- US retail media reaches about $71 billion in 2026, up ~18%, faster than search or social. [1]
- Amazon 79.7%, Walmart 8.0%, Target 1.5%. The top two take 89% of new 2026 dollars. [1][3]
- Costs aren't the story: Amazon Sponsored Products clicks +23%, cost per click (CPC) down 1% in Q4 2025. [4]
- The money is reallocated from TV, programmatic, and display, not found new. [2]
- Plan Amazon and Walmart as core performance lines. Treat the rest as a test budget.
There is a version of the retail media story that is all upside: a channel racing past $70 billion, growing faster than search and social, with the shelf itself turned into ad inventory. It's true. It's also the most concentrated channel in digital advertising, and the concentration is the part that should shape your plan.
That headline is the easy part of the story. The number that should shape your plan is who actually captures it, because retail media is the most concentrated channel in digital, and the split is not close.
Amazon and Walmart held a combined 87.7% of US retail media spend in 2025, and eMarketer expects them to capture 89% of the incremental dollars in 2026, $9.42 billion of $10.53 billion in net-new spend. [3] The duopoly is already on the income statements, not just in the forecasts. Amazon reported $21.3 billion of advertising services revenue in the fourth quarter of 2025 alone, up 23% year over year (a worldwide figure, not just US retail media). [5] Walmart's global advertising business closed its fiscal 2026 at nearly $6.4 billion, up 46%, with Walmart Connect in the US growing 41% in the holiday quarter. [6] The blunt reading: for a brand with real retail distribution, the planning question is how much to put into Amazon and Walmart, not whether. Everything else (Target, DoorDash and Instacart, each near $1 billion in US ad revenue, plus the long tail of category networks) is coverage and experimentation, not scale. [1]
The cost story is the opposite of what you'd expect
Rapid growth usually means price inflation. Retail media, so far, is the exception.
Clicks and sales up more than 20% on Amazon Sponsored Products, CPCs down 1%, demand-side platform (DSP) spend up 31%: that is an efficiency-expansion phase, not a cost crunch. [4] Sponsored products alone were a roughly $38 billion US line in 2025, the workhorse the whole channel runs on. [3] Inventory is growing faster than demand can bid it up. The implication for 2026 is unambiguous: the limiting factor is how much profitable sponsored-search coverage you can actually claim, not how expensive the clicks have become. Maximize that before you treat retail-media display and video as scaled performance. For brands that also sell on these platforms, that ad line stacks on top of the marketplace fee card, which is why it belongs in cost of goods, not marketing.
The budget isn't new. It's being taken from somewhere.
The uncomfortable half of the story: retail media growth is largely funded by cannibalizing other lines.
Share of marketers reallocating into retail media from each channel: linear TV 52%, traditional advertising 48%, programmatic 45%, display 45%. [2] Search and social are increasingly pooled in rather than raided outright.
That is why the framing is shifting from "retail media" to "commerce media," a single budget category absorbing retail, affiliate, connected TV (CTV), and social commerce. [3] Forty percent of advertisers already see retail media as a full-funnel channel, just behind social at 52%. [1]
Stop running retail media as an incremental test line.
The first wave of money came out of TV, open-web programmatic, and display. The second wave is lower-funnel search and social budget moving into retailer-owned audiences, where the measurement ties straight to sales. If you keep these siloed while competitors optimize at the commerce-media level, you lose the efficiency they're compounding.
What to do with this in 2026
- Make Amazon and Walmart core performance lines with real targets, not a 5% experimental carve-out.
- Max profitable sponsored-search coverage first. Scale display and video only once search is saturated.
- Fund it honestly: reallocate from linear TV, open-web programmatic, and generic display, where the spend is already leaking.
- Plan search, social, and retail as one commerce-media pool, optimized to sales, not three defended silos.
- Use the long tail (Target, Instacart, DoorDash, category networks) for coverage and joint business plans, not scale.
