facebook's fraud lawsuit for inflating its claimed reach for advertisers just posted 95 files (3,646 pages!) - including a lot of the evidence dcn pushed to unseal. includes internal email threads behind prior claims. looks damning. will link to prior threads in a bit. /1
In the thousands of pages, we learn Publicis in France was especially concerned suggesting clients significantly reduce spend leading to their largest client, P&G, supposedly ceasing all spend. They estimated differences of 40-50% in claimed reach vs census in certain targets. /2
We also see Alex Schultz (now CMO) breaking down the source of inflated reach which follow-up docs show he accurately assessed as mostly due to #2 - lying about age and #3 - fake and duplicate accounts. Both listed as "millions" of the inflation... /3
a reminder, this issue surfaced from a press report, caused a flurry of emails by COO, CFO, CMO, etc (now unsealed) where decision was made not to inform investors. That's why a top Senator wrote letter last week and it's part of a SEC complaint, too. Will link in a minute. /4
Back to the docs, we learned there was special sensitivity around "SUMA" (single user multiple accounts) which Schultz indicates wasn't part of the press cycles and that was considered important. Later internal analysis suggested SUMA was 40% of the inflated #s. /5
We also learn "Ineligible Accounts" was considered a massive contributor to inflated reach. How much is still redacted so we're left with an engineer's note of 15-50%. Reminder, Facebook dismisses all of this as "potential reach" that doesn't matter as they don't bill on it. /6
But on that note, smoking gun, we have an internal document that very clearly states why Facebook's reach estimation matters. Even in the case of direct response advertisers buying on clicks, they say it matters. For brand advertisers, it's arguably the most important metrics. /7
there are hundreds of pages of internal debates by emails and slides and chat threads whether Facebook should stop claiming potential reach of "people" and move back to calling them "accounts" in order to avoid what has been said to be fraud (hence this lawsuit). /8
there is also considerable expert analysis of Facebook's inflated reach. One document is nearly 1,000 pages and this chart caught my attention. I imagine it will catch yours, too. Quarter 18 is when Cambridge Analytica broke wide open and Zuckerberg was first called to DC. /9
As someone who has recently experienced Facebook deceptively working to keep me from deleting my account and instead keeping it "inactive," the numbers of inactives being counted towards an inflated reach number is all sorts of suspect. Here is the table for the data. /10
the expert's 923 pages take a variety of data sources to model facebook's inflated reach by country and demographic building to the larger numbers. it's remarkable the SEC and/or independent auditors don't do this to all their data considering prior incidents with FB. /11
We also see again how Carolyn Everson played role to have a council of advertisers "ready on our behalf' with press. Everson is same super senior executive who recently left after other unsealed emails showed her waving a flag of concern over this issue. /12
on a side note, I would also love to see redactions like this considering Facebook claims it doesn't have underage users on FB and Instagram (because "they're not allowed") when this clearly shows they stick to what the user tells them "for a ton of reason"..ahem, legal. /13
press should dig into these new files - happy to help act as a guide. it's critical facebook have external accountability probing the integrity of its #s. here is a thread of prior information summarizing them. /14
and again, here is a direct link to prior unsealed emails including a senior executive matching point that there is no question inflated reach estimations impacted budgets including for small and medium-sized marketers. /15
Here is @SenWarren's letter to the SEC on this concern from just two weeks ago. I frankly don't why there haven't been hearings launched over it. I've long called for a longitudinal audit of Facebook's user accounts. warren.senate.gov/imo/media/doc/… /16
Justice takes time. What he knew when. AOC will remember this, “Their lawsuit says Zuckerberg—facing the risk of personal liability over the data privacy scandal—got himself out of trouble by agreeing to pay a larger-than-necessary $5 billion fine with shareholder money.” 1/3
Here is the full report from Bloomberg on Zuckerberg’s deposition which apparently was cut short and late on docs on Dec 3rd. Board members Thiel, Andreessen, others all being deposed these weeks. Press allowed Facebook to rewrite history on this. 2/3 news.bloomberglaw.com/business-and-p…
Here is a good thread that will get you into the details. Sheryl Sandberg also deposed although her assumed prior SEC deposition was sealed. We did finally get Zuckerberg’s which showed his nerves and that the scandal was on his mind much earlier. Thanks to @zamaan_qureshi 3/3
Overnight: FTC plans to call CEO Zuckerberg and (former) CTO Schroepfer in first 2 wks of trial (late April) in DC seeking to break up the company for abuse of monopoly laws. Also on their short list are some very big roles and names. Very likely driving behaviors at the top. /1
As I know their roles...
Chris Cox - chief product officer who took a break when scandals accelerated, and avoided testimony in UK.
Javier Olivan - now COO, key lieutenant
Sheryl Sandberg - former COO, on everything
Alex Schultz - key growth hacker, now CMO /2
Adam Mosseri - key lieutenant, now runs Insta
Dave Wehner - former CFO, deal approval including $19B with no real revenue WhatsApp
Fidji Simo - (former) always in mix, product leadership
Guy Rosen - Onavo alleged packet sniffing of WhatsApp, Snap /3
Woah. Google filed redacted versions of its Summary Judgment exhibits in Texas adtech antitrust case ahead of the March 31 trial. Although this case mirrors DOJ's case awaiting decision, it has even more eye-popping evidence. And an expert suggesting $29B in penalties. /1
"Project Bernanke" is an oldie but goodie that gets a lot of discussion. ICYMI, Google would allegedly increase the first highest and the second highest bid in its 2nd-price auctions then reinvest the "saved" funds back into other bids to Google wins more auctions. /2
The problem is while it may have ended up providing more revenue to the publisher (from/through Google) and no doubt to Google, Inc, it allegedly ignored the revenue which could have come through another channel if Google didn't manipulate the rev share as it ran the auctions. /3
US v Google II Closing arguments today.
70min for DOJ-> 95min for Google-> 20min for DOJ. Having predicted this case as better odds than search case (Google already lost in August), nothing changed my mind today. I wrote down: influence/$, complexity, deception, and arrogance. /1
I'm staying high level on perception first as findings of fact already covered much. On influence/$, DOJ pointed out early on and then again in rebuttal that every single witness presented by Google (except one) was paid or had grants from Google. /2
On complexity, Google again executed on its spaghetti defense with lead counsel, Karen Dunn, bookending trial by running over as she did in her opening. She appeared to skip dozens of slides, many minutes of close. And she loaded her slides while talking twice as fast as DOJ! /3
Bam. There is it. US Department of Justice has filed - requesting divestiture of Chrome as a remedy for court's finding against Google. Android at risk, too. /1
As it relates to Android, here is how DOJ puts it. Forced divestiture is option that "swiftly, efficiently, and decisively strikes at the locus of some anticompetitive conduct at issue here" but we're ok with tight behavioral remedies with option to divest if they don't work. /2
Revenue sharing for search default and/or preferential treatment - dead. Google's tens of billions to Apple - dead (last I saw it's about 15% of Apple's profits). Reminder, this all needs to be argued and approved by Court and then will get appealed. Still far off. /3
KA-BOOM. So when Google and its proxies (see so-called Chamber of Progress), friendly academics and analysts continue to suggest Chrome has nothing to do with the case, please ask them how many days they were at the trial. 1/3
This is super important. It’s an area @DCNorg (premium publishers) are intensely interested and concerned they get right around ability to restrict. The Court and trial made it clear they understood its importance during trial. We’ll be reading closely on Wednesday. 2/3
Here is the full report from Bloomberg who consistent with the entire trial showed up every day, did the hard work, and now got the massive scoop ahead of Wed filing. 3/3 bloomberg.com/news/articles/…