is most important for a direct marketer online: Sales. High CTR numbers do not seem to have a great correlation to the conclusion of a transaction, a number which the industry widely considers being very important – maybe not least because it is extremely easy to obtain. The wunderloop/FDM Travel study shows that this key performance indicator might not be very viable as a measure of actual campaign results

Copyright © 2009. Diplomica Verlag. All rights reserved.

and for calculation of campaign RoI.

6.4. Future outlook Although BT today faces some obstacles to realizing more extensive growth, the importance of behaviourally targeted advertising will increase; the development can therefore be forecasted as being positive. The following quotes give an impression of the 243 244

Skou, Interview 02.09.2008 Cp. Skou, Interview 02.09.2008

61

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

atmosphere within the industry. Jack Smith (24/7 Real Media): “In the coming years, behavioural targeting will continue to pervade digital media, enabling more effective microtargeting of audience across such digital marketing platforms as search or video, plus user-generated content such as blogs, wikis and podcasts.” 245 And further Bill Gossmann (CEO of Revenue Science): “Behavioural Targeting is being driven by demands of both consumers and the market. […] As IP-based media such as mobile and IPTV become more prevalent, the need (for BT) will only increase.” 246 New forms of media will therefore be one driver of the future development of BT. Experts within the industry have understood the importance and the potential influence of BT: According to the ENIGMA GfK study 93% of the interviewed experts see targeting in general as one of the most important factors in online advertising. However, this will only be realised if the industry establishes more transparency and standardisation, further clarifying the facts behind the technology. 247 Along with that might go a fundamental change within the BT industry. Most BT approaches are predicated on what might be termed ‘the affinity hypothesis’: If a user consumes a lot of finance content or answers yes to being finance interested in a questionnaire, this user will also be more likely to click on a finance-related ad. So far the BT industry has had a very hard time proving this is the case. 248 However, according to a Starcom USA, Tacoda and ComScore study from February 2008, the phenomenon “Natural born clickers”, a group representing 6% of all online users, are responsible for 50 percent of all click-throughs. The problem with ‘natural born clickers’ is that they only click on the advertisements but never buy, meaning there is no conversion. 249 This implies that the currently industry-wide used key performance indicator CTR does not express anything about the actual result; the Return-onInvestment cannot be expressed by it. What the BT industry needs to learn from this is to measure what actually matters. CTR will not continue to be a viable KPI for measuring campaign performance – we need to focus on the users actually performing the task we wanted them to as a result of the Copyright © 2009. Diplomica Verlag. All rights reserved.

campaign. That means profiling the users of the goal page and finding more of these users in the universe covered by the BT vendor. Instead of relying on hypotheses of who the target group is, BT should empirically discover this and use the discovery as well for 245

Cp. Anfuso, Dawn: Where does BT go from here?, November 2006, in: http://www.imediaconnection.com/content/12466.asp / 05.09.2008 246 Cp. Anfuso, Dawn: Where does BT go from here?, November 2006, in: http://www.imediaconnection.com/content/12466.asp / 05.09.2008 247 Cp. Market Report 2006, p. 36 248 Cp. Skou, Interview 02.09.2008 249 Cp. Moran, Gwen: Natural Born Clickers, September 2008, in: http://www.entrepreneur.com/magazine/entrepreneur/2008/september/196280.html / 05.09.2008

62

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

campaign optimization as to inform the media planner, the advertiser and the creative agency of these learnings. Profiling clickers as well and comparing these to the converters will give valuable insight as to whether the creative appeals to the right target groups. This learning should be fed back to the creative agency. With regard to section 2.5. (Media selection programmes), the empirical evaluation of the goal page will compensate the disadvantage of the current approach in order to select media online, the ranking method. While the ability to optimize a running online campaign based on data in real time, which for BT means that the definition of the target group can be adjusted, constitutes a great advantage compared to offline optimization, it still implies wastage in the optimization period. Evaluating the goal page and defining the actual target group before the campaign runs addresses these issues and makes optimization dispensable, representing a substantial saving on the media budget. Looking at the broader issues facing BT, mainly reach and the depth and quality of profiles, behaviourally targeted ad exchanges – such as wunderloop’s Connect – seem a very interesting development.

