The finding
After one bad month (a 128% ACOS in January, against a December run-rate near 49%), this brand concluded that Amazon no longer worked. It paused all 73 of its Sponsored Products campaigns, let its hero products run out of stock, and effectively walked away from the channel.
The account-level data told a completely different story. Across three years, the Amazon business had generated about $157,500 in ad-attributed revenue on roughly $89,600 in spend, a 1.76x lifetime blended ROAS. In the five months before the brand pulled the plug, it was running at 1.86x blended ROAS on a profitable streak. The channel was not failing. The campaign structure underneath it was, and the blended numbers were hiding that.
The data
- 73 lifetime Sponsored Products campaigns: all now paused
- 55% of all lifetime campaigns: structurally unprofitable
- 414 of 900 lifetime targets: generated zero sales
- $13,747 spent: at an ACOS above 100%
- Aug to Dec 2025: 1.86x blended ROAS, 53.7% ACOS, a 5-month profitable streak
- January 2026: 128% ACOS (the single bad month that triggered the pause)
- Lifetime: roughly $157,500 ad-attributed revenue on roughly $89,600 spend (1.76x)
- Brand search demand: still running at about 25,000 searches per year
The Account Diagnostic read the account at the target level, not the blended level, which is where the real picture lived.
ALL LIFETIME CAMPAIGNS, BY STRUCTURE
A single blended ACOS averaged these together, which is why one bad month looked like a dead channel.
The leak mechanic
Here is what the blended ACOS was concealing. Every single profitable campaign on the account, without exception, was defensive: branded keywords, own-ASIN product targeting, and Auto loose-match that caught the brand's own long-tail. The most profitable target on the entire account was product-targeting on the hero ASIN, at 4.38x lifetime ROAS. In other words, the only thing the account was profitable at was capturing demand it already had: people searching the brand by name, or already looking at its listings.
A real acquisition layer, campaigns built to bring in new customers who were not already searching the brand, had never been built. The 55% of unprofitable campaigns and the 414 dead targets were dragging the blended average down far enough that one bad month tipped it into looking like a broken channel. The brand read a structural problem (no acquisition engine, plus a pile of dead weight) as a channel failure, and abandoned three years of accumulated buyer signal because of it.
The recommendation
Do not abandon the channel. Rebuild the structure. Cut the dead targets and the structurally unprofitable campaigns, keep the proven defensive core that was always working, and for the first time build an actual acquisition layer on top: non-brand and category campaigns designed to reach new customers, not just re-catch the ones the brand already has.
Restock the hero products so there is inventory to sell. Sequence it as a phased relaunch (Foundation, Acquisition, Scale) rather than a flip-the-switch restart.
The projected impact
The value of this finding is the diagnosis itself. It is the difference between a brand permanently walking away from a profitable channel and a brand rebuilding it correctly.
Modeled conservatively against the account's own pre-pause performance, the corrected relaunch path projects roughly $298,600 in total revenue over 12 months (ad-attributed plus organic), exiting at an annualized run-rate near $486,000. That first year alone is close to twice the brand's entire three-year historical Amazon revenue. Every one of these figures is a modeled 12-month projection, not a recovered or guaranteed number.
This is the kind of finding a single Account Diagnostic surfaces.
Most accounts are hiding at least one. Seven days, a 30 to 50 page report, and a prioritized action list.