Retail Media Examples: What They Actually Look Like

The five things people mean when they say 'retail media example'

"Retail media example" almost always means one of five things, and most explainers roll them into a single blur. Picture a shopper searching "cast iron skillet" on Amazon and seeing a sponsored listing for HexClad above the organic results — that's onsite sponsored advertising, the oldest and biggest category, sold on a cost-per-click auction tied to the search term.

A step up from that is onsite display: a banner on the Amazon homepage, a video shelf unit on a category page, a placement on a brand's own storefront. Same retailer, same first-party data, but the shopper isn't searching — they're browsing, and the ad is trying to earn attention rather than close intent that already exists.

Then it leaves the retailer's own pages. Offsite display, video and audio — run through Amazon's DSP — follows a shopper who looked at a wallet on Amazon and later sees that brand's ad on a news site, a podcast app, or a connected TV show. In-store digital does the same job physically: a checkout screen or endcap display, built on loyalty card data instead of browsing data. Offsite retargeting is the fifth: a retailer's pixel or clean room shares browsing and cart data with an ad partner, who serves the ad on social or the open web. All five are "retail media." None is more real than the others — they just sit at different distances from the point of sale.

A worked example: what one of these campaigns turns into

Here's what that mix turns into when you follow it through an actual account, not a slide. Across 30 advertisers we manage on Amazon DSP in July 2026, the combined onsite and offsite media delivered 6.04x return on ad spend on 78.4 million impressions at a $4.00 CPM, with a blended $1.42 cost-per-click.

Impressions and clicks aren't the number that matters, though. Those campaigns converted to 57,137 attributed purchases at a blended $5.49 cost-per-acquisition — and 20.1% of those buyers were new to the brand. That last figure is the one worth sitting with. A campaign that drives cheap purchases from people who already buy the brand every month is optimizing for people who were never going to stop. A campaign pulling in new buyers at a similar CPA is doing something a sponsored-ads-only strategy usually can't.

The mistake that inflates every retail media example you'll read

The mistake almost every retail media writeup glosses over: last-click attribution can't tell you whether the ad caused the sale. If a shopper sees a display ad, then searches the brand name three days later and clicks a sponsored product ad, the sponsored ad gets the credit and the display spend looks like it did nothing — even if the display ad is the reason they searched at all. Run the sequence differently and display gets credit sponsored ads earned on their own.

We've made this mistake ourselves. Early DSP campaigns get judged on last-click ROAS because it's the number sitting right there in the dashboard, and it's easy to keep a campaign running because the ROAS looks fine when it's actually just riding on demand the brand already had. The only way to know is a holdout or matched control — hold a comparable audience out of the campaign, measure the difference — reconciled in Amazon Marketing Cloud, the one place DSP and sponsored ads stop double-counting each other's sales.

When the numbers on your own example look wrong

If your retail media numbers don't match what you expected, check these before assuming the media itself is broken:

• High ROAS, low new-to-brand share: the campaign is probably capturing demand that already existed. Not worthless, but it's retention spend, not growth spend — label it that way before you report it as acquisition.
• Cheap CPM, expensive CPA: the impressions are landing on the wrong audience. A low CPM paired with a high CPA usually means reach without relevance — check targeting before you touch the creative.
• Two channels claiming the same sale: if DSP and sponsored ads reporting both show the identical purchase as attributed, that's double-counting, not double performance. Dedupe in AMC before comparing either channel's number to a target.
• A holdout that shows no lift: that's not a failed test, that's the test working. Turn the spend off, rebuild the targeting, and re-test — don't keep paying for media a control group would have bought anyway.

Where this fits if you run Amazon ads

One category above — onsite and offsite Amazon DSP — is what Dr. DSP runs as a managed product. To be clear on the acronym: this is Amazon's Demand-Side Platform, the programmatic buying tool for display, video and audio, not the Delivery Service Partner courier franchise that shares the same three letters.

Dr. DSP is built by Full Circle, which has managed more than $500M in revenue across 100+ brands. There are 70+ brands live across the Full Circle and reMKTR group. There's no published price — a demo, the first 30 days free, and a rate set on the call against your actual media budget. Orbit, the reporting and guardrail software, is included at no extra cost either way.