IndustryPace
Ad industry commentary from Pesach Lattin. Since 2010.

Pace Notes

Friday, November 6, 2026

The ad business, read one turn ahead.

Three items today. A buyer's kit for the bot that got an ID, a profile of the strategist dentsu hired to dig out, and Criteo's quarter, read against both. The Pick is the one to forward to whoever signs your data contracts.

From ADOTATThe Field Kit: LiveIntent Sells "Authenticated" Reach. A Bot Got Authenticated.

LiveIntent Sells "Authenticated" Reach. Nobody Audits the ID Riding on Your Bid.Pick

Start with the finding. A peer-reviewed study at ACM's Internet Measurement Conference followed the HTTP Archive crawler, a bot that says it is a bot and has sat on the IAB Tech Lab Spiders and Bots list since 2013, through header bidding from 2022 to 2025. Seventeen of the eighteen ID vendors it met gave it an identity anyway, LiveIntent, ID5, Criteo, The Trade Desk and Experian's Audigent among them. Only Flashtalking said "Bot." LiveIntent tells Prebid its graph is authenticated daily through newsletter engagement. The crawler has never opened a newsletter.

ADOTAT's follow-up kit explains why a buyer should care, and it is about money, not manners. An ID is what turns a cheap impression into an expensive one. It lifts the bid, it puts the impression in your "authenticated audience," and it often drags a data fee onto your invoice. The kit traces the damage through five stops: the page, the auction, the targeting, the counting and the invoice. At the counting stop, vendors sent different IDs to competing exchanges on 18% of page views. One visit, two "people." Your reach looks bigger and your frequency cap leaks.

Here is the gap nobody owns. The Media Rating Council told Jack Neff it does not audit or accredit any identity provider. Your verification vendor grades the impression. Nobody grades the identity riding on it. Kean Graham of MonetizeMore put the incentive plainly: attaching an ID costs the vendor nothing and refusing costs them a bid. He sells bot filtering, so weigh it. He is still right.

The kit's answer is homework, not outrage. Get your ID roster by vendor from your DSP. Join your invalid traffic flags to the IDs that still carried a fee. Ask each vendor for 30 days of the requests it refused. Have a developer point an honest headless browser at a staging page and see who hands it an ID. Then put "fee follows validity" in the contract. As of Thursday morning no vendor had answered ADOTAT's questions. Selling identity is a business model. Selling identity you never checked is the problem.

The full story is at ADOTAT

Expanded from ADOTAT's Nov. 5 Field Kit, which draws on the peer-reviewed ACM IMC '26 paper by researchers at Indiana University, NJIT, Adalytics and DeepSee.io, ADOTAT's Nov. 4 report on it, vendor documentation, the Media Rating Council's statement to Jack Neff, and Kean Graham's written answers to ADOTAT on Nov. 3 and 4. The study covered HTTP Archive crawler traffic in header bidding to six SSPs from January 2022 to May 2025; the researchers could not decrypt the IDs, so nothing here says what any ID resolved to. Zeta Global (LiveIntent), Experian (Audigent), Mediaocean (Flashtalking), Criteo, Lotame and the paper's authors had not responded as of Thursday morning; ID5, The Trade Desk, Yahoo, PubMatic, OpenX and Magnite had not answered earlier ADOTAT inquiries. The checks and clauses are a buyer's tools, not legal advice. ADOTAT discloses that Graham's company sells the invalid traffic filtering he recommends, that two of the paper's authors work at Adalytics and DeepSee.io, and that the publisher has known executives at several companies named here for about two decades, including as a competitor. ADOTAT is reader-supported, with no paid placement in editorial.

From ADOTATChrissie Hanson Studied Japanese Before Her First Day at Dentsu. That Tells You How She Works.

Carat Says It Builds Brands. Dentsu Lost a Record ¥327.6 Billion.

Chrissie Hanson is dentsu's North America Media CEO and Global Brand President of Carat. Dentsu announced her on March 13, she started in August, and she came from running OMD USA, which Ad Age named 2025 US Media Agency of the Year. Her pitch for Carat fits in one line she gave Campaign: "We're in the business of building brands."

