FLEDGE: clarification on support for post-view conversion measurement

[RLemonnierScibids]

Post-view conversion measurement is central to adtech performance measurement, as last-touch models mixing post-view and post-click probably represent the majority of direct response campaigns.

This is as we see one of the most basic needs of advertisers regarding optimization: being able to report on conversions and get the raw material to optimize towards high-performing contexts (by using the reporting API to report on the performance of each context).

PV conversion measurement is a bit trickier to implement than PC conversion measurement in a privacy-preserving way, as it has been mentioned in several repositories/issues, and if I understand well this issue summarizes the current “state-of-the-art” regarding the topic. Even if it would take some testing to find the best one, there are already some ideas in it which would at least provide a first answer.

Our assessment is that continuing to support post-view measurement is mandatory if the objective of FLEDGE is to be widely adopted in an industry where post-click measurement is not the norm (and is not adapted to e.g acquisition campaigns whose objective is to generate a lot of small increments in probability to convert through multiple exposures).

Thus the question is more on FLEDGE general agenda: is it safe to assume that the removal of 3rd party cookies will not happen before a measurement solution for both post-view and post-click has been implemented?