A restaurant owner checks Google Maps on Monday: 4.3 stars. On Tuesday: 4.2 stars. No new reviews. No deleted reviews. Nothing visibly changed. Yet the number moved.
This happens more often than most owners realize. Google restaurant ratings fluctuate for reasons that have nothing to do with new reviews. Spam filters run silently. The rounding algorithm crosses thresholds. Old reviews get removed. Display caching creates temporary mismatches between what you see and what guests see. Understanding how Google calculates and displays your rating removes the panic from these shifts and helps you focus on the changes that actually matter.
Why do Google ratings fluctuate?
Google ratings can fluctuate due to review removals (spam filtering), algorithm updates, or new reviews being posted. Google does not use a simple average — recent reviews carry more weight. A single removed fake review can shift your rating by 0.1-0.3 points.
How Google calculates your restaurant rating
Google uses a weighted average of all visible reviews on your listing. The formula is straightforward: add up all star ratings, divide by the number of reviews. A restaurant with 100 reviews totaling 430 stars has a 4.30 average.
But what Google displays is rounded to one decimal place. That 4.30 shows as 4.3. A 4.25 also shows as 4.3. A 4.24 shows as 4.2. This rounding is where most “mysterious” fluctuations come from.
The rounding thresholds that matter
| Actual average | Displayed | Note |
|---|---|---|
| 4.249 | 4.2 | Rounds down |
| 4.250 | 4.3 | Rounds up (midpoint rounds up) |
| 4.349 | 4.3 | Rounds down |
| 4.350 | 4.4 | Rounds up |
| 4.449 | 4.4 | Rounds down |
| 4.450 | 4.5 | Rounds up |
If your real average is 4.251 and one review gets removed by Google’s spam filter, your average might drop to 4.248 — and your displayed rating jumps from 4.3 to 4.2. One thousandth of a point. Zero meaningful change in quality. But it looks like a significant drop.
6 reasons your Google rating changed overnight
1. Google removed a review as spam
Google’s spam detection runs continuously. It removes reviews it suspects are fake, incentivized, or from accounts with suspicious patterns. When a 5-star review gets removed, your average drops. You’re not notified when this happens. Common — Google removes millions of reviews per quarter.
2. The algorithm refreshed its rounding
Google occasionally recalculates displayed ratings from raw data. If your true average was sitting at 4.2499 and a recalculation yields 4.2501, you’d see a jump to 4.3 with no new reviews. Occasional — especially after Google Maps updates.
3. A reviewer edited their review
Reviewers can change their star rating at any time. A guest who left a 5-star review last year might edit it to 3 stars after a bad revisit. Uncommon but impactful when it happens.
4. A reviewer deleted their Google account
When someone deletes their Google account, all their reviews disappear across every business. Rare for individual impact, but happens at scale.
5. Display caching lag
Google Maps, Google Search, and the Google Business Profile dashboard don’t always update simultaneously. You might see 4.3 on Maps and 4.2 in Search for the same listing. This isn’t a real fluctuation — it resolves within 24-72 hours. Common after receiving new reviews.
6. Review velocity changed your position weight
Google factors review recency into how prominently your rating appears in search results. If you had a burst of 5-star reviews 3 months ago and none since, Google may weight recent reviews more heavily. Ongoing — affects ranking more than the displayed number.
What changes your rating vs. what doesn’t
| Event | Affects rating? | Affects ranking? | You’re notified? |
|---|---|---|---|
| New review posted | Yes | Yes | Yes |
| Review removed by spam filter | Yes | Yes | No |
| Reviewer edits star rating | Yes | Yes | No |
| Reviewer deletes Google account | Yes | Slightly | No |
| You reply to a review | No | Yes (small boost) | N/A |
| Display cache refreshes | No (cosmetic) | No | No |
| Google algorithm update | Sometimes | Yes | No |
| Competitor gets more reviews | No | Yes (relative) | No |
| You add photos to your listing | No | Yes | N/A |
| Review age (older reviews) | Same weight | Less weight over time | No |
When a rating drop actually matters
Matters
- Your rating crossed a threshold (4.0, 4.5) that affects consumer behavior — 87% of guests won’t consider below 4.0
- Multiple legitimate negative reviews in a short period — indicates a real service issue
- Your rating dropped AND your review count dropped — Google may have flagged reviews as suspicious
- Consistent downward trend over 30+ days — not noise, a pattern
Noise (ignore)
- A 0.1 drop with no new negative reviews — likely spam filter or rounding
- Rating differs between Maps and Search — caching lag, resolves in 24-72h
- Rating changed then changed back within a week — algorithm recalculation
- One negative review on an otherwise strong profile (100+ reviews, 4.3+) — statistically insignificant
The math: how one review affects your rating
| Reviews | Current | 1-star impact | 5-star impact | Note |
|---|---|---|---|---|
| 20 | 4.30 | -0.17 | +0.04 | One bad review hurts a lot |
| 50 | 4.30 | -0.06 | +0.01 | Moderate impact |
| 100 | 4.30 | -0.03 | +0.01 | Barely moves |
| 200 | 4.30 | -0.02 | +0.00 | Negligible |
| 500 | 4.30 | -0.01 | +0.00 | Statistically invisible |
The best defense against rating fluctuations is volume. A restaurant with 200+ reviews is nearly immune to individual review impact. A restaurant with 20 reviews can lose 0.2 stars from a single bad day. This is why consistent review generation matters more than chasing a specific number.
What you can actually do about rating fluctuations
- Build review volume. At 200+ reviews, individual spam filter actions or bad reviews barely move your average.
- Maintain review velocity. 10 reviews this month matters more than 50 from last year.
- Reply to every review. Replies don’t change your star average but affect ranking and how prospective guests interpret negative reviews.
- Monitor weekly, not daily. Compare month-over-month, not day-over-day.
- Flag genuinely fake reviews. Google removes about 55% of legitimately flagged reviews, but it takes 5-14 days.
- Don’t buy reviews. Purchased reviews are removed in waves. When 20 fake 5-star reviews get removed simultaneously, the rating drop is far worse than the original problem.