Newsletter · · Ashutosh Agarwal

AI Drives Ad Growth While Draining the Clicks That Fund the Web - Weekly Online Ads Podcast Recap - Week of August 16, 2026

Weekly online advertising podcast roundup for the week of August 9-16, 2026. AI was the biggest growth engine inside ad platforms and the biggest threat to the traffic they run on, while The Trade Desk, Reddit, Tencent and Google's new AI search ads drove the week's loudest debates.

Weekly Online Ads Podcast Recap

Week of August 16, 2026: AI Drives Ad Growth While Draining the Clicks That Fund the Web

What the podcasts were saying about online advertising this week, covering the last 7 days (Aug 9–16, 2026). It was a heavy week: Trade Desk's post-earnings unravelling, Reddit's earnings shock, Tencent's record ad quarter, Google's new AI search ads, and a bigger argument about whether AI is quietly draining traffic from the whole open web.

Executive Summary

  • AI is now the dominant force in online ads, and it cuts both ways. It is the biggest growth engine inside platforms (Tencent's ad revenue +22% YoY to 43.6B RMB purely from smarter AI targeting, not more ads) and the biggest threat to the traffic those platforms live on (AI Overviews cut source click-throughs ~60%; ~68% of Google searches now end without a click).

  • The Trade Desk (TTD) was the most-debated name. Revenue growth collapsed from 19% to 3% YoY; two straight misses after 32 beats; Q3 guided to negative growth. The bear case: Amazon's new DSP at 10-11% take rate vs TTD's 20%+, plus a Publicis fee dispute, threatens the premium. Stock near a 7-year low ($13, ~$6B cap), "buying opportunity" or "value trap."

  • Reddit (RDDT) delivered the earnings shock. Revenue +61% to $805M, net income +183% to $253M, strong Q3 guide, yet the stock fell ~10-20% on a US user wobble, heavy Google-search dependency, and management quietly dropping logged-in-user disclosure. Debate rages over whether to renew the "shockingly low" $60M/yr Google data-licensing deal.

  • Google is rebuilding search ads for AI, the ad becomes a conversation with a business-trained agent inside AI mode, called "the biggest shift in Google Ads" in eight years.

  • Retail media keeps maturing beyond Amazon/Walmart into marketplaces (Best Buy, Lowe's, Ulta) and AI chat: retailers were ~40% of early ChatGPT ad impressions, with Best Buy the top advertiser.

  • Ad fraud is a ~$300B problem out of a ~$750B industry ("the GDP of Denmark"), worst in open web, CTV, and audio.

  • Other threads: creators as marketing's "new operating system" (Omnicom/OMD); cookies not dead, so publishers build their own identity tools (ID5, UID2); Tencent (TCEHY) record ad quarter vs an AI-capex cash burn; Roku's (ROKU) free AI "slop" channel as a zero-cost content bet; and creative software (Adobe, Canva, Figma) wrestling with AI as feature-or-existential-threat.

(a) Dominant Themes

1. AI is now the single biggest force in online advertising, and it cuts both ways.

The through-line across nearly every podcast this week is that artificial intelligence is simultaneously the biggest growth engine inside ad platforms and the biggest threat to the traffic those platforms depend on.

On the growth side, the clearest example came from China. On Deep Values (Aug 14), the hosts broke down Tencent's Q2 2026 results and explained how AI juiced advertising without cramming in more ads. Tencent's marketing-services division (its ad business) "surged 22% year over year to 43.6 billion RMB" even though WeChat already has "1.439 billion monthly active users", a base the hosts called "practically the entire connected population of China." The trick: "They rolled out an upgraded AI-driven ad recommendation model and this automated campaign management solution called AM+… the number of ads you see on WeChat actually remained below the industry average. Instead, they made the ads exponentially smarter." Better targeting lifts click-through rates, which lets Tencent charge more per impression: "Tencent is literally absorbing marketing budgets away from e-commerce rivals simply because their AI model converts better." Gross margin hit 57.8%, up from 47.5% three years ago Deep Values.

