Quality20 August 2026 · 4 min read
The rating moved three weeks before the reviews explained it.
A packaging supplier changed a component without flagging it. The first measurable consequence was not a review — it was the average rating drifting down while the review count barely moved.
By SellerMate
Calder & Finch Supplements
Health & Household
- Marketplace
- Amazon.co.uk (UK)
- Catalogue
- 4 ASINs, one hero SKU
- Rules in place
- 1
- Rating floor reached
- 4.33.9 → 4.3Where the slide stopped, versus where a “below 4.0” threshold would have caught it.
- Detection lead time
- 34 daysBetween the movement alert firing and a 4.0 threshold rule firing.
- Units shipped after detection
- ~900~4,800 → ~900Affected units that still went out before the seal specification was reverted.
- The situation
- A packaging supplier switches to a thinner induction seal without notice. Units start arriving with the seal broken. Almost nobody reports it, because a customer with a broken seal requests a refund rather than writing a review — and the ones who do write take two to three weeks to get round to it.
- Rules in place
- Star rating drops more than 0.1
- How it plays out
- The rating goes 4.5 → 4.4 and the alert fires. That date brackets the shipment window, which identifies the batch. The seal specification is reverted before the next production run, and the rating bottoms out at 4.3 rather than continuing down. Under a conventional “below 4.0” threshold the same discovery arrives about five weeks later.
The reviews eventually told us it was the seal. What we could not have worked out from the reviews was when it started, and that is the only part that told us which pallet to pull.
Why the date is the useful output
Knowing the rating fell is worth something. Knowing which day it started to fall is worth a great deal more, because it turns an open-ended quality investigation into a lookup.
Reviews describing a physical defect lag the shipment by two to four weeks — the time to receive, use, get annoyed and write. So the day the rating turns sits much closer to the shipment date than the day the explanatory reviews appear. Working backwards from the turn gives a shipment window, and a shipment window gives a batch.
Without the history you get “the rating is 4.1 and there are complaints about the seal”. With it you get “it started on the 14th”, which is an answer.
Why a movement rule beats a threshold here
A “below 4.3” rule is the right default for steady-state monitoring. It is the wrong rule for the sixty days after any change to what ships in the box, because it only fires once the damage has accumulated enough to cross a line.
A movement rule — more than 0.1 in either direction — fires on the first step. It is noisier, which is exactly why it belongs in a risk window rather than in the permanent setup: run it through the window, then relax back to a threshold.
Reading the review count alongside it
Every check records the review count next to the rating, and the relationship between the two is what identifies the cause.
A rating falling while the count rises means new negative reviews, which is the case here. A rating falling while the count falls means Amazon removed reviews. A rating falling while the count stays flat usually means recency weighting or a variation merge. Three unrelated problems that look identical if the star number is all you have.
- Rating down, count up → new negatives. Read them for the shipped date.
- Rating down, count down → review removal. Not a product problem.
- Rating down, count flat → weighting or a catalogue change.
The takeaway
Tighten the rating rule to a 0.1 movement for sixty days after any supplier, packaging or formulation change. It is the cheapest insurance available on a physical product.
Last updated 23 August 2026
Set the rule while you are thinking about it.
Check any listing free. No Amazon login, no card.
Check a listing