PUFFIN, Personal User Floor For Impression Negotiations, is an enhancement for TURTLEDOVE and related proposals, intended to enable users to incentivize the production and distribution of desired ad-supported content.
New proposals for enabling interest group based advertising are predicted to have two key effects on the web advertising market.
-
Additional incremental ad revenue for publisher sites because of increased advertiser knowledge of audience composition, similar to the effect shown by Johnson et al. for the third-party cookie.
-
Leakage of ad revenue from high-engagement, high-reputation sites to lower-engagement sites where the same audience appears to be available.
While proposed post-cookie ad systems address the leakage of user data away from high-reputation sites, the revenue leakage is still a problem for web users who are members of the audience for that content and prefer to see more of it produced.
In the past, it has been difficult to take advantage of the immediate revenue-boosting effects of cross-site tracking without accepting the revenue leakage effects. However, moving the advertising market into the browser allows for adjustments to the rules of that market, in the interests of the user and of the ad-supported sites whose content they prefer.
The advertising literature generally focuses on the use of advertising by the seller of a good or service to produce desired behavior in the buyer. Today's web advertising ecosystem, however, is more complex. Advertising placement decisions are encoded in software run in the buyer's browser, which is the agent of its user. The rules of the market can therefore be designed to produce desired behavior in the seller.
The browser user seeks to obtain more value, in the form of ad-supported content of interest, for a given "cost" in advertising exposure. Browser-based ad auctions are an opportunity for users to choose to receive interest-group advertising selectively, in order to incentivize increasing investments in the ad-supported content they want.
In TURTLEDOVE and related proposals, a contextual ad and an interest-group ad participate in an auction. The interest-based ad wins (and is displayed to the user) if it can outbid the contextual ad.
A PUFFIN is a persistent per-browser floor price that the interest-group ad bid must also exceed in order to win the auction. Each PUFFIN is calculated based on an exponentially weighted rolling average of winning contextual ad bids. Losing bids and interest-group bids never affect PUFFIN.
The browser maintains one PUFFIN for each ad unit type that can be detected automatically. For example, a video ad and large leaderboard ad would each have a different PUFFIN from a small, below-the-fold ad.
With PUFFIN, in order for the interest-based ad to win the in-browser auction, it must beat not only the highest-bidding contextual ad, but also the PUFFIN for ads of the same category that have previously appeared in the same browser. If the category cannot be detected or is ambiguous, the interest-based ad must outbid the lowest applicable PUFFIN. This enables the positive, revenue-enhancing uses of interest-based advertising, but limits the revenue leakage effects.
Because PUFFIN is intended to incentivize production of high-engagement content, browsers may choose to adjust PUFFIN based on site engagement score, with a higher effective PUFFIN applied to low-engagement or previously unvisited pages.
One alternative considered would be for in-browser ad auctions to apply floating, periodically recalculated floor price to all impressions. However, applying the same floor to both contextual and interest group ads would deprive the user of valuable information about site reputation. Users can evaluate the trustworthiness of a site by observing which contextual ads the site is willing to run. PUFFIN preserves that information for the user by applying in-browser floors to interest group ads only.
PUFFIN provides for no additional reporting beyond what is available in the existing ad auction. PUFFIN is confidential and per-browser, and is not available to any script.
FIXME
FIXME
-
Johnson, Garrett and Shriver, Scott and Du, Shaoyin, Consumer Privacy Choice in Online Advertising: Who Opts Out and at What Cost to Industry? (June 19, 2019). Simon Business School Working Paper No. FR 17-19, Available at SSRN: https://ssrn.com/abstract=3020503 or http://dx.doi.org/10.2139/ssrn.3020503
-
Evan Davis, John Kay, Jonathan Star. Is advertising rational?