AI
AI Buying Platforms Will Steer 27% of U.S. Spend
Madison & Wall sees automated campaign types taking 27% of U.S. ad spend by 2030 as Performance Max and Advantage+ decide where the dollars go.
Madison & Wall forecasts that AI buying platforms will steer 27% of U.S. ad spend by 2030, up from 12% in 2026 on a $479 billion market outside political ads. The firm, a media and advertising research shop, put that share at 2% in 2023.
That 27% is not a new channel. It is Google’s Performance Max, Meta’s Advantage+, and similar tools choosing the bid, the placement, and which query or feed item gets the dollar, which is why social, search, and commerce keep growing while TV and print stall.
Automated Campaigns Already Hold 12% of U.S. Spend
The September update from Madison & Wall puts global ad revenue at more than $1.3 trillion in 2026, an 11% rise after the industry first crossed $1 trillion in 2024. The firm called that pace unusually rapid. Global revenue grew 12.9% in the second quarter of 2026, and the same shop expects 9.8% in the third quarter once U.S. midterm political ads are stripped out.
The Interactive Advertising Bureau, working from more than 200 brand and agency buyers, separately raised its 2026 U.S. ad spend forecast to 12.3% growth, up 2.8 points from 9.5% in January. IAB chief executive David Cohen tied the revision to a strong first half around the Winter Olympics and the FIFA World Cup, plus tools that make it easier to find customers.
Those two 12-ish figures are not the same thing. IAB’s 12.3% is total U.S. spend growth. Madison & Wall’s 12% is the slice of a $479 billion U.S. market, excluding political ads, that already runs through automated or AI campaign types in 2026. By 2030 that slice is $158 billion, or 27% of the U.S. market.
THE AUTOMATED SHARE PATH
| Year | Share of U.S. ad spend | What Madison & Wall attached to it |
|---|---|---|
| 2023 | 2% | Baseline before the current tools were default |
| 2026 | 12% | On a $479 billion U.S. market, excluding political ads |
| 2030 | 27% | $158 billion of U.S. advertising |
Luke Stillman, managing director at Madison & Wall, did not dress that up as extra demand. He called it a transfer of who holds the budget.
This is a share shift.
Luke Stillman, managing director, Madison & Wall
From 2% in 2023 to 12% in 2026 is 10 points in three years. From 12% in 2026 to 27% in 2030 is another 15 points in four years, if the forecast holds. Reddit, Pinterest, and TikTok now sell their own automated campaign types, so the 27% is not a two-company private club, even if Google and Meta set the pace.
Performance Max and Advantage+ Pick the Placements
Performance Max is Google’s all-in-one campaign. The advertiser sets a goal, loads assets, and names a budget. Google’s system then builds the ad and decides where it runs. The company’s own product page describes a single campaign across every Google channel, spanning Search, YouTube, Gmail, Maps, Display, and Discovery.
Advantage+ is Meta’s parallel product, a suite that can automate audience, budget, placement, and creative combinations across Facebook and Instagram. Meta says advertisers who use the shopping version of that suite generate $4.52 in revenue for every dollar spent, 22% more than campaigns run the old way. Chief financial officer Susan Li told analysts in July 2026 that Advantage+ end-to-end solutions had passed a $75 billion annual revenue run rate, up from $60 billion in 2025, and that the company wants deeper adoption because using several of the tools together compounds the gains.
The media plan used to be a list of audiences, keywords, and placements a buyer could change. These products treat that list as a suggestion. Stillman said cheaper AI creative is also freeing cash that marketers then put back into media, which in practice means more money for the same automated pipes.
WHAT THE BUYER STILL SETS
- The goal: A conversion, a purchase, a store visit, or a lead the system is told to chase.
- The assets: Images, video, headlines, and a product feed the system can remix.
- The budget: A daily or campaign cap, not a placement-by-placement split.
- The signals: Customer lists and audience hints that steer the start, then fade as the model finds its own mix.
Bids, queries, landing pages, and which surface gets the next impression sit on the other side of that line. Google’s newer AI Max format also writes headlines and can pick the landing page, and Dynamic Search Ads are being retired into it. On Meta, Advantage+ is becoming the default for new campaigns, which means the targeting dropdown is no longer the job. The job is a clean purchase event the model can trust, because a fuzzy lead or view-content ping will send spend into junk.
Retail Search Budgets Are Already Inside the Box
The 12% national share still looks modest until you look at the accounts that live in these tools every day. Tinuiti, a performance agency that publishes client benchmarks, said Performance Max has accounted for 60% to 70% of ad spend among its retail clients since the fourth quarter of 2025. Buyer tallies circulating with the same forecast put about 30% of Google search spend in AI Max or Performance Max.
Scott Hendler, associate director of paid search at Ars X Machina, said Performance Max already takes well over 50% of his clients’ search budgets and that he can see the rest following. Becca Shih, a performance specialist at Roast, put her shop’s search mix through Performance Max at 11% to 12%, in line with the national figure, and said the format is becoming the default on the major platforms. John Dawson, vice president of strategy at Jellyfish, did not stop at Madison & Wall’s 27%.
We don’t think automation in media stops at 20% or 30%, we think it gets to 90%. AI is entering every part of the marketing lifecycle, and that will transform how media is planned, bought and optimized.
John Dawson, vice president of strategy, Jellyfish
Shih’s caution still stands. Give the system thin conversion data or sloppy creative, and it will spend with confidence in the wrong place, the same way a chatbot will answer a bad prompt. Danny Weisman, co-founder of indie shop Obsessed, said brands lean in because the platforms push the checkbox, and that some of his clients already run as much as 30% of their dollars through Advantage+. The agencies that still sell audience architecture as the product are selling a joystick the interfaces are putting away.