The context is the reason this hire matters. Dentsu posted a record ¥327.6 billion loss for 2025. That is the hole. Hanson spent four and a half months of garden leave with family, books and Japanese lessons. Dentsu is headquartered in Tokyo. Most American media bosses would have spent that time on the phone with clients. She spent it learning how her new owners talk.

The record says strategist. Carat London, Mediacom, Global Chief Strategy Officer at OMD Worldwide, a law degree from Durham, the IPA Excellence Diploma. In October Carat dropped "Designing for people" for "Difference Engineered", and planning and strategy moved to the top of the house with Bram Meuleman and Sarah Stringer. She also said something agencies rarely say: 5 to 15% of a typical plan is "dead reach," audience that was never going to matter.

Why it matters beyond one résumé: this has been the year of the fine print, with platforms shifting liability for ads nobody approved and buys sold as one thing and delivered as another. A media chief who reads contracts is worth more than one who reads decks. The open question is speed. Can a brand-first pitch win budgets while rivals sell automation and price? Watch for her first big account win and whether the waste she named shows up cut in client plans.

The full story is at ADOTAT

Expanded from ADOTAT's Nov. 5 profile, which draws on dentsu's and Carat's March 13, 2026 announcement of the hire, Hanson's interviews with Ad Age and Campaign (Oct. 14, 2026), and dentsu's reported 2025 results. Carat's 2026 Cannes award came before Hanson's start date, and Carat has not said who made the leadership appointments named here. Hanson was not interviewed for the original profile, and no one at dentsu or Carat reviewed it before publication. ADOTAT is reader-supported, with no paid placement in editorial.

ElsewhereInvesting.com: Earnings call transcript: Criteo falls as q3 results show growth pressure

Criteo Sells "Commerce Intelligence." Its Contribution ex-TAC Fell 15%.

Criteo reported Q3 on Nov. 4. Revenue was $416 million. Media spend through its platform rose 9% to $1.2 billion. And Contribution ex-TAC, the money Criteo keeps after paying for the media, fell 15% at constant currency to $240 million. More traffic through the pipe, less left in the pipe. Shares fell about 4% in premarket trading, per Investing.com.

CEO Michael Komasinski did not dress it up, calling it a challenging year. Retail media contribution fell 23%, with a $20 million hit from reduced scope at two retail media clients Criteo does not name; it says the underlying business grew 10%. Performance media contribution was $189 million, down 13%. The bright spot Criteo pointed to: CTV, AI platforms and social now account for 13% of Commerce Growth campaigns, up from 6%. Guidance held at a full-year contribution decline of 10% to 12%.

Now the ADOTAT angle. Criteo's pitch is "commerce intelligence," large-scale shopping data plus AI that turns intent into sales. On the day these results came out, ADOTAT reported that Criteo was among the seventeen ID vendors that gave a self-declared bot an identity in the ACM IMC study. Criteo had not answered ADOTAT's questions by Thursday morning. Intent data is only as good as the shopper behind it.

Nobody is saying the bot explains the quarter. It doesn't. But a company whose margin is shrinking while spend grows has every reason to keep every ID it can attach. Buyers moving money into its newer CTV and AI channels should ask Criteo the Field Kit question first: send me 30 days of what you refused to identify.

Read the original at Investing.com

Figures, guidance and the CEO's remarks come from Investing.com's Nov. 4 earnings call write-up, with the release date confirmed against Criteo's Nov. 4 release on PR Newswire; Contribution ex-TAC is Criteo's non-GAAP measure, and the premarket share move is as Investing.com reported it. The article was published slightly more than 48 hours before this edition. The bot detail comes from ADOTAT's Nov. 4 and Nov. 5 reports on the ACM IMC '26 paper; the researchers could not decrypt the IDs, and nothing here says the study bears on Criteo's results. This is commentary, not investment advice.