Amazon advertising · Attribution
Your ad bought the click. A reseller closed the sale.
Brand halo attribution credits third-party purchases back to your Sponsored Brands campaigns. On a catalogue with resellers on it, that makes your worst-exposed campaign look like your best one.
The short answer
Amazon’s brand halo attribution counts any purchase of your brand’s ASINs inside the attribution window — including purchases made through the “Other Sellers” link, where a reseller, not you, ships the unit and keeps the retail margin. Your ad console reports that sale as campaign revenue.
If you do not hold the featured offer on the pages your Sponsored Brands traffic lands on, you are paying retail-level advertising costs to move inventory you already sold at wholesale — and reading the result as retail performance. In one account we manage, close to 40% of Sponsored Brands attributed revenue sat on ASINs the brand does not sell at all.
To find your own number: pull the Sponsored Brands attributed purchases report, split promoted from brand halo, and cross the halo ASINs against your featured offer percentage. It takes about ten minutes.
What brand halo attribution counts, and what it does not tell you
Brand halo is the portion of a campaign’s attributed sales that came from products other than the ones in the ad. It is matched on the brand of the listing, not on the seller account that paid for the advertising.
Every Sponsored Brands campaign reports two kinds of sale. The first is straightforward: someone clicked your ad and bought one of the products in the creative. The second is brand halo — they clicked your ad and bought something else of yours instead. A different size, a bundle, a product they found in your Store.
That second bucket is useful. It is also where the problem hides, because “something else of yours” is defined by the brand on the listing, not by who is selling it. As Code3’s breakdown of the attribution model puts it: “if a shopper makes a purchase from a third-party via the ‘Other Sellers’ link, that sale would be included in Brand Halo reporting.”
Nothing here is a bug, and nobody at Amazon is hiding it. It is an attribution model built around the brand rather than around the seller account paying for the ads — which works fine until somebody else is selling your products.
| Promoted | Brand halo | |
|---|---|---|
| What triggered it | The shopper bought an ASIN that was in the ad creative. | The shopper bought a different ASIN carrying your brand. |
| Who sold the unit | Whoever held the featured offer at purchase. | Whoever held the featured offer at purchase — including a reseller, reached through the Other Sellers link. |
| How it appears in your ROAS | As revenue. | As revenue, indistinguishable from the first row unless you split the report. |
| What you actually earned | The full retail margin. | The wholesale margin, banked whenever you sold the unit to the reseller. |
Why the featured offer decides whether Sponsored Brands spend is an investment or a subsidy
The featured offer — the buy box — determines who fulfils the sale on any page your advertising sends traffic to. If you do not hold it, the click is a cost to you and the revenue is somebody else’s.
Buy box share is the upstream constraint on everything downstream in a retail media account. Most advertisers treat it as a merchandising problem and keep it out of the media conversation entirely, which is exactly how it goes unnoticed for a quarter at a time.
Sponsored Products is targeted at the ASIN, so featured offer problems there tend to surface as eligibility and impression loss. Sponsored Brands behaves differently: it routes traffic to a Store or a multi-ASIN landing page, and the shopper then navigates to whatever page they like. The campaign has no view of who holds the featured offer on the page where the click lands. That structural difference is the whole reason this leak hides inside Sponsored Brands rather than inside the rest of your Sponsored Ads.
What we found in one account: nearly 40% of attributed revenue on ASINs the brand does not sell
TNOMADS account data — not an industry benchmark
In one account we manage, close to 40% of Sponsored Brands attributed revenue sat outside the brand’s own catalogue, while the session-weighted featured offer share across that catalogue was under half.
The account belongs to a leading connected outdoor technology brand operating across e-commerce, retail and subscription services: physical products that carry proprietary software, an app, cellular connectivity and recurring data plans. It sells on Amazon through a single Seller Central account, with no vendor relationship, and a long tail of resellers on its listings.
Our approach started from an anomaly rather than a problem. The Sponsored Brands campaign carried the best ROAS in the account, which is what prompted the audit — in a catalogue with known buy box pressure, the strongest-looking campaign is the one worth checking first.
The data showed close to 40% of the attributed revenue sitting on ASINs outside the brand’s own catalogue — reseller listings. Upstream, the session-weighted featured offer share was under half: most of the traffic being paid for landed on pages where somebody else was the default offer.
The part that actually costs money
It is not the individual sale. Those units were bought from the brand at some point, so the revenue is thinned to wholesale margin rather than lost outright. The cost is in the decision that follows: the most reseller-exposed campaign posts the best ROAS, so it wins the budget argument every month, and that budget buys more clicks into pages the brand does not control.
How to audit your own brand halo exposure in ten minutes
Split your Sponsored Brands attributed purchases into promoted and brand halo, check the halo ASINs against your own catalogue, and weight your featured offer share by sessions. This is the sequence we run.
