Pace Notes
Monday, November 9, 2026The ad business, read one turn ahead.
Three items today. Affle's deal for Perion, read from the paperwork the trade press skipped, a retail analyst who says shoppers are fine with ads, and The Trade Desk on the morning it is expected to report. The Pick is the one to send to anyone whose retail media runs through Perion.
From ADOTATWire & File: Affle Calls Perion "Synergies." The Trade Press Skipped Discover-Tech.
Affle Pitches Outcomes. In 2021 It Agreed to Buy the Discover-Tech Name.Pick
The headline numbers first. Affle is paying $11.70 a share in cash for Perion, about $420 million, a 34.5% premium to the Oct. 30 close. Citibank is lending, the deal should close in early 2027, and Perion leaves Nasdaq and Tel Aviv. Perion pulled its guidance and cancelled its Nov. 16 earnings call. In March 2023 the market valued Perion at more than $1.8 billion. That is the part everyone wrote. It is the least interesting part.
Look at what the money actually buys. Perion spent about $374 million on acquisitions like Hivestack and Greenbids and on $196.8 million of its own stock. Take out the $228.3 million in cash on its balance sheet and Affle is paying roughly $192 million for the operating company, ADOTAT's arithmetic. That company posted $2.8 million in adjusted EBITDA in Q2, down 61%. Perion rebranded as AI-native. The market priced the rebrand below the cost of building it.
Now the paperwork. On Jan. 26, 2021, Affle told Indian exchanges its Dubai unit had agreed to buy the business assets of an Israeli company called Discover Tech, listing the platform, the brand name, the IP and the business relationships. In October, Citizen Lab tied Discover-Tech to Candiru spyware delivered through ads to at least 45 Al Jazeera devices. Citizen Lab says it has no reason to believe Affle controlled that operation and that the scope of the 2021 sale is unknown. So the question is simple: if Affle bought the name in 2021, how was a company still trading as Discover-Tech and still certified by Google into 2026? Affle did not answer Citizen Lab or India's The Wire.
Provenance matters because of who sits downstream. Perion's public listing descends from Conduit's toolbar business, the one security vendors flagged as unwanted software, and its later Bing dependence ended when Microsoft cut it off. Affle built on ads after every text message and app screens on a phone's first boot. And Perion runs ad buying for Target, Albertsons and Best Buy Canada. Asked about the price, Affle's North America chief told AdExchanger "a number is a mindset thing." Nobody asked about Discover-Tech. Perion's clients should ask in writing what Affle bought in 2021 and what it did with the name, before the deal closes.
Expanded from the Affle and Perion File in ADOTAT's Nov. 6 Wire & File, which draws on Perion's SEC filings and Q2 2026 results, Affle's Indian exchange filings and press releases from 2009 to 2026, June 30, 2026 shareholder data, Citizen Lab's Oct. 15 report as described by its researchers and by India's The Wire, and AdExchanger's deal coverage. The roughly $192 million net price for Perion's operating business is ADOTAT's arithmetic. The scope of Affle's 2021 asset purchase is not public, and nothing here alleges Affle took part in Discover-Tech's spyware activity. ADOTAT sent questions to Affle and Perion and said any answer would run as a dated update. The same issue's Fox and Roku items are not expanded here. ADOTAT discloses that Pesach Lattin has done contract work for Fox5 in Las Vegas and has known many ad tech executives for roughly 20 years, including as a competitor while running the ad network Vizi, and that sponsor Troutman Amin had no say over editorial. ADOTAT is reader-supported, with no paid placement in editorial.
From ADOTATEveryone Says Consumers Hate Ads. Ricardo Belmar Says the Spend Disagrees.
Ricardo Belmar Says Shoppers Accept Ads. He Clicks Past Amazon's Sponsored Ones.
Ricardo Belmar runs the Retail Razor Podcast Network and used to be a retail marketing director at Microsoft. On The ADOTAT Show he argued the thing nobody says on a panel: consumers are fine with ads. His evidence is the money. Ad budgets keep going up, and if nobody clicked, he says, advertisers would stop paying. Streaming is his second exhibit. It was born as the way to skip commercials, and then every streamer added a cheaper ad tier and signed up more people. That is not a revolt. That is people doing math.
