points by bbor 1 year ago

  "by engaging in a systematic campaign to seize control of the wide swath of high-tech tools used by publishers, advertisers, and brokers, to facilitate digital advertising."

IMO this is, by far, the most slam-dunk case against google, and the slow discovery of that fact is part of the reason I broke down while working there. This article is good, but just to make a complex issue a bit clearer:

There are people who sell Display Ad space ("Publishers") and people who buy ad space ("Advertisers"). Google AdSense is a well-known name, but that's actually their division for small blogs and such -- the big bucks come in from news sites, homepages, and other high-traffic publishers that use a different product altogether. Google isn't a monopoly on the publisher side (say, ads on your "Google Home" screen in the Google app), and obviously isn't a monopoly on the advertiser side; the issue comes about in the middle of the two, the "Ad Exchange" side. Google runs the show, and has spent roughly infinity dollars to keep it that way, even though this whole segment (Display Ads) is a pretty small slice of their overall revenue. The graphic published by the DoJ does the best job of explaining the overall issue, and the pieces involved: https://www.justice.gov/opa/pr/justice-department-sues-googl...

For example, they staved off Facebook's short-lived attempt to break into the exchange market (a meta-market, if you will!) by drowning the issue in money, then eventually doing the corporation equivalent of "I'll pay you $5 to fuck off": cartel formation. https://en.wikipedia.org/wiki/Jedi_Blue

  According to the draft lawsuit, Facebook agreed to reduce its participation in header bidding in return for "information, speed, and other advantages" that would come from staying with Google. Facebook allegedly would receive a guarantee of 90% of auctions regardless of the bids; 300 ms to bid (vs 160 offered to others), along with the ability to identify 80% of smartphone users and 60% of web users.

But IMO by far the most egregious behavior was that time they made a cool $230M off "project Bernanke", so-named because it was deemed a "bailout" of Google's struggling Advertiser clients by favoring them in the Exchange auctions. Somehow said with a straight face! The absolute fucking brazenness of stealing from others to please your clients and grow your pocket book while honestly seeing yourself as the good guy helping out little mom n' pop advertisers is breathtaking.

Article: https://adtechexplained.com/google-project-bernanke-explaine... Major points to anyone who can find the actual slides, they're made for middle managers so very digestible, and they have the hilariously illegal "ATTORNEY-CLIENT PRIVILEGE" warnings up top that Google is now known for. But somehow they're nowhere to be found as of today, even on Kagi.

If nothing else, once the dust settles and the redactions clear this will be a fascinating snapshot of how normal, kind people are made to do wildly asocial things with a smile on their face. I never once met a single person at Google who wasn't kind, intelligent, and well-spoken. And yet...

materielle 1 year ago

To add onto what you’re saying, this is surely part of Google’s strategy: obfuscation through complexity.

The anticompetitive schemes they engage in are so complicated and require so much background context that it’s basically impossible to explain to even subject matter experts, much less the casual reader.

  • bbor 1 year ago

    Well said, totally agree. While looking for the Project Bernanke slides I stumbled across this absolute classic internal guidance document, courtesy of the Texas DoJ: https://www.justice.gov/d9/2023-11/417885.pdf

      Don't define markets and estimate shares... For example, a field like advertising sweeps worldwide, and includes lots of different products and services. Defining a "market" more narrowly, or even defining one at all, can create issues if economists aren't involved. There's no problem with referring to a "market segment,'' "sector," "business/' etc., but just try and avoid defining the "market" for what we offer. Similarly, estimating "market share" is complicated, and it implies a defined market, so guessing or approximating isn't very reliable.
    

    Not to be trite, but it reminds me of 1984 more than a bit;

      If you want to keep a secret, you must also hide it from yourself.
mathteddybear 1 year ago

No, project Bernanke wasn't about that.

At that time, the exchange was a second-price auction, and all parties could submit up to two bids (presumably, the top two bids from their own collection of advertisers). Let's call the Google bids G1 > G2.

Since Google already implemented automated bidding strategies, they would submit to this auction (1+a)G1 and (1+b)G2 for certain fixed small value parameters a,b. Project Bernanke computed on historical data the optimal values of these a,b parameters.

Cue government discovery misunderstanding documentation

  • bbor 1 year ago

    Hmm, you clearly know what you’re talking about and this contradicts the available info, so thanks for sharing! I’m a little confused though. To bring it back to simple terms: you’re saying that the project was simply to start bidding more on ad space…? How exactly does that help (/“bailout”) their customers, and why would that be its own project? “Determining how much to bid for ad space” is already the job of half of Bayview campus, so I’m confused by this benign explanation.

    At the very least it sounds like they were using their position as auctioneer to fine-tune their bidding strategies, which seems like a textbook example of monopolistic behavior. But even that would be a step up from what I/the article above accuse them of.

    • mathteddybear 1 year ago

      The values of G1 and G2 are computed by a complex algorithm, however, that algorithm is agnostic of the position of the ad in the auction. Unlike the constant factors (1+a) and (1+b) applied on top of that.

      Other companies in that auction could apply this kind of optimization, too. Perhaps the improvement is not as large for smaller participants, and so, not worth looking into.

frays 1 year ago

Great comment. Will be interesting to see what Google looks like in 5 years from now.