On the threat side, the same AI wave is eroding the clicks that fund web publishers and search-dependent businesses. On The Investor's Podcast (Aug 16), the hosts noted that Google now has "a billion Gemini users and click-through rates for original sources plunged by roughly 60% because users can find exactly what they're looking for answered in the AI overview" rather than clicking through to the source website The Investor's Podcast. On Alex and Annie (Aug 12), marketers cited a Spark Toro statistic that "roughly 68% of Google searches end without a click," pushing brands toward "Answer Engine Optimization (AEO)" for AI Overviews Alex and Annie. And on The Programmatic Digest (Aug 12), Ad Butler's head of product Rob Janes described publishers facing "this existential crisis right now" as readers migrate "from search to AI tools," adding candidly, "I can't remember the last time I actually looked through the list of websites and went to a content website" The Programmatic Digest.

2. Google is rebuilding search ads for the AI era, the ad becomes a conversation.

The most concrete look at how Google plans to keep making money as search shifts to AI came from Whole Squad Eats (Aug 10). Host Phil Risher, who says his firm manages over $1 million in Google Ads and has spent eight years in home-services marketing, called it "the biggest shift in Google Ads that I've seen." The mechanic: a search ad in AI mode now shows a "talk to the business" option that "opens an AI agent that's trained on your Google business profile and website data," with a new "promote" tab that turns those conversations into a paid channel. His summary of the strategic stakes for Google: it solves "AI mode, trust signals, and monetization all at once." The takeaway for advertisers is that the old playbook is dead: "The old way is search ads, landing page, form… The new way is that your business profile is the decision point", so the advantage shifts from "a nicer website" to "how well AI understands your business" Whole Squad Eats.

3. Retail media keeps maturing, beyond Amazon and Walmart, into marketplaces and AI chat.

On the Retail Media Breakfast Club (Aug 10), a panel of four analysts (Kiri Masters, Andrew Lipsman, Sarah Marzano of eMarketer, and Debbie Aho Williamson) recorded at Cannes dug into where retail media networks (RMNs, the ad businesses retailers run against their own shopper data) go next. Key points they agreed on: where an RMN "sits in an organization dictates what it is allowed to be" (club retailers reinvest ad profits into member experience; others send them straight to the bottom line); the "60 plus RMNs" in the long tail beyond Amazon and Walmart need a different playbook, with Best Buy turning "their stores into media platforms"; and marketplaces are the real growth engine: "Marketplace where the money at… Best Buy, Lowe's, and Ulta. They all implemented marketplaces." A striking data point from Sensor Tower: in the first months of ads on ChatGPT, "almost 40% of ad impressions… were from retailers," and "Best Buy was the top advertiser in that time period", all product ads, no brand ads Retail Media Breakfast Club.

4. Ad fraud is a $300 billion problem hiding in plain sight.

Two podcasts this week put a hard number on how much ad money never reaches a real human. On Ecommerce Conversations (Aug 14), ad-fraud specialist Eric Beecham (of Beardbrand) estimated the global ad industry at "750 billion a year and growing" with roughly "$300 billion" lost to fraud: "if ad fraud was a country, it would have the GDP of Denmark." He explained that fraud is worst in the open web, connected TV, and audio, and detailed a scheme where a middleman spends half a client's budget on legitimate Netflix/Hulu/Disney+ inventory and pockets the other half by "spoofing" cheap web traffic as premium TV in the log-level data, a structure he compared to the mortgage tranches in The Big Short. His memorable frame: "the margins on fraud is near 100 percent" Ecommerce Conversations. On The Programmatic Digest (Aug 11), Dr. Augustine Fou (Fou Analytics) argued the industry's own euphemisms ("non-human traffic," "invalid traffic") "just cover up the problem. Just call it fraud." He warned against relying on AI for fraud detection ("there's no AI in Foo Analytics… it was built from the ground up to be deterministic") and cited client Georgia Pacific, which cut fraud and saw "increases in sales on their product pages on Amazon" partly by over-indexing on iPhone users because "bots don't like to pretend to be iPhones" The Programmatic Digest.

