Why did my product rating drop? Four causes and how to tell them apart
When a product rating drops, the cause is usually one of four things: something about the product or its delivery changed, the product has so few reviews that a single bad one moves the average, you changed who you ask or when, or some reviews were removed or regrouped. Find the date the drop started, then compare what reviews were about before and after it. The topic that grew is your cause. On Google, first check which rating fell: the store rating covers a rolling 24-month window, so it can slip with no new complaint at all.
What this comes down to
- With few reviews, one bad review moves the average a lot; with hundreds, the same review barely registers.
- A real drop usually starts on a date, and something changed on or just before that date: a supplier, packaging, carrier or listing.
- Changing when or how you ask for reviews changes who answers, which can move the average with no change to the product.
- Compare topics before and after the drop per hundred reviews; the topic that grew is where to look.
First, rule out small numbers
An average of a few reviews is fragile. The arithmetic, which you can check by hand:
| Reviews so far | Average before | After one 1-star review |
|---|---|---|
| 10 | 4.6 | 4.27 |
| 50 | 4.6 | 4.53 |
| 500 | 4.6 | 4.59 |
If your product has a few dozen reviews and dropped by a few tenths after one or two bad ones, you may not have a trend. Read those reviews, reply if they need it, and watch the next month before changing anything.
Before you look for a problem, check whether you are looking at arithmetic.
Find the date it started
A real problem has a start date. Sort the reviews by date and find where the low ratings begin to cluster. Then list what changed in the weeks before that date:
- A new supplier, batch, formula or material
- New packaging, or a new warehouse or carrier
- A new product page, photos, title or price
- A peak season that stretched delivery and support
Customers review what they received, usually days or weeks after it shipped, so look a little earlier than the drop itself.
Compare what the reviews were about
Split the reviews into before and after the date, tag both sides with the same topics, and compare mentions per hundred reviews. An illustration, not real figures: before the drop, leaks appears in 1 of every 100 reviews of a water bottle; after it, in 12. Everything else is flat. That is the cause, and it points straight at the new lid supplier.
The Review Analyzer tags topics, sentiment and severity on each review, so the before and after sets can be compared on the same labels. The guide on when to analyse covers fair comparisons in more detail.
Did you change who you ask?
A drop with no new complaint topic can mean the reviewers changed, not the product. If you moved the review request earlier, you may now hear from customers who have not used the product yet. If you started asking every customer instead of a few, you now hear from the quiet middle, who tend to rate lower than the enthusiasts who wrote before.
That second kind of drop is not bad news. The new average is likely closer to what customers actually think, and it is the honest one to build on. The guide on when to ask covers the timing.
Reviews that were removed or regrouped
Sometimes the reviews did not get worse; the set changed. Platforms remove reviews that break their rules. Google’s product ratings policy, for example, removes reviews written with a conflict of interest, such as by employees. Remove a batch of positive reviews and the average falls with no new complaint.
On Google, check which rating fell
Google shows two different star ratings, and each can move without a single new complaint:
| Rating | What it rates | Can fall with no new complaint |
|---|---|---|
| Product rating | One product, from reviews you and others send Google | A review feed that stopped updating (the program asks for uploads at least once a month); changed GTINs, MPNs or brand, which Google says its display depends on |
| Store rating | Your whole store, per country | Its 24-month window: if your best stretch was two years ago, the average falls as those reviews age out; below about 100 eligible reviews or 3.5 stars, it may stop showing at all |
A store rating that slipped while product reviews look steady can be the window moving, not the product. The differences between the two are covered in product ratings vs store ratings.
Questions
Why did my product rating drop suddenly?
Check whether the product has few reviews, where one bad review moves the average a lot. If not, find the date the drop started and look for what changed just before it: supplier, packaging, carrier, listing, or how you ask for reviews.
How much can one bad review lower my rating?
It depends on how many reviews you have. With 10 reviews averaging 4.6, one 1-star review brings it to about 4.27. With 500 reviews at 4.6, the same review brings it to about 4.59.
Can asking more customers for reviews lower my rating?
It can. Asking every customer brings in people with moderate experiences, who tend to rate lower than enthusiasts. The new average is likely closer to what customers really think.
Read next
When to analyze customer reviews: a monthly habit and five extra moments
Reply daily, but count topics in monthly batches. The five moments that call for an extra analysis, and how to compare reviews before and after a change.
Why customers leave negative reviews, and where each fix lives
Six causes behind negative reviews, from the expectation gap to delivery and service, who owns each fix, and how to find the top cause in your own reviews.
Product ratings vs store ratings on Google: what is the difference?
Google shows two kinds of stars. Where product ratings and store ratings come from, how many reviews each needs, and why product stars often fail to appear.