By aggregating reach across many different sites and

sharing the profiles between these sites, the BT ad exchange sees a greater part of the individual user’s behaviour and makes him targetable in many more places. Furthermore, it will enable a substantial reduction of the transaction costs in online media booking, since the media agency only has to upload the creative into one system, from which it is proliferated to all the publishers where the target group can be found. As the aggregation into a single point of sale is done via technology and not expensive human resources, fees for using this kind of technology can be reduced, leaving more value to be distributed between the advertiser, media agency and publisher. Another efficiency gain comes from the fact that only the individual users fitting to the target group need to be purchased. For that reason the effective CPM of the publisher can go up, while the effective CPM for the advertiser (the price paid for actually putting a thousand ads in front to the target group) can go down – and the user sees more relevant content.

Copyright © 2009. Diplomica Verlag. All rights reserved.

Looking at the relationship between media consumption (25% going to online) and media ad spend (10% going to online), it is clear that an imbalance exists. This imbalance can be explained by the lack of branding budgets going into online, and I would argue that the situation arises because the online advertising industry today does not adequately support the needs of brand advertisers. Two developments are necessary here in order to shift significant branding budget online: Firstly, online has to make available to branding advertisers single points of sale offering reliable and relevant targeting (such as socio-demographic and product-related attributes) 63

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

with a reach close to that of TV. This is particularly important for global players, e.g. Beiersdorf or Coca-Cola, who do not directly sell over the internet but instead aim to influence customers earlier in the AIDA-cycle. Secondly, online has to furnish the demand-side with easy-to-use and inexpensive tools to measure the brand-impact of online branding campaigns. Interestingly, ad exchanges with targeting have the potential to deliver on both requirements, but so does very large, technologically sophisticated networks such as Google, Yahoo! and Platform A (comprising AOL and advertising.com). A third possibility is for a technology owner to make a targeting standard open – in Germany an example of this is United Internet Media, which has declared its Target Group Planning methodology an open standard in the hope of winning significant market share to achieve this position. 250

251

One development which already becomes apparent might change the entire advertising industry, not only the online media, and might especially influence the media planners. According to the latest findings of the Yahoo! Close-the-loop-study integrated campaigns of display (banner advertisement) and search-campaigns (search-engine marketing) result in a 1560 percent increase of reach and an increase of two percent of the conversion rate as well as an increase of page impressions by 390%. 252 Integrating both display and search campaigns means leading the consumer through the entire AIDA process, so while ‘attention’, ‘interest’ and ‘demand’ is covered by the display campaign, ‘action’ is covered by the search campaign. This trend can be seen for instance by Google’s acquisition of DoubleClick in order to provide one single interface for integrated campaign planning. As already explained in section 3.3 Google aims at changing the value chain of the entire industry through disintermediation, enabling either the advertisers or the creative agencies to set up their campaigns themselves with attendant savings. But not only has this formed the online media since Google is the most powerful player in the industry, the company has also acquired a radio and print platform and investment in TV is planned. Google hereby redefines the rules of the business, since its platform enables advertisers and Copyright © 2009. Diplomica Verlag. All rights reserved.

advertising agencies to run integrated campaigns across all channels in coordination without any media agencies involved. It will enable the realisation of advantages of a mixed-media strategy as presented in section 2.5. in conjunction with BT. As discussed in section 5.5.2 this is currently not the case within the BT industry, since the different approaches are handled isolated from offline advertising.

