US AI-powered ad spend jumps 63% to $57 billion in 2026, Madison and Wall reports

US advertisers will spend $57 billion through AI-powered ad platforms in 2026, a 63 percent increase from the prior year, according to a new report from advertising analytics firm Madison and Wall. To put that in context, human-managed ad spend is growing at just 5 percent over the same period. The gap tells you something concrete about where the advertising industry is heading: not gradually toward machines, but already there for a large portion of the market.

The two products doing most of the work are Google's Performance Max and Meta's Advantage+. Both automate targeting, bidding, and budget allocation with minimal human input. Advertisers feed in creative assets and a conversion goal, and the systems decide where, when, and to whom the ads appear. That sounds straightforward, but it represents a fundamental change in what media buyers actually do day to day.

What Performance Max and Advantage+ actually do

Performance Max, launched by Google in 2021 and expanded significantly in 2022 and 2023, runs ads across all of Google's inventory simultaneously, including Search, YouTube, Gmail, Display, Maps, and Discover. The advertiser sets a budget and a conversion target, such as a purchase or a lead form submission, and the system allocates spending across channels in real time based on predicted conversion probability. There is no channel-level budget control and very limited placement reporting.

Meta's Advantage+ works similarly within Meta's ecosystem, placing ads across Facebook, Instagram, Messenger, and the Audience Network. Advantage+ Shopping Campaigns, the version most widely adopted by e-commerce advertisers, use Meta's AI to identify which users are most likely to purchase based on browsing behavior, purchase history, and engagement signals across Meta's platforms. Advertisers can set a minimum percentage of budget for retargeting existing customers, but the rest is fully automated.

Why advertisers are shifting budget at this pace

The 63 percent growth rate is not primarily driven by advertisers experimenting with AI out of curiosity. It is driven by return on ad spend comparisons. Madison and Wall's report found that advertisers using Performance Max and Advantage+ at scale reported average return on ad spend 22 to 35 percent higher than comparable manually managed campaigns across Q3 and Q4 2025. That kind of performance gap is hard for a media buyer to argue against when they are presenting quarterly results to a CMO.

The cost efficiency argument is also real. Running a manually managed Google Search campaign with tightly controlled keyword lists, bid adjustments by device and time of day, and separate ad groups for different audience segments requires meaningful human hours. Performance Max collapses much of that into a single campaign structure. For advertisers with large SKU catalogs or frequent promotional changes, the operational saving is significant regardless of the performance difference.

What advertisers give up when they automate

The trade-off is control and transparency. Performance Max does not report performance by channel in the standard interface. An advertiser cannot see whether their budget is predominantly going to YouTube versus Search versus Display, which makes it very difficult to understand what is actually working. Google added some asset group-level reporting in 2023 and search term insights in 2024, but the visibility remains substantially lower than manually structured campaigns.

Advantage+ has similar limitations. Advertisers have reported difficulty distinguishing how much of their reported conversions are incremental new customers versus existing customers who would have purchased anyway. Meta's attribution model gives the platform credit for conversions that occurred within a 7-day click or 1-day view window, which can overstate the actual impact of a specific campaign. Independent measurement using tools like Northbeam or Triple Whale often shows lower incrementality than Meta's native reporting.

The 5% growth rate for human-managed spend and what it means

Human-managed ad spend growing at only 5 percent is not a sign that traditional media buying is disappearing immediately. It is growing, just slowly. The categories where human management retains a stronger foothold are brand advertising on premium placements, political advertising where placement control is legally and reputationally important, and niche B2B campaigns where audience targeting precision matters more than scale.

Madison and Wall's analysis projects that AI-managed spend will account for 38 percent of total US digital ad spending by the end of 2026, up from 24 percent at the end of 2025. If that trajectory holds through 2027 and 2028, AI-managed campaigns will represent the majority of US digital ad dollars before the end of the decade, which has significant implications for what skills media agencies need and what Google and Meta can charge for their inventory.

How this affects media agencies and in-house teams

The shift toward automated campaigns is already changing agency headcount decisions. Several large holding company agencies, including WPP's GroupM and Publicis Media, reported in their 2025 annual reviews that the ratio of media buyers to managed spend has declined over the past two years as automated campaign types have taken a larger share of client budgets. Fewer people are needed to manage the same dollar volume when the optimization work is done by the platform.

For in-house marketing teams, the practical change is in where time gets spent. Campaign setup, keyword management, and bid adjustments consume less time. Creative testing, measurement architecture, and feed quality management consume more. An Advantage+ campaign is only as good as the product catalog data feeding into it and the creative assets the system has to work with. Teams that have invested in those inputs are seeing better results from automation than teams that fed the same campaign types with weak creative and incomplete data.

Google and Meta's financial interest in the shift

Both Google and Meta have a direct financial incentive to push advertisers toward automated campaign types. When the platform controls optimization decisions, it also controls where within its own inventory the budget gets allocated. That gives Google and Meta the ability to direct spend toward inventory that is harder to sell through manual placements, including lower-engagement display formats and off-peak time slots, while still reporting aggregate conversion numbers that satisfy performance targets.

Google's revenue per search query has declined as query volume has grown and AI Overviews have reduced click-through rates on traditional search results. Performance Max allows Google to offset some of that per-query revenue pressure by moving advertiser budget across its entire network rather than limiting it to search. Meta's Advantage+ similarly allows Meta to fill its full inventory more efficiently. The 63 percent spend growth number benefits both platforms disproportionately compared to what manually managed campaigns would have generated on the same budgets.

Madison and Wall's next quarterly update is scheduled for publication in May 2026 and will include revised full-year projections alongside data from Q1 actual results.