Amazon Advertising Examples: What Each Format Actually Does

The four formats, and what each one is actually for

Amazon advertising examples usually mean one of four formats, each built for a different job. Sponsored Products is a single-product ad that slots into search results and product pages, priced on cost-per-click. Sponsored Brands puts a headline, your logo, and up to three products in a banner above search results — it exists to hold shelf space against private label and multipacks. Sponsored Display retargets shoppers who viewed your listing but didn't buy, and can run off Amazon too. Amazon DSP is the programmatic layer: display, streaming TV (Prime Video, Twitch), and audio, bought against Amazon's shopping and streaming signals rather than a keyword.

These aren't interchangeable, and most pages that show Amazon advertising examples quote one CPC or one ROAS as if it applied to the whole account. It doesn't. A search ad chasing an existing want and a streaming TV spot introducing a shopper to a category for the first time are different products, bought at different prices, for different reasons.

A worked example: what a full-funnel campaign actually produces

Here's what a full-funnel campaign looks like when you don't cherry-pick the best line item. Across 30 advertisers we managed in July 2026, the portfolio delivered 6.04x return on ad spend, measured across the whole book, not the best-performing product. That ran on 78.4 million impressions at a $4.00 CPM, with a blended $1.42 cost-per-click. Blended cost per acquisition landed at $5.49 across 57,137 attributed purchases, and 20.1% of those purchases came from a shopper new to the brand.

Two things in that number set matter more than the headline ROAS. First, the $1.42 blended CPC is not the number most listicles quote — they usually cite $0.41, which is an online-video-only figure, not a whole-book average. If you're benchmarking your own account against a single quoted CPC, check which format it describes before deciding you're overpaying. Second, a 6.04x ROAS on a mixed book — some formats charge for attention with no click at all, others only charge on conversion — is a more honest number than a 6.04x on Sponsored Products alone, because the awareness spend is doing work a last-click report will never credit.

Examples by brand category

The right example depends on the funnel stage a brand is filling, not which format is trendy. A cookware brand like HexClad, competing on crowded generic search terms, leans on Sponsored Products to defend branded keywords and Sponsored Brands video to differentiate on a results page full of look-alike pans. A grooming brand like Beardbrand or a footwear brand like Walkize, selling into categories where the shopper often doesn't know the brand yet, gets more from Sponsored Display and DSP to build awareness before search intent ever forms. Ridge, selling a considered-purchase item like a wallet, fits a retargeting-heavy Sponsored Display setup, since shoppers tend to browse before they buy.

A hobby brand like The Woobles or Epic Gardening, whose audience finds them through interest rather than a specific product search, does more with streaming TV and off-Amazon display to introduce the brand, then narrows to search once demand exists. BK Beauty sits in between: enough branded search volume to run Sponsored Products hard, but a visual product that benefits from Sponsored Brands creative. None of this is a guarantee that a given brand's mix will land these results — it describes what each format is built for, not a promise of what it will do for any specific account.

The common mistake: reading every example through last-click

The most common mistake in reading Amazon advertising examples is judging every format against last-click attribution, then concluding that display and DSP "don't work" because they don't show sales the way a search ad does. Last-click attribution was never built to answer that question and can't. It credits whichever ad sat closest to the purchase, which systematically favors search over the awareness spend that helped create the search in the first place.

The second common mistake — one we've made ourselves early in an account — is letting Sponsored Display and DSP report the same purchase that Sponsored Products already claimed, so the combined total looks better than it is. The fix isn't a smarter dashboard. It's reconciling the two in Amazon Marketing Cloud so a purchase gets counted once, and running actual holdouts or matched-control tests to see whether display spend caused an incremental sale or just took credit for one Sponsored Products would have closed anyway.

When your own numbers don't match the example

If your account doesn't look like the numbers above, work through it in this order before assuming something's broken:

• Check the format mix first. A $3 CPC on Sponsored Products in a competitive category isn't the same story as a $3 CPC on a DSP display placement — compare like against like.
• Check the attribution window and overlap. If Sponsored Display and Sponsored Products are both claiming the same order, your blended ROAS is inflated and the real number is lower than it looks.
• Check new-to-brand share before cutting an awareness format. A format with a mediocre ROAS but a high new-to-brand percentage may be doing exactly the acquisition job it was bought for.
• If a fix didn't work, don't re-run the identical test. A holdout that shows no lift once is a result. Running the same campaign again hoping for a different answer usually just spends more to confirm what you already know.