250

Cp. Skou, Interview 02.09.2008 Cp. United Internet Media, in: http://united-internet-media.de/tgp.html/ 10.09.2008 252 Cp. Yahoo!Close the loop, 2007, p.2 251

64

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

As Google is probably the company in the world that knows the most about the largest group of the global online population, their next move in targeting will be very interesting to watch. Recently they launched a product in beta called AdPlanner for sociodemographic targeting across the Google publisher network, marking the first step in attracting more branding money, and this will undoubtedly continue to develop as Google brings more and more display functionality to the integrated platform through the DoubleClick acquisition. 253 This development seems inevitable, and indeed today media agencies are threatened on their raison-d’être; therefore they need to focus on the improvement of their advisory performance. They need to move out of the increasingly commoditized function of simple

Copyright © 2009. Diplomica Verlag. All rights reserved.

media booking in into more value-added services. 254

255

253

Cp. Google, in: http://adwords.blogspot.com/2008/06/introducing-google-ad-planner.html/ 05.09.2008 Cp. http://blog.metaroll.de/2008/08/25/google-adplanner-das-ende-der-online-mediaplanung/ 05.09.2008 255 Cp. SinnerSchrader, in: http://www.mediabrief.de/2006/02/ mediaplanung_am_scheideweg_alle_macht_de.html / 05.09.2008 254

65

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

7. Summary and final remark Online advertising will expand in the next few years, although the spending compared with the media consumption is currently still low. Along with the increase of advertising spend the demand for efficient reach of target groups will be more important. BT can be an answer to that demand since it is able to address online consumers exactly according to their interest and attitudes. However, the online industry in general and the BT industry itself is described as being immature and the relations between the individual market players are complicated and therefore quite often very inefficient. Furthermore, the development of the online market is very diverse from an international point of view. Whereas the U.S. and some countries in Europe are highly progressed, the rest of the world needs fundamental developments to reach the same level and in order to provide the foundation for BT. BT fits into the general media planning process since it implies the same objectives such as reach, frequency, continuity, message weight and of course the definition of the right target group. It can furthermore comply with the requirement for a profound information basis since it delivers real time data concerning the target group, which then also allows optimization of the campaign parameters while it is running. This means the best media placing can be selected and afterwards evaluated regarding its performance. However, BT cannot contribute to the calculation of the budget; it is not a method to determine the most cost-efficient way of allocating the advertisement spending. Therefore it also does not support the establishment of a media strategy or the most efficient media mix respectively, since it is examined only regarding its abilities to find the right target group online. This is its major task within media planning and the way it is performed implies several advantages and disadvantages. The strength of BT is the realisation of the maximum reach within a target group while minimizing wastage, as well as an improvement of contact quality and advertising impact.

Copyright © 2009. Diplomica Verlag. All rights reserved.

BT is a learning system, which implies a great advantage regarding offline market research, since it is always up-to-date and provides a basis for prompt optimization in order to increase effectiveness. Frequency and continuity can be planned more accurately compared to offline media. Based on frequency capping BT is able to guarantee that a user sees an advertisement fitting his interests according to a schedule, which is also an important feature regarding brand impact. An increase of branding effects and user involvement is created by the relevant advertisement based on BT, which can imply a higher tolerance by the online 66

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

users since they are actually targeted with products and services they are really interested in. Most importantly, BT addresses the major objective of media planning: efficiency. Various case studies have already shown that behaviourally targeted advertising can drastically increase efficiency. A weakness is though that this statement has not been proven yet: There are no exhaustive studies supporting the claim of the industry that BT can actually increase efficiency. At present BT is not appropriate for mass advertising, since it has too limited reach among the entire online population. Firstly, the target criteria narrow down the broad reach and secondly, although one target group can be fully reached, BT is only possible across one network or ad exchange and currently those are too small in order to provide great reach. Furthermore the technology, which is as varied as the different BT provider in the market, is still under development implying quality issues, and lacking standards and transparency causing on the one hand higher costs and on the other confusion with marketers. This is supported by the fact that the relations within the online advertising industry are quite destructive, since knowledge gathered by BT about the target group is not used within the offline part of advertising. Overall, BT is seen to isolated from both the industry itself and the marketers, and therefore it cannot contribute to a mixed media strategy. Consequentially BT is threatened by the scepticism of the marketers resulting from their fear that behaviourally targeted advertising might actually harm their brands, since the technology is insensitive to the context in which it places the advertisements. The technology is dependent upon how many web sites are within the reach of a campaign, but some publishers feel that their web pages are commoditized by BT and might therefore be reluctant towards it. One major concern of advertisers and publishers are however the protection of the consumer’s privacy. Whether behaviourally targeted advertising might threaten privacy is Copyright © 2009. Diplomica Verlag. All rights reserved.