5. Creators are becoming "the new operating system" for marketing.

On Adspeak (Aug 11), Chrissie Hanson, CEO of OMD USA (part of Omnicom), described "the decentralization of creativity in real time," with brands handing creators "the microphones" to be "active partners in the production, not simply participants in a campaign." She stressed that creator work now demands the same rigor as retail media, citing a Starbucks Frappuccino campaign with Walmart run through three creators that drove "145 million impressions, 5 million products purchased, 7 points [increase] in ad recall, 8.9% sales lift, 5.5 in ROAS." She also flagged the "$700 billion global social commerce" opportunity and research showing "42% of purchases that come from an influencer inspired journey were completely spontaneous" Adspeak.

6. The cookie is (still) not dead, so publishers are building their own identity tools.

Rob Janes on The Programmatic Digest (Aug 12) reminded listeners of the industry's whiplash: for most of 2025 everyone braced for Google to kill third-party cookies and scrambled for "identity resolution," and then "near the end of 2025, they're like, oh, by the way, we're not going to do that now." The upshot: publishers built identity tools internally anyway, driving "massive growth for like ID5 and UID2." He also floated an emerging idea: publishers charging AI companies to scrape their content: "if an AI agent is scraping your site, then you need to be able to go back to Anthropic or OpenAI… you scraped my page… you need to pay me for that" The Programmatic Digest.

7. Programmatic buying is finally coming to podcast and audio ads.

On Right About Now (Aug 14), industry analyst Jeff Umbro (Podcast Perspectives) described a shift toward true programmatic advertising in podcasting: Simplecast pushing to AdSwiz, Spotify building programmatic alongside its private marketplace, iHeart developing "Omni" with Triton Digital, and YouTube enabling dynamic audio insertion tied to AdSense, letting advertisers buy across shows using listener demographic data rather than direct show-by-show deals Right About Now.

(b) Active Debates

1. The Trade Desk (TTD): temporary macro speed bump, or the start of structural decline?

This was the single most-debated ad name of the week, driven by The 7investing Podcast (Aug 14), which walked through Trade Desk's "ugly" Q2. Revenue growth collapsed "from 19% year over year… in the second quarter of 2025 to just 3% growth in the second quarter of 2026," net income margin fell from 13% to 9%, and adjusted EBITDA margin from 39% to 34%.

  • Bull side (CEO Jeff Green's view): the slowdown is macro, Trade Desk sells to "consumer discretionary spend, travel agencies, gyms, consumer products," which pull back when inflation and gas prices bite. Green calls it temporary, is hiring a new commercial C-suite (three senior client-facing execs in three days in July, many from Google), and the host sees a possible "buying opportunity" at ~$13/share (a seven-year low, ~$6B market cap), with a bull case of "$40 by the year 2027, 2028."

  • Bear side (the host's growing worry): Trade Desk missed estimates "for two quarters in a row" after "32 quarters in a row" of beats, and guided Q3 to negative year-over-year revenue growth for the first time. Two structural threats loom: a fee dispute with major agency client Publicis (which audited "the AI-related products" Trade Desk was charging for), and Amazon launching a competing demand-side platform (DSP, the software agencies use to buy ads) at "only… 10 or 11% of the take rate as opposed to the 20% plus" Trade Desk charges. The host's blunt conclusion: "I think personally the take rate is going to be coming down," and "the stock actually feels pretty appropriately priced right now" until growth returns The 7investing Podcast.

2. Reddit (RDDT): wildly under-monetized winner, or a company with a hidden user problem?

Reddit's Q2 print triggered a genuine two-sided debate across four podcasts.

