Retail Media Networks for DTC Brands: A 2026 Performance Playbook

Quick answer: Retail media networks (RMNs) are ad platforms owned by retailers like Amazon, Walmart, Target, and Kroger, letting brands buy ads across the retailer's site, app, in-store screens, and off-site inventory using its first-party purchase data. In 2026 they are the third force in digital ads behind Google and Meta, projected at USD 176B globally. DTC brands allocate 15 to 35 percent of paid budget.

Every DTC founder we talk to in 2026 is trying to figure out how much to spend on retail media and how to measure it. Across 22 accounts we ran in H1 2026, RMN spend produced 2.1x to 5.4x last-click ROAS with a measurable halo on organic retail sales.

By Nand Patel, Founder, New Public Media.

What Is a Retail Media Network in 2026?

A retail media network is an ad business built on a retailer's first-party shopper data, letting brands buy on-site sponsored placements, off-site programmatic, and in-store screens and audio. The category expanded from Amazon Advertising (launched 2012) into Walmart Connect, Target Roundel, Kroger Precision Marketing, Instacart Ads, Best Buy Ads, Home Depot Retail Media Plus, and dozens more.

The unlock is closed-loop measurement. Because the retailer owns both impression and purchase, brands see exactly which ads drove which sales without cookies or probabilistic modeling. That is why RMNs kept growing while everyone else lost signal.

Key takeaway: RMNs sell you access to a retailer's shoppers using the retailer's own purchase data, and closed-loop attribution is why the category exists.

Why Retail Media Networks Became the Third Force in Digital Ads

Retail media became the third pillar because retailers have three things advertisers ran out of: first-party purchase data, cheap traffic already paid for by the retail business, and closed-loop sales attribution. Per eMarketer's 2026 retail media forecast, global retail media ad spend will hit USD 176.9B in 2026, roughly 23 percent of all digital ad spend, up from 11 percent in 2020.

The margin math explains the push. A grocery chain runs on 2 to 4 percent EBIT. An ad business on the same shopper data runs at 60 to 80 percent margin. Every major retailer above USD 5B in revenue has stood up an ad network by 2026.

On our client accounts (2024 to 2026), Meta plus Google share of DTC paid budgets fell from 84 percent to 61 percent, and RMN share climbed from 6 percent to 27 percent.

Key takeaway: Retail media is the third pillar because it monetizes shopper data at software margins and now takes 20 to 30 percent of DTC ad wallets.

Amazon Ads vs Walmart Connect vs Target Roundel vs Kroger Precision Marketing

The four largest RMNs differ on ownership, buyer base, inventory depth, and minimums. Pick the one whose shopper actually buys your category, not the biggest logo.

Amazon still owns roughly 74 percent of US RMN ad spend in 2026 per Insider Intelligence. Walmart Connect grows 40 percent+ YoY, and its Vizio CTV integration made it the only RMN with a fully owned smart TV footprint. See Amazon Advertising, Walmart Connect, and Target Roundel for inventory sheets.

Key takeaway: Amazon is the default for scale, Walmart for mass CPG and CTV, Target for premium DTC, Kroger for grocery-adjacent categories.

How Much Should a DTC Brand Actually Spend on Retail Media?

DTC brands should spend 15 to 35 percent of paid budget on retail media once they cross USD 100K/mo in retailer revenue on the platform, split roughly 70 percent on-platform sponsored ads and 30 percent off-platform DSP. Below that floor, spend is directional and often loses money for lack of conversion signal.

Practical budget guidance by DTC revenue tier (annual ecommerce revenue):

On-platform vs off-platform rule of thumb: 70 percent on-platform (Sponsored Products, Sponsored Brands, on-site display) for direct conversion, 30 percent off-platform (DSP, CTV, off-site programmatic) for upper-funnel demand. For a deeper cut on channel mix at scale, see How Nine Figure DTC Brands Structure Their Paid Strategy.

Key takeaway: Spend 15 to 35 percent of paid budget on retail media above USD 100K/mo in retailer revenue, weighted 70/30 on-platform to off-platform.

The Ad Formats That Actually Move the Needle

Six formats generate almost all retail media performance in 2026: Sponsored Products, Sponsored Brands, Sponsored Display, DSP video and CTV, in-store audio and screens, and off-site programmatic.

Do not neglect creative. On CTV, in-store shopper footage plus a 6-second promo tag outperformed generic brand spots by 34 percent CTR in our 2026 tests.

Key takeaway: Sponsored Products carry 60 to 75 percent of RMN spend, but the leverage in 2026 is DSP CTV plus in-store screens where CPMs are still cheap.

Attribution: Closed-Loop Reporting, MMM, and the Halo Problem

Closed-loop retail media reporting overrates the platform because it only sees the last click inside a walled garden of high-intent shoppers. Amazon and Walmart both attribute the sale to their ad if the shopper clicked in the last 7 to 14 days, ignoring everything upstream on Meta, Google, or organic.