still up for discussion at several governmental authorities, and currently the most prevalent legislation is the EU Privacy Directive, which forms the basis for most national legislation in Continental Europe. Consumer themselves do not seem to feel bothered too much by BT, but as stated this might be owing to the fact that they are ill-informed of the actual behavioural tracking going on. Large sums are invested to develop both technology and reach, and furthermore standardization can be expected by the major players in the online world such as Google and Yahoo!. 67

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

One major opportunity is seen in the practice of combining the target methods as described in section 4, which implies the advantage of refining the target group since not all aspects can be covered by BT, and an elimination of concerns regarding the branding effect could be realised this way. New orientations of advertisers such as the personalisation of consumer communication or new technologies such as mobile devices, IPTV and online video will be key drivers for BT in the next few years. Those developments characterize the nature of the industry, quickly moving and highly innovative. In conjunction with the overall development of the industry BT will gain more importance and will be more influential. The technology might shift the focus to target discovery as currently executed by wunderloop by empirically evaluating the profile of users visiting the goal page and defining the exact target group in order to avoid imprecise hypotheses. This enables the advertiser to reach the target group in the most efficient way possible while being highly transparent and therefore creating greater trust in the capabilities of the technology. The advertising industry as a whole might undergo a great transformation in the next years. First of all this is absolutely essential in order to realize BT’s full potential by creating an active relationship between the offline and online interfaces in order to plan the target group and the media cross-channels in the most effective way. Furthermore Google, being the greatest online player, tries to extend its influence into the offline world. This might imply that in the future media planners have to drastically change their business concepts if they do not want to be overrun by technology, disintermediating them and saving their fees. So although BT still faces some obstacles and disadvantages for optimal media planning, it also offers a lot of opportunities, and with its development it will ameliorate the concerns of advertisers. Therefore media planners have to understand its potential and should track its development in the near future to find the best way to position themselves in the value chain, avoiding exclusion from the future execution of media planning. BT will grow in the next few years and can be recommended for media planning, since it is Copyright © 2009. Diplomica Verlag. All rights reserved.

already a highly efficient way to target the right consumers online. Its current limitations and obstacles always have to be considered when applying the technology to an advertising campaign, but a well thought-through procedure with clearly set objectives and a process, which allows optimization, are the foundation for successful media planning in conjunction with BT. In the future, however, it is very probable that the technology and the industry will develop to a stage where it overcomes its current restrictions and then provides an even more efficient form of analysis for media planning. 68

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

8. Bibliography 8.1. Books

1. Arens, William/Weigold, Michael F./Arens, Christian: Contemporary advertising, 11. Edition, Boston 2008, McGraw-Hill/Irwin

2. Barban, Arnold M./Cristol, Steven M./Kopec, Frank J.: Essentials of media planning, 2. edition, Chicago 1987, NTC Business books

3. Bogner, Thomas: Strategisches Online Marketing, Wiesbaden 2006, Deutscher Universitäts-Verlag

4. Bruhn, Manfred: Kommunikationspolitik – Systematischer Einsatz der Kommunikation für Unternehmen, 4. Edition, München 2007, Verlag Franz Vahlen

5. Bürlimann, Martin (2004): Web Promotion – Professionelle Werbung im Internet, 3rd Edition, Zürich 2004, Midas Management Verlag

6. Dannenberg, M./Wildschütz, F.H.: Erfolgreiche Online-Werbung, 2. edition, 2006 Göttingen, Business Village Verlag

7. Gärtner, Sandra: Die Währung für Online-Werbung, Köln 2007, EUL Verlag

8. Gay, Richard/Charlesworth, Alan/Esen, Rita: Online-Marketing – a customer-led approach, New York 2007, Oxford University Press Inc.

Copyright © 2009. Diplomica Verlag. All rights reserved.