  • Bull side: Revenue jumped "61% to $805 million," net income rose "183% to $253 million," and Q3 guidance of "$860 to $870 million" beat Wall Street, all covered on The AI Podcast (Aug 11) The AI Podcast. On The Joseph Carlson Show (Aug 12), the host framed Reddit as "an old company with young economics" (revenue up from "$93 million" in 2020 to "$500 million" trailing twelve months), powered by hyper-granular contextual ads ("if you go to a running subreddit, it'll advertise you running shoes"), and predicted a jump from a "$30 billion market cap" to "$50 or $60 billion" The Joseph Carlson Show. On The Investor's Podcast, one host noted management "hit a 30% operating margin" years ahead of his estimate and remains "quite bullish… it shows you how under-monetized this platform actually is."

  • Bear side: The same Investor's Podcast co-host delivered a "post earnings rant," arguing Reddit has a "user problem, especially in the US." US daily active users slipped (to 53.2 million from 53.5 million, per The AI Podcast), and, most damning, "more than half of daily Reddit users visit the website via Google search," making Reddit acutely exposed to AI Overviews cutting referral traffic. He accused management of choosing "to just hide it" and to stop reporting logged-in user metrics "right as this metric… might be turning against them." CEO Steve Huffman warned referral traffic had gone "choppy," and the stock fell ~10-20% after earnings. The debate crystallized in an analyst's on-call question (which the hosts replayed): why so confident about reaching "a hundred million logged in users in the US and 1 billion globally" when this quarter's daily count ticked down?

3. Should Reddit renew its $60 million data-licensing deal with Google?

A sub-debate with real money at stake. Multiple podcasts noted Google "pays Reddit $60 million a year to use their data." On The Investor's Podcast, one host called that "shockingly low… I think they should get multiples of that," given Reddit is "responsible for about five to 10% of the raw training data for LLMs and as much as 40% of real-time search citations." The counterpoint from the same conversation: leverage may already be gone: "even if Reddit says, I'm not licensing my data anymore, those LLMs [wouldn't] actually have a problem. I think we're just past that point." Reddit has signaled it may not renew when the deal comes up next year, precisely because "Google's AI products now compete with Reddit."

4. AI Overviews: is Google cannibalizing itself?

Beyond Reddit, hosts debated whether Google's own AI answers threaten Google's core business. As The Investor's Podcast put it, "Google is almost in a way siphoning traffic from websites and… maybe cannibalizing itself" because AI overviews "can perhaps not be monetized in the same way… [as] classical search traffic." The bull rebuttal running through the week: Google is the one player with both the AI and the ad machine to re-monetize the new interface (see the Whole Squad Eats AI search-ads mechanic above).

5. Is AI-referred retail traffic a brand-new measurement problem, or an old one in new clothes?

On the Retail Media Breakfast Club panel, Kiri Masters argued brands are "severely undercounting the traffic coming from AI" because it arrives "without appropriate referral tags," so it "looks like direct traffic." Andrew Lipsman pushed back: "It's not totally dark… you can reconcile multiple data sources." Sarah Marzano offered a third view: retailers "have been grappling with imprecise purchase journeys for a really long time" (word of mouth, in-store discovery), so AI is "an old [problem] that is simply wearing new clothes." The panel explicitly "didn't settle it."

6. Creative software (Adobe, Canva, Figma): is AI a feature they can absorb, or an existential threat?

On The Twenty Minute VC (Aug 13), the hosts debated the three big creative-software players after Canva CEO Melanie Perkins disclosed growth slowing "from 30%" to "20% by the end of this year," partly because frontier-model AI features "cost real money." The framing: Adobe ("$23 billion, growing at 12%"), Figma ("1.4 billion, growing at 40%"), and private Canva ("around 3.6 billion, growing at 20%"). The bear (Jason): the "prosumer market" is "the most exposed" because "everyone is ChatGPT fluent," and agents increasingly skip these tools entirely: "our agents never even suggest these products… It never once occurred to it [to use Canva]." The more hopeful read: Canva building its own cheaper in-house image model could relieve the margin pressure, and enterprise-anchored Figma is more defensible than prosumer Canva. Separately, on The Joseph Carlson Show, the host made a contrarian bull case for Adobe as a turnaround, noting the stock is down "60 to 70% from its all-time highs" ($660 to ~$255) while revenue "just grew 13% year over year", "one of the clearest cases of the fundamentals not matching the story" The Joseph Carlson Show.