Retail media isn't a land grab you can still win. The land is mostly claimed. The advantage left is operational: buy the two networks that matter with discipline, fund them from the lines they're already draining, and measure the whole commerce stack as one system while your competitors still run it as three.
Sources
- eMarketer · Retail media networks: who is winning and how marketers should allocate budgets in 2026
- Encodify · The state of retail media2024 marketer survey on budget sourcing
- Osmos · Retail media networks 2026: build, buy & benchmark guide
- Tinuiti · Retail media trends and outlook for 2026 (Digital Ads Benchmark Report)
- Amazon · Amazon.com announces fourth quarter resultsQ4 2025 earnings, February 2026; advertising services revenue is a worldwide line
- Marketing Dive · Walmart raked in $6.4B from ads last yearcoverage of Walmart's Q4 fiscal 2026 results, February 2026
Frequently asked questions
How big is US retail media in 2026?
eMarketer forecasts roughly $71 billion in US retail media ad spend for 2026: $71.09 billion on its December 2025 forecast (+17.8% year over year from $60.32 billion in 2025), and $71.98 billion (+18.7%) on a later update. Either way, plan for about $70 billion to $72 billion growing near 18%, faster than both search and social ad spending.
Who controls retail media, and is it really just Amazon and Walmart?
For scaled buys, effectively yes. Amazon Ads held 79.7% of US retail media spend in 2025, Walmart Connect 8.0% and Target Roundel 1.5%. Combined, Amazon and Walmart are on pace to absorb 89% of all net-new US dollars in 2026, about $9.42 billion of $10.53 billion in incremental spend. Everything else is category coverage and experimentation, not scale.
Are retail media CPCs rising?
Not on the leaders, at least not yet. Tinuiti's Q4 2025 benchmark showed Amazon Sponsored Products clicks up 23% and sales up 22% year over year while CPCs fell 1%. Amazon DSP spend rose 31% with CPMs (cost per thousand impressions) also down 1%. The constraint in 2026 is inventory and retail distribution, not cost inflation. So the play is to max out profitable sponsored-search coverage before scaling display and video.
Where does retail media budget actually come from?
Mostly from legacy channels, not new money. Encodify's survey found retail media budgets most commonly reallocated from linear TV (52% of respondents), traditional advertising (48%), programmatic (45%), and display (45%). Search and social are increasingly folded into the same 'commerce media' pool rather than defended as separate silos.
Should smaller retailers' networks be part of the plan?
As targeted tests and category coverage, yes. As a scale channel, no. DoorDash and Instacart each generate close to $1 billion in US ad revenue and can matter for specific categories, but eMarketer expects the share of essentially every other network to stay flat or decline through 2027. Treat the long tail as joint-business-plan and experimentation budget.
What is commerce media, and how is it different from retail media?
Commerce media is the wider budget category that retail media is folding into, a single pool that also absorbs affiliate, CTV, and social commerce. The shift matters because the second wave of money moving in is lower-funnel search and social budget heading into retailer-owned audiences where measurement ties straight to sales. Brands that keep these siloed lose the efficiency competitors compound by optimizing at the commerce-media level.
Should I put more budget into sponsored search or into display and video?
Max out profitable sponsored-search coverage first, then scale display and video only once search is saturated. Sponsored products were roughly a $38 billion US line in 2025 and are the workhorse the whole channel runs on, and the current constraint is inventory and distribution, not cost. Amazon Sponsored Products clicks rose 23% while CPCs fell 1% in Q4 2025, so the limiting factor is how much profitable search coverage you can claim.
Is retail media only a lower-funnel channel?
Not anymore for many advertisers. About 40% already see retail media as a full-funnel channel, just behind social at 52%, which is part of why the framing is shifting toward commerce media. Even so, the disciplined 2026 play is to make Amazon and Walmart core performance lines with real targets rather than a small experimental carve-out.
Mikołaj Salecki
Writes about media, tech, and AI business for people who actually run digital. Former agency lead. Skeptic of frameworks that read better than they perform.
Get the next issue by email.
One letter, once a week. Sharp coverage of media, tech, and AI business. No filler.