9. Hamm, Ingo: Internet-Werbung – Von der strategischen Konzeption zum erfolgreichen Auftritt, Stuttgart 2000, Schäffer-Poeschel Verlag

10. Johannsen, Nicolai: Banner-@advertising – Klickraten als Möglichkeit einer Mediaselektion im Internet?, Wiesbaden 2002, Deutscher Universitätsverlag

11. Kollmann, Tobias: Online-Marketing, 2. Edition, 2006 Göttingen, Business Village Verlag 69

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

12. Koschnik, Wolfgang J.: Standard-Lexikon für Mediaplanung und Mediaforschung in Deutschland, Band 2, L-Z, 2. Edition, München 1995, Saur Verlag

13. Lammenett, Erwin: Praxiswissen Online-Marketing, Wiesbaden 2006, Gabler Verlag

14. Löffler, Jens: Media – Planung für Märkte, 7. Edition, Hamburg 2004, Axel Springer Verlag 15. Mühling, Johannes: Targeting – Zielgruppen exakt online erreichen, München 2007, Verlag reinhard Fischer

16. Mühling, Johannes: Targeting – Zielgruppen exakt online erreichen, München 2007, Verlag Reinhard Fischer

17. Pepels, Werner: Marktsegmentierung: Erfolgsnischen finden und besetzen, 2. Edition, 2007 Düsseldorf, Symposion Verlag

18. Percy, Larry/Elliott, Richard: Strategic Advertising Management, 2. Edition, New York 2005, Oxford Press

19. Reich, Sandra: Die persuasive Werbewirkung von Online-Medien – Eine empirische Studie zu Wirkung von Werbung, Aachen 2006, Shaker Verlag

20. Schmahl, Diana: Moderne Online-Marketing-Methoden, Saarbrücken 2007, VDM Verlag Dr. Müller

21. Schumann, David W./Thorson, Esther: Internet advertising – Theory and reserach, New Jersey 2007, Erlbaum Associates Inc.

Copyright © 2009. Diplomica Verlag. All rights reserved.

22. Schweiger, G/Schrattenecker, G.: Werbung, 6. Edition, Stuttgart 2005, Lucius & Lucius Verlag

23. Strauss, Judy/El-Ansary, Adel/Frost, Raymond: E-Marketing, 4. Edition, 2006 USA, Pearson Prentice Hall

24. Tellis, Gerard J./Ambler, Tim: The SAGE Handbook of Advertising, Los Angeles 2007, SAGE Publications 70

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

25. Unger, Fritz/Durante, Nadja/Gabrys, Enrico/Koch, Rüdiger/Wailersbacher, Rainer: Mediaplanung – Methodische Grundlagen und praktische Anwendungen, 5. Edition, Berlin Heidelberg 2007, Springer Verlag

26. Wolf, Volkhard: E-Marketing, 2007 München, Oldenbourg Wissenschaftsverlag

8.2. Articles in collected editions 1. Deyle, Hanno-G.: Mit neuen Segmentierungsverfahren zu besseren Zielgruppendefinitionen, in: Diller, H.(ed.): Zielgruppen finden und überzeugen, 2006, Wiss. Ges. für Innovatives Marketing

2. Fittkau, Susanne: Nutzer und Nutzung des Internets in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketing-Börse GmbH

3. Hegge, Ulrich: Targeted Advertising, in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketingBörse GmbH

4. Sander, Carsten/Schott, Alexander: Bannerwerbung – Formate, Erfahrungen, Trends, in: Schwarz, Torsten (ed.): Leitfaden Online Marketing – Das kompakte Wissen der Branche, Waglhäusel 2007, marketing-Börse GmbH

5. Unger, Fritz: Mediaplanung – Vorraussetzungen, Auswahlkriterien und Entscheidungslogik, in: Scholz, Christian (ed.): Handbuch Medienmanagement, Berlin Heidelberg 2006, Springer Verlag

Copyright © 2009. Diplomica Verlag. All rights reserved.