On retail media he splits the difference. It is partly proof shoppers accept ads and partly retailers finally selling data they sat on for years. The episode puts the business near $50 billion. Belmar has Amazon and Walmart taking about 85%, and counts well over 200 other networks. Worth noting: the same week, ADOTAT's Wire cited Reuters putting the two giants at about 88% of an estimated $74.4 billion US market. Nobody counts this business the same way twice. His sharpest point is about stores, not screens. Associates field the same 40 easy questions out of every 100. Automate those, he says, and people get time for the 60 that turn into sales.
Then there is the anecdote that undoes the thesis. Belmar says he often sees a sponsored listing beside the organic one for the same product from the same seller, and he clicks the unpaid one. ADOTAT's paywalled fact check calls it what it is: avoidance with good manners. The seller paid for a click it would have earned free. In retail media that is paying for sales you would have gotten anyway. And rising spend proves advertisers believe ads work, not that consumers like them, especially when the seller of the ad also grades it.
He also made a call with a due date. Recorded in June, he predicted AI-assisted shopping could surge as early as this holiday season. ADOTAT says the announcements have arrived and the adoption has not been measured yet, and promises to check his math in January. His warning for brands holds either way: much of a product page sits in scripted code that chatbots cannot read. If the bot cannot see your product, it will not sell your product. Where he lands is the line worth keeping: consumers don't hate ads. They hate bad ads.
Expanded from ADOTAT's Nov. 9 piece on Pesach Lattin's ADOTAT Show interview with Ricardo Belmar, recorded in June 2026 and released in November, including its paywalled fact check. Quotes in the original are lightly edited from the recording. Belmar cited several figures from memory; ADOTAT checked them against public sources and rates them from holds up to plausible, depending on who counts. His account of OpenAI pulling instant checkout from ChatGPT was not independently confirmed by ADOTAT. The 88% and $74.4 billion figures come from ADOTAT's Nov. 6 Wire, citing Reuters. ADOTAT discloses that Belmar previously spoke with it as a source for its August 2026 reporting on Walmart Connect, that The ADOTAT Show is sponsored by Troutman Amin LLP and Incremental, which have no say over guests, questions or edits, and that ADOTAT uses AI tools, including Anthropic's Claude, in its production. ADOTAT is reader-supported, with no paid placement in editorial.
Elsewhere24/7 Wall St.: The Trade Desk Jumps 8% Ahead of Earnings; Magnite Gains 2%, AppLovin Dips
The Trade Desk Promises Greater Focus. It Cut 15% of Staff.
24/7 Wall St. reported this morning that The Trade Desk rose about 8% to $14 ahead of its third-quarter results, which the outlet expects today. Its read is blunt: management lowered the bar itself. In August the company guided Q3 revenue to at least $650 million and adjusted EBITDA to about $160 million. Q2 revenue had grown just 3% to $715.06 million, and it missed.
What followed the miss is the real story. A new CFO, CMO and chief commercial officer. A 15% cut to the workforce, with Jeff Green telling staff the company would run on smaller pods and scrums, "but with greater focus." And on Nov. 3, co-founder Dave Pickles, who left the CTO job in 2023, came back to lead a new CTO Council on AI workflows. The bull case the piece offers is retention above 95% and $269 million left on the buyback. The bear case has a name: Amazon, chasing the same budgets.
ADOTAT has been tracking this from the other side of the pipe. Its Nov. 6 Wire noted the guidance would leave Q3 about 12% below last year, which Adweek says would be the company's first year-over-year revenue decline as a public company, with the stock roughly 91% below its peak. The same week Magnite, the sell side The Trade Desk spent years telling buyers to route around, raised its guidance on 39% CTV growth. 24/7 Wall St. makes the same contrast, adding AppLovin's 55.5% revenue jump.
So today's report is not really about beating a number. Clearing a floor you set yourself is not growth. The question buyers should bring to the call is the one ADOTAT keeps asking: if the open internet's biggest buyer is shrinking while the pipes it disparaged are growing, who is the middleman now?
Read the original at 24/7 Wall St.
Stock moves, guidance, Q2 figures, the layoff, the leadership changes, the Pickles return and the Magnite and AppLovin figures come from David Moadel's Nov. 9 Market Movers piece at 24/7 Wall St., published at 8:46 a.m. ET; the article's headline and body say the stock rose 8% while its URL and summary bullet say 6%, and it notes the company had not confirmed the report date. Prices are intraday and will have moved. The article gives no analyst estimates, and this item was written before The Trade Desk reported. The 12% decline, first-decline and 91% figures come from ADOTAT's Nov. 6 Wire, citing Adweek. This is commentary, not investment advice.