7. Roku's AI "slop" channel: clever cost play, or embarrassing filler?

On Morning Brew Daily (Aug 13), hosts covered Roku launching "Fairground AI Creator TV," a free, ad-supported channel that "only plays AI generated content on a loop" (sample plotline: "chimpanzee smokes in zoo and gets arrested"). The bear read is aesthetic: reviewers called it "AI slop" and "categorically, catastrophically awful." The bull read is economic: it "doesn't cost anything to make," it feeds Roku's free ad-supported (FAST) strategy that "has paid off handsomely" (user base now "exceeds 100 million global households"), and it's a test: "if we can get viewership for this… we can pay people less." The hosts also noted Fox agreed to buy Roku in a "$22 billion" deal in June Morning Brew Daily.

8. Tencent (TCEHY): unassailable AI-ad compounder, or a cash-burning value trap?

On Deep Values, the debate was whether Tencent's record ad quarter is undermined by an AI spending spree. The paradox: a "monopolistic cash machine" (3.4 billion RMB non-IFRS net profit) that nonetheless posted "negative free cash flow of 13.8 billion RMB" because capex "skyrocketed by 176 to 190% year over year" to "roughly 52.8 billion RMB in a single three-month period" on GPUs and ASICs. Bull case: normalize for the hardware prepayments and free cash flow would have been "a highly positive 37.6 billion RMB"; the DCF scenarios ranged from a "$30.24 per share" no-growth floor to "$41.85" at 3% growth and higher at 12%. Bear case: three "permanent risks": demographic decline, Chinese state regulation, and "U.S. export controls on advanced GPUs" that could handicap its AI ambitions Deep Values.

(c) Tickers Mentioned

TTD, The Trade Desk

  • Bull: Revenue slowdown is macro-driven (consumer-discretionary advertisers pulling back), not structural; new commercial leadership + possible buying opportunity near a 7-year low with a path to ~$40 by 2027-2028.
  • Bear: Two straight quarterly misses after 32 consecutive beats; Q3 guided to negative YoY revenue growth; Publicis fee dispute + Amazon's cheaper DSP threaten the premium 20% take rate.
  • Source: "Trade Desk's Ugly Q2: What's Really Going Wrong | 7investing Live" (Aug 14). Speaker: 7investing host (with co-host Heather).
  • Quote: "We went from 19% year over year revenue growth… to just 3% growth in the second quarter of 2026… Amazon is now saying… we're only going to charge 10 or 11% of the take rate as opposed to the 20% plus you can pay me for the Trade Desk."
  • The 7investing Podcast

RDDT, Reddit

  • Bull: Q2 revenue +61% to $805M, net income +183% to $253M, Q3 guide $860-870M above consensus; deeply under-monetized with hyper-granular contextual ads and 30% operating margins achieved early.
  • Bear: US daily users slipped; over half of daily users arrive via Google search, so AI Overviews threaten referral traffic; management stopped disclosing logged-in user metrics; stock fell ~10-20% post-earnings.
  • Sources: "Crypto-Driven Revenue Boom for Reddit" (Aug 11); "9 Stocks That Could Make You Rich" (Aug 12); "TIP838: Google, Reddit, Amazon" (Aug 16).
  • Quote: "Their stock is still down 10% after hours because their CEO, Steve Huffman, warned that search referral traffic had become… choppy." The AI Podcast
  • Quote: "Reddit is responsible for about five to 10% of the raw training data for LLMs and as much as 40% of real-time search citations." The Investor's Podcast
  • Quote: "If you go to a running subreddit, it'll advertise you running shoes, Nike shoes… I believe Reddit is a company that we could look back in a couple years and see it at a $50 or $60 billion market cap." The Joseph Carlson Show