8.3. Journal articles 1. Alreck, Pamela L./Settle, Robert B.: Consumer reactions to online behavioural tracking and targeting, in: Database Marketing & Customer Strategy Management, 2007, Vol. 15, 1

2. Bannan, Karen J.: Behavioural targeting, in: B to B, Vol. 92, Issue 7, 2007

3. Blythe, Alex: Predict a click, in: Revolution, April 2008 71

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

4. Bryant, Melchior D.: Neue Trends im Behavioural Targeting, in: Absatzwirtschaft, Issue 4, April 2008

5. Croft, Martin: Behvioral Targeting – it could be the holy grail of online advertisement, in: Marketing Week: Interactive, 29.03.2007

6. Fösken, Sandra: Die Qualität der Zielgruppe ist ein wichtiger Aspekt, in: Absatzwirtschaft, Nr. 4, April 2007

7. Gengler, Barbara: Marktforscher erwarten starkes Wachstum für verhaltensbasierte Werbeformen im Internet, Computer Zeitung, Heft 33, 2007

8. Lemonnier, Jonathan: Contextual targeting boosts loyal following, Advertising Age, 4/14/2008 ad Network & Exchange Gui, Vol. 79

9. Matthewson, James A.: Behavioural Targeting: can online advertising deliver in 2006?, in: Journal of Direct, Data and Digital Marketing Practice, Vol. 7, No. 4, 2005

10. Micu, Anca C.: How did this ad get in my browser? A theoretical examination of online targeting and segmentation practices as they relate to media planning on the internet, American Academy of Advertising Conference Proceedings, 2005

11. Richard, J.I./Curran, C.M.: Oracles of „advertising“: Searching for a definition, Journal of Advertising, 31(2), 2002

12. Tsai, Jessica: Oh Behave!, CRM Magazine, Jan 2008, Vol. 12, Issue 1

Copyright © 2009. Diplomica Verlag. All rights reserved.

13. Eu-Gremium prüft Targeting-Methoden, in: Der Kontakter, Nr. 9, February 2008

8.4. Internet 1. Anfuso, Dawn: Where does BT go from here?, November 2006, in: http://www.imediaconnection.com/content/12466.asp / 05.09.2008

2. Barnett, Alex: http://blogs.msdn.com/alexbarn/archive/2005/10/08/478598.aspx/ 24.7.2008 72

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

3. BNet: http://findarticles.com/p/articles/mi_m0EIN/is_2007_May_31/ai_n27258672 / 30.08.2008

4. Bundesverband Digitale Wirtschaft (BVDW)e.V.: http://www.bvdw.org/presse/news.html?tx_ttnews[tt_news]=1367&tx_ttnews[backP id]=4&cHash=55c5427390/ 07.08.2008

5. Business Wire: http://www.businesswire.com/portal/site/google/index.jsp?ndmViewId=news_view& newsId=20060926005456&newsLang=en / 30.08.2008

6. Davis, Wendy: ValueClick Sues Tacoda Over Behavioral Targeting Patents, August 2008, http://www.mediapost.com/publications/?fa=Articles.san&s=87909&Nid=45708&p= 944842 / 21.08.2008

7. Denker, Helge: Targeting – Sein Verhalten macht ihn zum Ziel, March 2007, http://www.adzine.de/de/site/contentfcmsv445f073499f7dc/artikel.xml /04.06.2008

8. Denker, Helge: Werbung, die (ver)folgt), March 2006, in: http://www.adzine.de/de/site/contentfcmsv44419c077c4ed4/artikel.xml / 25.8.2008

9. Google, in: http://adwords.blogspot.com/2008/06/introducing-google-adplanner.html/ 05.09.2008

10. Gossmann, Bill: Behavioral Targeting – what we’ve learned so far, January 2008, in: http://www.imediaconnection.com/content/17905.asp / 25.8.2008

Copyright © 2009. Diplomica Verlag. All rights reserved.