Social, Search and Commerce Keep the Growth
Stillman said social, search, and commerce are the channels with the scale and the consumer data, so they are the channels AI can actually price. Madison & Wall’s 2027 forecast has social up 17%, search up 12%, and commerce up 14%. For 2026 the firm has TV up 1.3%, direct mail up 0.3%, and audio flat at 0%, with print publishing the only grouping in the red. On a webcast briefing, Stillman said legacy offline channels such as television, audio, and publishing were essentially flat or declining even while the wider market ran hot.
IAB’s September channel cut, which splits TV the way buyers now buy it, shows the same tilt with different labels. Linear television is down 1.5%. Connected TV is up 15.6%. Paid search, a slower cousin of Madison & Wall’s broader search line, is up 8.1%.
IAB 2026 U.S. CHANNEL GROWTH
| Channel | September projection | Change from January |
|---|---|---|
| Social media | 16.5% | +1.9 points |
| Connected TV | 15.6% | +1.8 points |
| Commerce media | 13.6% | +1.5 points |
| Digital video excluding CTV | 9.4% | -0.2 points |
| Podcasts | 8.7% | +0.1 points |
| Paid search | 8.1% | -0.1 points |
| Digital out of home | 7.0% | -0.4 points |
| Linear TV | -1.5% | +0.2 points |
Madison & Wall also estimated that Alphabet, Meta, and Amazon will take 60% of ad revenue in North America, 59% in Europe, the Middle East and Africa, and 53% in China. North America is still the largest market at 36% of the global total, and it grew 17% in the first quarter. Latin America led major regions at 18% in that quarter, on a $61 billion base that Madison & Wall tied to Brazil and Mercado Libre, and the firm expects that growth to run into 2027. A smaller base makes a double-digit rate easier to print. It does not change where the dollars concentrate.
IAB buyers are chasing the same pipes. Customer acquisition jumped nine points to 63% as a stated media goal. Adapting to AI-driven search is now the top investment challenge at 44%, and 86% of buyers say they are already changing, or expect to change, how they measure media within 12 months because of conversational tools and agents. The measurement gap is not slowing the spend. It is following it into social, commerce, and the automated campaign types that sit on top of both.
AI Companies Bought a Quarter of Super Bowl LX
The other half of the loop is that AI firms are advertisers, not only vendors. iSpot, the television ad analytics firm, found that 15 of 66 commercials in Super Bowl LX, or 23%, either used AI in the making or sold an AI product. That tally is on the Super Bowl LX commercial tracker iSpot published around the February 2026 game.
OpenAI returned with a 60-second Codex spot, “You Can Just Build Things,” after a 2025 debut that chief marketing officer Kate Rouch put at about $14 million; she said the 2026 buy was roughly consistent with that. Anthropic ran Claude spots built around the line that ads are coming to AI, but not to Claude, a swipe at OpenAI’s plan to sell ads in ChatGPT. Google showed Gemini in a family-move ad. Meta pushed AI glasses. Amazon ran Alexa+. GenSpark used Matthew Broderick. Svedka aired a mostly AI-made vodka spot. Microsoft’s Copilot work also showed up in the mix.
Madison & Wall treated that night as a sample of a broader habit. AI companies, still flush with venture money, are spending to grow users at a harder clip than most other categories. The same firms sell the buying tools that then absorb a rising share of everybody else’s budgets. Creative gets cheaper, the leftover cash goes into media, and the media goes into Performance Max and Advantage+. The circular part is not a metaphor. It is the same dollar showing up on both sides of the ledger.
The Boom Competes With Governments for Capital
The September forecast’s warning sits one step past the 27% share. AI companies are now large enough borrowers and equity raisers that they compete with governments and other firms for investor capital. If that contest lifts borrowing costs, the same boom that is juicing ad revenue becomes a drag on the rest of the economy. The report put that in one sentence.
In other words, the AI spending boom is simultaneously supporting economic activity and increasing the financial pressure on the broader economy.
Madison & Wall, September 2026 advertising forecast
That is the second-order risk the 27% figure does not show on its own. If the tools keep taking share, Google, Meta, Amazon, and a short list of other platforms keep the growth even in a slower GDP tape. If the capital binge breaks, the same AI companies that bought Super Bowl time and stuffed automated campaigns with budget are the first advertisers that can slam the door. Traditional TV, print, and audio, already flat or down on Madison & Wall’s 2026 sheet, have no automated share shift to hide in.
Stillman listed the usual reasons a marketer might pause, from the war in the Middle East to interest rates, the job market, and an AI assessment bubble. Budgets have not blinked. “They keep closing their eyes and spending. We’ve seen no impact on budgets,” he said. The 27% forecast assumes they keep doing that, and that Performance Max and Advantage+ keep being the place the money lands when they do.
-
AI3 months agoFable 5 Came Back Under a Commerce On-Off Switch
-
AI4 months agoGoogle’s SpaceX GPU Lease Has a Sept. 30 Deadline
-
CRYPTO4 months agoPlasma One’s XPL Locks Face a 1.81 Billion Cliff
-
APPS4 months agoDGO’s Rs 549 World Cup Pass Cost Fans Sleep and Data
-
AI4 months agoMoonshot AI’s $30 Billion Ask Became a $35 Billion Close
-
NEWS4 months agoColorOS 17 Device List Spans Oppo, OnePlus and Realme
-
GAMING4 months agoXbox Cuts 3,200 Jobs After Five Years of Thin Returns
-
GAMING3 months agoThe RTX 4050 Under Rs 70,000 Hides a Wattage Gap