GOOGL / GOOG, Alphabet (Google)

  • Bull: Rolling out AI Search Ads that turn a search ad into a conversation with a business-trained agent, solving monetization for the AI-mode era; still the best-positioned to re-monetize the new interface; "a billion Gemini users."
  • Bear: AI Overviews cut source click-throughs by ~60% and may cannibalize Google's own high-margin search clicks; ad monetization of AI answers is unproven; the $60M Reddit data deal looks cheap and may not renew.
  • Sources: "Google Just Revealed AI Search Ads | Ep 143" (Aug 10); "TIP838: Google, Reddit, Amazon" (Aug 16).
  • Quote: "A search ad now shows talk to the business option… It opens an AI agent that's trained on your Google business profile and website data… This is the biggest shift in Google Ads that I've seen." Whole Squad Eats
  • Quote: "Click-through rates for original sources plunged by roughly 60% because users can find exactly what they're looking for answered in the AI overview." The Investor's Podcast

AMZN, Amazon

  • Bull (as ad platform): Launching an open DSP at roughly half Trade Desk's take rate (10-11% vs 20%+), leveraging massive first-party shopper data and owned inventory (Prime Video, e-commerce), a formidable position "right there next to people who are clicking on the buy button."
  • Bear (for competitors): Amazon's entry pressures independent ad-tech economics industry-wide.
  • Source: "Trade Desk's Ugly Q2" (Aug 14).
  • Quote: "The most terrifying thing you can possibly hear if you're the CEO of a smaller company is Amazon is now competing against us." The 7investing Podcast

TCEHY (0700.HK), Tencent

  • Bull: AI ad targeting (AM+) drove marketing-services revenue +22% YoY to 43.6B RMB with below-average ad load; 57.8% gross margin; fortress balance sheet and aggressive buybacks; DCF floor ~$30/share.
  • Bear: Free cash flow swung to -13.8B RMB on a 176-190% capex surge for AI hardware; structural risks from demographics, regulation, and US GPU export controls.
  • Source: "Tencent Record Profits Fuel AI Arms Race" (Aug 14). Speakers: Deep Values hosts.
  • Quote: "This division surged 22% year over year to 43.6 billion RMB… the number of ads you see on WeChat actually remained below the industry average. Instead, they made the ads exponentially smarter." Deep Values

ROKU, Roku

  • Bull: Free ad-supported (FAST) strategy has "paid off handsomely"; 100M+ global households; AI channel is a near-zero-cost content experiment; Fox agreed to buy Roku for $22B in June.
  • Bear: The AI "slop" channel drew scathing reviews and raises brand-quality questions.
  • Source: "Bob Iger, Josh Kushner Buys LA Lakers…" (Aug 13). Speakers: Neil Freiman & Kyle Hagee, Morning Brew.
  • Quote: "Roku now has plot lines like chimpanzee smokes in zoo and gets arrested… Their user base now exceeds 100 million global households. And Fox reached an agreement to buy the company in a deal worth $22 billion in June." Morning Brew Daily

META, Meta Platforms

  • Angle: Referenced mainly on the ad-quality/fraud side. On Ecommerce Conversations, the fraud specialist noted Meta "tends to be more responsive" on fraud than the open web, but flagged its "audience network" as risky and cited past episodes where Meta "had to wipe out like 10% of our users because they were bots." Broader Meta AI-ad tooling (Advantage+) surfaced only tangentially this week.
  • Source: "Ad Fraud Expert on Scams, Bots, Spoofs" (Aug 14). Speaker: Eric Beecham, Beardbrand.
  • Quote: "You also see like these announcements where meta is like, hey, we had to wipe out like 10% of our users because they were bots." Ecommerce Conversations