11. Internet Usage Statistics; http://www.internetworldstats.com/stats.htm/ 24.7. 2008

12. Irwin, Tanya: ValueClick To Launch Predictive Behavioral Targeting, July 2008, http://www.mediapost.com/publications/?fa=Articles.san&s=86905&Nid=45235&p= 118697 / 22.8.2008

13. IVW, http://www.ivwonline.de/ausweisung2/search/ausweisung.php / 08.09.2008

73

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

14. JupiterResearch: http://findarticles.com/p/articles/mi_m0EIN/is_2006_June_14/ai_n26896408 /

15. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr= 4682/ 18.7.2008

16. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=1704 / 18.7.2008

17. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?stichwort =Optimierung&x=0&y=0/ 18.7.2008

18. Koschnik, Wolfgang J: FOCUS-Lexikon Werbeplanung, Mediaplanung, Marktforschung, Kommunikationsforschung, Mediaforschung, in: http://relaunch.medialine.de/PM1D/PM1DB/PM1DBF/pm1dbf_koop.htm?snr=6281 / 10.09.2008

19. Längin, Stefan: Mehr Effizienz durch Behavioural Targeting?, January 2006, in: http://www.adzine.de/de/site/contentfcmsv443d929d6473ee/artikel.xml / 24.08.08

20. Lawson, Cree: Why to choose a vertical ad network, in: http://www.imediaconnection.com/content/12998.asp / 07.09.2008

Copyright © 2009. Diplomica Verlag. All rights reserved.

21. Mason, Jeremy: The past, present and future of behavioural targeting, October 2007, in: http://www.imediaconnection.com/content/16891.asp / 30.08.2008

22. Mediensprache: http://www.mediensprache.net/de/werbesprache/internet/formen/banner.asp / 30.7.2008

74

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

23. Moran, Gwen: Natural Born Clickers, September 2008, in: http://www.entrepreneur.com/magazine/entrepreneur/2008/september/196280.htm l / 05.09.2008

24. Mr. Web: More scrutiny for online ad targeting, March 2008, in: http://www.mrweb.com/drno/printable/pn8116.htm /11.06.2008

25. Nuggad: http://www.nugg.ad/de/ 22.08.2008

26. Nugg.ad: http://www.nugg.ad/fileadmin/redakteure/downloads/Case_Studies/080505_Case_ Study_Ekstrabladet_de.pdf / 29.08.2008

27. Viralmythen: http://blog.metaroll.de/2008/08/25/google-adplanner-das-ende-der-onlinemediaplanung/ 05.09.2008

28. Schirnke, Marla: BT discipline for newbies, May 2008, http://www.imediaconnection.com/content/19242.asp / 23.08.2008

29. Shields, Mike: Study: Behavioural Targeting Succeeds on Web, February 2006, in: http://www.mediaweek.com/mw/news/interactive/article_display.jsp?vnu_content_i d=1002074845 / 24.08.2008

30. SinnerSchrader: http://www.mediabrief.de/2006/02/mediaplanung_am_scheideweg_alle_macht_de.

Copyright © 2009. Diplomica Verlag. All rights reserved.

html / 05.09.2008

31. Smith, Jack: Retargeting will feature in your future, January 2006, in: http://www.imediaconnection.com/content/7907.asp /22.08.2008

32. Welch, Sarah: 3 themes shaping behavioural targeting, 2008, in: http://www.imediaconnection.com/content/18549.asp / 24.08.2008

33. United Internet Media, in: http://united-internet-media.de/tgp.html/ 10.09.2008 75

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

34. Werbelexikon: http://www.werbe-lexikon.info/fachbegriffeaz/aidaformel.html / 22.08.2008

35. Wood, Jeff: Behavioral Targeting: 2004 in Review, http://www.imediaconnection.com/content/4858.asp / 07.08.2008

8.5. Miscellaneous 1. Berman, S.J./Battino, B./Shipnuck, L./Neus, A.: IBM Global Business Service – The end of advertising as we know it, 2007, IBM Global Business Services

2. EIAA Media Scope Europe: Media Consumption Study 2006

3. Euromonitor International, data exported 09/04/2008

4. Hallerman, David: eMarketer : Behavioural Targeting: Marketing Trends, New York 2008

5. Interactive Advertising Bureau Europe: Digital Europe – Tracking the growth of online marketing spend, Digital Insight Report, 2006

6. Interactive Media/ENIGMA/GfK: Market Report 2006: Erfolgsfaktor Targeting – Funktionsweise, Wirkung und Marktbedeutung

7. Jupiter Research: Behavioural Targeting & The Purchase Funnel Opportunity Consumer survey prepared for AOL + RevenueScience, April 2007

8. McRoberts, Brian/Terhanian, George H., Fleishman Hillard/Harris Interactive:

Copyright © 2009. Diplomica Verlag. All rights reserved.

