Chrome Web Store ratings: what a 0.1 drop really means

Where Chrome Web Store ratings actually sit

We looked at every crawled extension with at least 10 ratings. Fewer than that and one angry review swings the average by a full star.

Lifetime averages from our crawl of 65,655 extensions, observed 3 Oct 2026.

Almost half sit at 4.5 or higher. So 4.5 isn't a mark of excellence. It's roughly the median ticket to the conversation.

Why a 0.1 drop is louder than it sounds

The store shows one decimal. A tool at 4.46 displays 4.5. A tool at 4.44 displays 4.4. That two-hundredths of a star is invisible in the data and very visible on the card.

So what does the drop cost in installs? Honestly, we can't tell you yet. Our own history started on 2 Oct 2026, and measuring that needs months of before and after.

For current numbers, open the Tools category page.

Big tools rate lower, and that's normal

Observed 3 Oct 2026. User counts are the store's rounded buckets.

Small tools get rated mostly by fans and friends. Big tools get rated by people who had a bad Tuesday. Compare a rating only against tools of the same size.

What tends to cause a drop

Ratings rarely fall for no reason. When we read the newest reviews on listings that lost stars, the same causes come up again and again.

Most of these are fixable within a week. The faster the fix, the fewer one-star reviews pile up.

How to recover from a drop

A lifetime average moves slowly, so recovery takes patience. These steps help.

1. Find the causein the newest reviews.
2. Ship a fix, and say what it fixes in the update notes.
3. Reply to each complaintwith the fix.
4. Ask happy users for a ratingat a moment the tool helps them.
5. Track the count and the averageweekly, with dates.

Don't offer anything in return for ratings. Google's policies treat incentivised ratings as spam, and our guide to bought installs shows how that ends.

Ratings by category

Some categories rate higher than others, so compare within yours. Average ratings by category in our 3 Oct 2026 crawl:

A 4.2 in Accessibility sits above that category's average. The same 4.2 in Travel sits well below. Our categories pages show each one with its counts.

The perfect 5.0 trap

A flawless score looks great on a card. It usually means very few people rated. In our 3 Oct 2026 crawl, 16,181 extensions showed a 5.0 average. Of those, 14,727 had fewer than 10 ratings.

That's why the rating count belongs next to the stars every time. A 5.0 from three people tells you almost nothing, and a 4.5 from thousands are a much stronger sign.

Reading a rating like a buyer

If you're installing a tool, a few quick checks beat staring at the stars.

• Look at the count first. Under 20 ratings, the average can swing a lot.
• Sort reviews by newest. The lifetime average can hide a recent problem.
• Check the last update. A drop in rating after an update often points to a bug.
• Compare with tools of the same size, using the table above.

Highly rated big tools exist. On 3 Oct 2026, Sider had a 4.92 from 114,491 ratings and GoFullPage a 4.89 from 84,154. Scores like that at that scale are rare, which is what makes them meaningful.

For builders: what a rating is worth

We can't yet measure how much a rating change moves installs, since that needs months of our own readings. What we can say is where your rating sits against others, and that's useful on its own.

1. Find your size band in the table above.
2. Find your category average in the category table.
3. If you're below both, the newest reviews will usually tell you why.
4. If you're above both, protect it: ship small, test updates and reply to every complaint.

You can see how rivals in your space compare on our niche pages, with ratings and counts dated.

What to do with a rating you're watching

1. Note the rating and the rating count together, with the date.
2. Put it in its size band from the table above.
3. Check again in 30 days. One reading tells you a level, not a direction.
4. Read the newest reviews on the store when it moves, as our reviews guide suggests.

We're biased, since we build Ext Watch, but this is the thing it's for. Our extension pages show rating and rating count with the date we saw them, and they won't call a rating's direction until we have 30 days of our own readings.

Next time you glance at that half-second number, check the size band first.