The fix is a three-layer stack:

1. Platform reporting (Amazon Attribution, Walmart Luminate) for tactical decisions inside the network.
2. Media mix modeling refreshed quarterly, with a separate coefficient per RMN. On our accounts, MMM typically discounts Amazon last-click ROAS by 25 to 45 percent once upstream Meta and search are properly credited.
3. Geo or holdout incrementality tests, run quarterly per network, to sanity check MMM.

For the deeper attribution stack post-cookies, see DTC Attribution Without Cookies and our companion piece on Amazon DSP for DTC Brands.

Key takeaway: Retailer closed-loop reports overrate the network by 25 to 45 percent, and only MMM plus incrementality tests correct the halo.

How Retail Media Fits Alongside Meta and Google

Retail media sits at the bottom of a flywheel that starts with paid social and ends with organic retail lift, not as a Meta or Google replacement. Meta and TikTok create demand, Google captures branded intent, Amazon and Walmart convert the last click, and the review flywheel feeds the next cohort.

The layered budget looks like this on our scaled DTC accounts:

The mistake is treating Amazon as demand creation. Brands that shift budget from Meta into Sponsored Products to hit an ACOS target see Amazon revenue plateau in 30 to 60 days, because branded search on Amazon caps at the size of the upstream demand pool.

Key takeaway: Retail media converts demand created upstream by Meta and Google, and starving upstream to feed Amazon caps growth within 60 days.

A 90-Day Retail Media Launch Plan for DTC Brands

A 90-day launch delivers a running Amazon program with clean tracking, a first Walmart test, and one incrementality readout.

1. Week 1 to 2: Audit retail sales by network, SKU, and margin. Set the USD 100K/mo revenue floor per network. Baseline organic sales for 4 weeks.
2. Week 3: Stand up Amazon Sponsored Products with 3 campaign types (branded defense, generic category, competitor conquest). Budget USD 8K to 15K/mo.
3. Week 4: Add Sponsored Brands on top 10 SKUs. Set 15 to 30 percent ACOS targets. Turn on Amazon Attribution for off-Amazon tracking.
4. Week 5 to 6: Layer Sponsored Display retargeting. Build first AMC audiences (24-month lookback). Book Walmart Connect kickoff if annual Walmart revenue exceeds USD 500K.
5. Week 7 to 8: Launch Walmart Sponsored Search on top 20 SKUs. Mirror Amazon structure. Budget USD 3K to 8K/mo.
6. Week 9: Start Amazon DSP self-service if retail revenue exceeds USD 100K/mo. Begin with 3 line items: retargeting, conquesting, category prospecting.
7. Week 10 to 11: Run first geo holdout (2 states off, 8 states on) for 14 days. Ingest results into MMM.
8. Week 12: Full readout. Publish the retail media dashboard. Set Q2 budget on incrementality-adjusted ROAS.

Want our team to run this on your accounts? Book a call or see our performance marketing services.

Key takeaway: A 90-day plan produces a running Amazon program, first Walmart test, and one incrementality readout before you scale.

Reporting Cadence, KPIs, and Team Structure

Retail media reporting should run weekly, monthly, and quarterly, owned by one person inside the brand who talks to both ecommerce and paid media leadership. Splitting ownership between the Amazon team and the paid team is the most common failure mode.

Team structure by brand size:

Core KPIs regardless of size: TACOS (target 8 to 15 percent), new-to-brand rate (target 25 to 40 percent), incrementality-adjusted ROAS (above blended paid benchmark), and share of voice on category head terms.

Key takeaway: One owner, three cadences, four KPIs (TACOS, NTB rate, incrementality-adjusted ROAS, share of voice) runs a healthy RMN program.

Frequently Asked Questions

What is a retail media network in simple terms?

A retail media network is an ad platform run by a retailer. Instead of buying ads on Facebook or Google, brands buy placements on retailer surfaces like Amazon search results, Walmart product pages, Target's app, or Kroger checkout screens. The retailer targets ads using its own shopper purchase data and reports back exactly which ads drove which sales, which is why the category grew so fast.

How much should a DTC brand spend on retail media in 2026?

DTC brands doing above USD 100K/mo in retailer revenue should allocate 15 to 35 percent of paid media budget to retail media, weighted 70 percent to on-platform sponsored ads (Sponsored Products, Sponsored Brands) and 30 percent to off-platform DSP and CTV. Below the USD 100K/mo revenue floor per network, spend is directional and rarely proves ROI. Above USD 25M ARR, retail media commonly reaches 30 percent+ of paid mix.

Do retail media networks replace Meta and Google ads?

No, RMNs capture demand that Meta and Google create upstream. Cutting Meta or Google to fund Amazon typically caps Amazon growth within 30 to 60 days because branded search on retailers is capped by upstream demand. The right structure is a flywheel: paid social creates demand, search captures intent, retail media converts, and reviews compound.