Digital Influence Index Study, June 2008

9. Rasmussen, Andrea/Ude, Carolyn/ Landry, Edward: Interactive Advertising Bureau: HD marketing 2010 – Sharpening the conversation, New York 2007

10. Skou, Kasper, Senior Vice President Product and Business Development wunderloop, Interview 28.7.2008 / 02.9.2008

76

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

11. The 2007 Entertainment, Media & Advertising Market Research Handbook, USA 2007, Richard K. Miller & Associates

12. VanBoskirk, Shar/Forrester: US Interactive Marketing Forecast 2007 to 2012, Cambridge USA 2007, Forrester Research Incorporation

13. VanBoskirk, Shar/Forrester: The Reality of Behavioral Ad Targeting, Cambridge USA 2006, Forrester Research Incorporation

14. Yahoo!, Close the loop, 2007

8.6. List of figures

Figure 1: Based on Bruhn (2007): Classic marketing instruments (4Ps) in the marketingmix. Bruhn, Manfred: Kommunikationspolitik – Systematischer Einsatz der Kommunikation für Unternehmen, 4. Edition, München 2007, Verlag Franz Vahlen, p. 11

Figure 2: Based on Koschnik (1995): Criteria for media selection in the media planning process Koschnik, Wolfgang J.: Standard-Lexikon für Mediaplanung und Mediaforschung in Deutschland, Band 2, L-Z, 2. Edition, München 1995, Saur Verlag, p. 1236

Figure 3: % Time Spent Online vs. % Share of Online Adspend across Europe EIAA Media Scope Europe: Media Consumption Study 2006 (Hours – 2003, 2004, 2005 actual, 2006-2010 prediction, Adspend – 2003, 2004, 2005 actual and 2006-2010 prediction)

Copyright © 2009. Diplomica Verlag. All rights reserved.

Figure 4: Value-chain in online advertising (Based on Skou) Skou, Kasper, Senior Vice President Product and Business Development, wunderloop, Interview 28.7.2008

Figure 5: Classification of banner Dannenberg/Wildschütz (2006) Dannenberg, M./Wildschütz, F.H.: Erfolgreiche Online-Werbung, 2. edition, 2006 Göttingen, Business Village Verlag

77

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Figure 6: wunderloop’s connect marketplace Skou, Kasper, Senior Vice President Product and Business Development, wunderloop, Interview 02.09.2008

Figure 7: US behaviourally targeted online advertising spending 2006-2012 Hallerman, David: eMarketer : Behavioural Targeting: Marketing Trends, New York 2008

Figure 8: nugg.ad’s technology system Nugg.ad: http://www.nugg.ad/en/products/technology.html / 29.08.2008

Figure 9: Evaluation of the campaign results Nugg.ad: http://www.nugg.ad/fileadmin/redakteure/downloads/Case_Studies/ 080505_Case_Study_Ekstrabladet_de.pdf / 29.08.2008

Figure 10: wunderloop’s intelligent re-targeting or micro-targeting Skou, Kasper, Senior Vice President Product and Business Development, wunderloop,

Copyright © 2009. Diplomica Verlag. All rights reserved.

Interview 02.09.2008

78

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,

Unser gesamtes Verlagsprogramm finden Sie unter: www.diplomica-verlag.de

Copyright © 2009. Diplomica Verlag. All rights reserved.

Diplomica Verlag

Behavioural Targeting: An Online Analysis for Efficient Media Planning? : An Online Analysis for Efficient Media Planning?, Diplomica Verlag, 2009. ProQuest Ebook Central,