- Pull the Sponsored Brands attributed purchases report. It carries the purchased ASIN and flags each purchase as Promoted or Brand Halo — the split the standard campaign view does not give you.
- List the halo ASINs against your own catalogue. Anything that is not a SKU you sell is revenue a third party earned and your campaign reported.
- Pull featured offer percentage by ASIN from your business reports, weighted by sessions rather than averaged flat. In practice we found a flat average is the single most misleading number in this audit — it lets dormant SKUs flatter the ones carrying real traffic.
- Recalculate ROAS on own-catalogue revenue only. That is the number to plan against. Keep the reported figure beside it so the gap stays visible to everyone reading the deck.
- Repeat in thirty days. Buy box share moves weekly; a single reading tells you where you are, not which way you are going.
What to do once you know your exposure
Do not pause the campaign. Fix the featured offer, re-point the traffic at ASINs you control, and report two ROAS figures instead of one.
The instinct is to switch the campaign off. That is usually wrong — the campaign is reporting accurately on a situation created elsewhere, and pausing it buys a cleaner report rather than better margin.
What changed after the optimisation in the account above was the order of operations rather than the spend level. We fixed the featured offer where it was recoverable, since buy box share is the constraint everything else sits under. We re-pointed Sponsored Brands at the ASINs the brand actually controls instead of the Store page while exposure stayed high. We began reporting own-catalogue ROAS and reported ROAS side by side, so no budget decision was made on the blended figure alone. And we took the reseller question to whoever owns distribution — a reseller holding the buy box on your best ASIN is a channel problem that advertising is being asked to absorb, and no bid adjustment fixes it.
Where this recurs across several retailers rather than one, it usually belongs in a retail media strategy conversation rather than a campaign one. If you would rather have someone else run the numbers, that is what our retail media audit covers.
Questions we get asked about brand halo and resellers
Does brand halo attribution include sales made by other sellers?
Yes. Brand halo is defined by the brand on the listing, not by the seller account running the ads, so a purchase made through the Other Sellers link on your ASIN is included in the halo figure your campaign reports. This is the single most misunderstood part of Sponsored Brands reporting, and the reason a campaign can post strong ROAS while your own sales stay flat.
How do I find out how much of my ad revenue went to resellers?
Pull the Sponsored Brands attributed purchases report, which breaks each purchase into Promoted or Brand Halo and names the purchased ASIN. Cross the halo ASINs against your own catalogue: anything you do not sell is revenue a third party earned. Then check your session-weighted featured offer percentage on the ASINs that remain, because owning the ASIN is not the same as owning the sale.
Is inflated ROAS from brand halo a reporting error?
No. The attribution is working exactly as documented — it measures the influence of your ad on purchases of your brand, which is a reasonable thing to measure. The error is on the reading side: treating a brand-level attribution figure as a seller-level revenue figure, and then setting budgets with it. The fix is a second number, not a support ticket.
Should I stop running Sponsored Brands if resellers hold my buy box?
Usually not. Sponsored Brands does work that Sponsored Products cannot — category defence, new-to-brand acquisition, Store traffic. What should change is where it points and how it is judged. Re-point the campaign at ASINs you control, judge it on own-catalogue revenue, and treat the buy box as the thing to fix rather than the thing to advertise around.
How often should we check featured offer share?
Monthly at minimum, weekly on any catalogue with active resellers or recurring suppression. Buy box share moves faster than most reporting cycles, and a SKU can lose the featured offer and keep absorbing ad spend for weeks before anyone connects the two. Session-weight it so the ASINs carrying real traffic drive the number.
Does the same problem exist on Walmart, Instacart or other retail media networks?
The attribution mechanics differ by network, but the underlying exposure does not: wherever a retailer attributes sales at the brand or item level and a third party can win the sale, your reported ad revenue can include units you did not sell. On networks without a buy box the equivalent leak is usually out-of-stock or unavailable items still absorbing spend. The discipline is the same — check availability before you scale spend.
Want this run on your account?
We manage Amazon Sponsored Ads for brands where the buy box is contested and the reporting does not tell the whole story — reported ROAS beside the revenue you actually earned, ASIN by ASIN.
See how we run Amazon PPC Or book a discovery call and we will look at your attributed purchases report together.Sources and disclosure
- External: Code3 — A Deep Dive into Brand Halo Attribution. Source for the attribution window, brand-level matching, and the inclusion of third-party purchases made through the Other Sellers link.
- External: Intentwise — Sponsored Brands attributed purchases report. Source for the Promoted vs Brand Halo field and the purchased-ASIN detail the audit above relies on.
- TNOMADS first-party: all account figures are from a live client engagement we manage. They are rounded, and the client, category and absolute spend are withheld. They describe one account and are not presented as a category benchmark.
Published 14 September 2026. Last updated 14 September 2026.