ADBE, Adobe

  • Bull: Consistent ~13% revenue growth despite an AI-disruption narrative; stock down 60-70% from highs ($660 to ~$255), "fundamentals not matching the story"; strategic pivot to acquire low-end AI users to monetize later.
  • Bear: New AI image/video tools (ChatGPT, Higgsfield, etc.) threaten core creative demand; part of the broader "AI as feature vs. existential threat" debate.
  • Sources: "9 Stocks That Could Make You Rich" (Aug 12); "20VC: Canva Slashes Growth…" (Aug 13).
  • Quote: "Adobe is down massively from $660 per share… now it's down to $255… Can you see the disruption in the revenue? I sure can't. It just grew 13% year over year." The Joseph Carlson Show

FIG, Figma

  • Angle: Cited as the fastest-growing of the three creative-software players ("1.4 billion, growing at 40%"), more enterprise-anchored and thus seen as more defensible than prosumer Canva, but it "traded down 20%" after warning that agentic-product usage would impair gross margins.
  • Source: "20VC: Canva Slashes Growth…" (Aug 13).
  • Quote: "Figma missed the quarter in a sense. Figma traded down 20%, but they burned the tokens." The Twenty Minute VC

Canva (private)

  • Angle: Growth slowing from 30% to ~20% as expensive frontier-model AI features compress margins; debated as the most AI-exposed of the creative tools given its prosumer base; building its own in-house image model to cut costs. (Private, no ticker.)
  • Source: "20VC: Canva Slashes Growth…" (Aug 13).
  • Quote: "Going into this year, they're growing at 30%. And the CEO, Melanie Perkins, disclosed… they're probably going to be growing 20% by the end of this year." The Twenty Minute VC

APP, AppLovin

  • Angle: On Mobile Dev Memo (Aug 12), discussed as a mobile-attribution powerhouse ("60-80% market share") in the context of digital advertising's M&A resurgence, referenced alongside a syndicate of Google, Meta, ByteDance, and Amazon. (AXON specifics were not the focus.)
  • Source: "Season 7, Episode 28: Digital Advertising's M&A Resurgence (with Maor Sadra)" (Aug 12).
  • Mobile Dev Memo Podcast

MGNI, Magnite (and the SSP landscape)

  • Angle: Named by Ad Butler's Rob Janes as an example of the supply-side platforms (SSPs, the tech publishers use to sell ad inventory) that an independent ad server lets publishers compare against each other ("how well is Magnite comparing against… cargo or anybody else").
  • Source: "205. Conversational Ad Ops, AI agents, and MCP with Rob Janes" (Aug 12).
  • The Programmatic Digest

OMC, Omnicom (via OMD) & the agency world

  • Angle: OMD USA CEO Chrissie Hanson framed creators as marketing's "new operating system" and demonstrated the ROAS rigor agencies now apply to creator campaigns (the Starbucks x Walmart example: 145M impressions, 8.9% sales lift, 5.5 ROAS). Publicis also surfaced as the agency in dispute with Trade Desk over AI-product fees.
  • Source: "Why Creators Are Becoming the New Marketing Operating System" (Aug 11).
  • Quote: "You worked with three creators and together you drove 145 million impressions, 5 million products purchased… 8.9% sales lift, 5.5 in ROAS." Adspeak

Retailer-advertisers referenced (Best Buy, Walmart, Lowe's, Ulta, Costco, Starbucks, Nike): Best Buy was named the "top advertiser" on ChatGPT and a leader in turning stores into media platforms; Best Buy, Lowe's, and Ulta all launched marketplaces; Walmart and Starbucks anchored the creator-commerce case study; Nike appeared as the archetypal contextual-ad example on Reddit. These were discussed as advertisers/RMN operators rather than as ad-stock pitches Retail Media Breakfast Club.