Amazon Attribution

You run a Google Shopping ad that lands on your Amazon listing. The sale closes on Amazon. Your Shopify pixel never fires, your ROAS report shows the campaign spending money against zero revenue, and the conversion that actually happened is invisible to every tracker you own.

Amazon Attribution is the tag that closes that gap. It measures how traffic from non-Amazon channels - search, social, email, display - performs once it reaches an Amazon product page, reporting the on-Amazon outcome back to the advertiser who paid for the click.

Two brands in the teardowns run straight into this

Most D2C accounts in the series drive every ad to their own site. Two do not.

Graza runs five Google Shopping ads whose landing URLs point at amazon.com/graza/oliveoil - “Amazon Olive Oil - Shop On Amazon,” “Buy With Prime,” “Graza Sizzle Oil - Buy On Amazon” - a dual-marketplace play its Google teardown calls unique in the whole set. Magic Spoon does the same at smaller scale: three search and Shopping ads driving to www.amazon.com, all last shown in October 2025.

Both brands keep their primary conversion tracking on their own Shopify sites. That is the exact split Amazon Attribution exists to reconcile - Google budget spent on one side of the wall, the purchase recorded on the other.

Why a tag instead of a pixel

On a D2C site you drop a Google or Meta pixel on the thank-you page and the conversion reports itself. Amazon does not allow third-party pixels on its product detail pages, per Pattern’s teardown of the tool, which is the whole reason a separate product exists.

Instead you wrap each ad’s landing URL in an Attribution tag. Amazon then reports clicks, detail-page views, add-to-carts, and purchases against that tag, using its own logged-in shopper data as the source of truth (Amazon’s own guide).

The tool is free. Access is gated behind an active Brand Registry account with brand-representative status, available to sellers in the US, Canada, UK, Germany, France, Italy, and Spain, with vendors and KDP authors eligible through the ad console (Amazon Ads support).

The measurement stays wrong in opposite directions

Attribution tags tell you Amazon’s half of the story and nothing else. The tool has no visibility into what happens after a shopper leaves Amazon and enters a brand’s own funnel, and the Amazon conversion cannot be fed back into Google or Meta for automated bidding, because the data lives inside Amazon’s closed ecosystem, per Improvado’s 2026 breakdown.

So a brand running graza.co and an Amazon listing off the same Google campaign gets two partial pictures. The Shopify side undercounts because it never sees the Amazon sale. The Amazon side undercounts because it is click-only and last-touch - a shopper who clicks a mobile ad and buys later on the Amazon desktop app slips the tag entirely.

Practitioners put a number on the cost of stitching those halves together by hand: 12 to 18 hours a month reconciling incompatible exports across the two systems, by Improvado’s estimate.

The window narrowed on January 1, 2026

Amazon’s attribution logic is not fixed, and it tightened this year rather than loosened. On January 1, 2026 Amazon retired the blanket 14-day view-through window across its view-based placements and replaced it with a shopping-signal enhanced last-touch model, one that uses machine learning to judge whether an ad view actually influenced the purchase, documented by AdBeacon.

The change hit view-through inventory - Sponsored Brands and Sponsored Display vCPM, plus DSP in-store placements - and left click attribution untouched. DSP advertisers reported double-digit declines in attributed revenue in 2026 reporting purely from the tighter model, with no change in actual sales, per Code3’s analysis.

For the Graza and Magic Spoon case - a click on a Google Shopping ad - the click-based path is the one that survives. The lesson sits underneath the mechanics: the number an Amazon report hands you is a function of a model Amazon can revise, and it revised it downward.