ad arbitration mechanism of DFP, it could result in better targeting of ads, thus better ad performance on the publishers' sites and consequently a higher attractiveness of these sites for advertisers.
DoubleClick currently does not use the data it has collected in the past to offer better targeting to new advertiser customers. This is because DoubleClick is contractually prohibited from using data created through the use of DFA except for limited purposes, none of which involves using data about user behaviour for the purpose of improving ad serving to publishers or advertisers other than the one advertiser on behalf of which the data was generated and collected.
Thus, DoubleClick can currently only use the data created through DFA to improve the service to the advertiser whose ads were served when the data was first recorded. For example, it can offer advertisers the ability to specify a sequence of ads that are to be shown to each user across all the web pages onto which the advertiser's ads are served. It is important to note that for this service to work, it makes no difference whether DFA has 190, 10 000 or just 1 advertiser customer. The technical working of the ad serving and the results for each advertiser would be the same in each case. Consequently, in this contractually restricted framework, the data created through the use of DFA does not bring about a network effect that could attract other advertisers to DoubleClick's ad serving tools. The same is true for publisher customers. For this scenario to work, it is not necessary that the web pages on which the advertiser's ads appear use DFP. They could use any competing publisher-side ad serving technology. As a result, data about user behaviour as it is created through the use of DFA does not bring about a network effect on the publisher side. As discussed in more detail below (cf. paragraphs 258 et seqq.), the merger is unlikely to change this to any significant extent.
Data created by the use of DFP
When the user requests the web page of a publisher using DFP, the publisher's web server calls DFP to add the ad tags to make sure ad space is filled. DFP executes its ad arbitration mechanism to decide from which advertiser, or from which ad network, an ad should be taken.
As a result of this way of operating, for each ad, DFP has a record about which advertiser or ad network has been selected to fill a specific ad space at a given time for a given web page of the publisher's whole website as well as the IP address of the user who had requested the web page. This record can sometimes be supplemented by information derived from what is known about the advertisers and the ad networks insofar as this information is provided by them, in particular the price they agreed to pay for the selected ad.
These records are then aggregated and presented to the publisher in order to
track the performance of his/her website. For example, the publisher could track the development of prices paid by different ad networks or, more plausibly, he/she could analyse which of the individual pages of his website created the most ad revenue. This type of analysis could be used as an input to uncover web page design flaws or to create new web pages that are optimised in order to bring in advertisement revenues.