Star Rating
A star rating is the average score, usually from 1 to 5 stars, that guests give a business on review sites, maps and delivery apps.
| Star Rating at a glance | |
|---|---|
| Also called | Average rating, review rating, overall rating, bubble rating (Tripadvisor) |
| Scale | 1 to 5 stars on most platforms |
| Formula | Total stars ÷ number of ratings that count |
| Shown as | One decimal (such as 4.3), plus star icons in half-star steps on many sites |
| Key platforms for restaurants | Google, Yelp, Tripadvisor, Uber Eats, DoorDash, booking platforms |
| Varies by platform | Which reviews count, recency weighting, rounding |
| Typical shape | J-shaped, with most ratings at 5 stars and a smaller peak at 1 star |
| Research | Luca (HBS, 2011) and Anderson and Magruder (Economic Journal, 2012) on Yelp ratings and restaurant demand |
| Related metrics | Review volume, review velocity, review recency, reputation score |
Key takeaways
- Your star rating is the average of the ratings a platform counts, and each platform decides which ratings count, how recent ones are weighted and how the result is rounded.
- The same restaurant can show different star ratings on Google, Yelp, Tripadvisor and delivery apps, so comparing locations works best within one platform.
- Rounding matters, because guests see a rounded number or half-star icons, and research on Yelp found that crossing a half-star threshold changes restaurant demand.
- The distribution behind the average, especially the share of 1- and 2-star ratings, often tells you more than the average itself.
- A star rating summarizes one signal, and a reputation score combines it with review volume, recency, responses and sentiment.
What is Star Rating?
A star rating is the average score guests give your business on a review platform, usually on a scale from 1 to 5 stars. On Google, Yelp, Tripadvisor and delivery apps it appears next to your name, often before a guest reads a single review.
Behind that one number sit three decisions each platform makes for itself: which ratings count, whether recent ratings carry more weight, and how the result is rounded for display. Understanding those decisions helps you read your ratings accurately and compare locations fairly.
For a restaurant brand, the star rating is often the first filter a guest applies when choosing where to eat or order. It also feeds into how visible you are in local search, and it is a core input to any reputation score.
How to calculate a star rating
The basic calculation is a simple average:
Star rating = (5 × number of 5-star ratings + 4 × 4-star + 3 × 3-star + 2 × 2-star + 1 × 1-star) ÷ total number of ratings
Suppose one of your locations has 200 Google ratings: 110 five-star, 50 four-star, 18 three-star, 10 two-star and 12 one-star. The total number of stars is 550 + 200 + 54 + 20 + 12 = 836.
Dividing 836 by 200 gives an exact average of 4.18. Google would show this as 4.2, while a site that draws its star icons in half-star steps would show four full stars, because 4.18 is closer to 4.0 than to 4.5.
The same distribution also tells you that 80% of ratings are 4 or 5 stars and 11% are 1 or 2 stars. Those two shares are worth tracking alongside the average.
Star rating calculator
Enter how many ratings you have at each star level. You'll see your exact average, how it would typically be displayed, and how sensitive it is to new reviews.
The calculator uses a plain average and conventional rounding. Platforms that filter reviews or weight recent ratings will show a somewhat different number.
How each platform calculates its star rating
Each platform documents its method to a different degree. The table summarizes what each platform has published about its calculation, together with how the rating appears to guests.
| Platform | Which ratings count | How the average is built | How it's displayed |
|---|---|---|---|
| All ratings published on Google for the place | Average of those ratings, on a 1–5 scale | One decimal, with star icons | |
| Yelp | Recommended reviews only | Average of recommended reviews | Half-star icons, plus a number to one decimal |
| Tripadvisor | All posted review ratings | Simple average | Nearest tenth since early 2025; bubble icons in half steps |
| Uber Eats | Consumer ratings of orders | Average that gives priority to the most recent ratings | Rating with a rating count |
| DoorDash | Ratings from verified DoorDash orders | An algorithm that may give more weight to recent feedback | Rating on the store page |
Google star rating
Google's Business Profile Help page, Understand review scores for local places and businesses, explains that scores run from 1 to 5 stars and that the review score is the average of all ratings published on Google for that place.
The same page notes that it can take up to two weeks for the score to update after a new review.
Yelp star rating
Yelp's recommendation software sorts reviews automatically. According to Yelp Support, reviews that aren't currently recommended don't count toward a business's overall star rating.
As a result, a restaurant can have more reviews on its page than the ones its Yelp average is based on. Yelp's star icons move in half-star steps, and its numerical rating adds precision to one decimal without changing the stars.
Tripadvisor bubble rating
Tripadvisor calls its star rating a bubble rating. Its FAQ on the refined bubble rating explains that from late February 2025, ratings for restaurants, hotels and attractions are rounded to the nearest tenth instead of the nearest half.
The calculation itself stayed the same, a simple average of all posted review ratings, and the bubble icons still round to the nearest half. Tripadvisor's popularity ranking is a separate measure that also considers the recency and quantity of reviews.
Delivery app star ratings
Delivery apps rate merchants from order-level feedback. Uber's ratings policy describes merchant ratings on Uber Eats as averages of consumer ratings that give priority to the most recent ones.
DoorDash's merchant help describes its store rating as based on verified DoorDash orders, calculated by an algorithm that may give more weight to recent feedback. Because neither app publishes a full formula, your own order-level data is the best way to understand what moves these ratings.
Star rating rounding and display
Guests rarely see your exact average. They see a rounded number, a row of star icons, or both, and small differences in your underlying average can produce visibly different results.
| Exact average | One decimal | Nearest half star |
|---|---|---|
| 4.18 | 4.2 | 4.0 |
| 4.24 | 4.2 | 4.0 |
| 4.26 | 4.3 | 4.5 |
| 4.74 | 4.7 | 4.5 |
| 4.76 | 4.8 | 5.0 |
Rounding thresholds have real consequences. Luca's Yelp study used them to isolate the effect of ratings on revenue, and Anderson and Magruder's Learning from the Crowd (Economic Journal, 2012) found that an extra half star made San Francisco restaurants sell out 19 percentage points more often at the 7 p.m. peak.
For you, the practical lesson is to know where each location sits relative to the next threshold. A location averaging 4.24 on Yelp shows the same four stars as one averaging 3.76, and a modest improvement could lift it to four and a half.
Star rating distribution
Two restaurants with the same star rating can have very different guest experiences. The distribution of ratings across 1 to 5 stars shows whether an average reflects consistent good service or a mix of delighted and disappointed guests.
Online ratings tend to follow a J-shaped distribution: many 5-star ratings, a smaller cluster of 1-star ratings and relatively few in between. Nan Hu, Paul Pavlou and Jie Zhang described this pattern in Overcoming the J-shaped Distribution of Product Reviews (Communications of the ACM, 2009) and argued that the average alone is a poor proxy for quality.
| Location | 5★ | 4★ | 3★ | 2★ | 1★ | Average |
|---|---|---|---|---|---|---|
| A (consistent) | 40% | 50% | 10% | 0% | 0% | 4.3 |
| B (polarized) | 70% | 15% | 0% | 5% | 10% | 4.3 |
Location B has the same 4.3 average with a 15% share of 1- and 2-star ratings, which usually points to an inconsistency such as slow service at certain dayparts. Sentiment analysis of the low-rated reviews tends to show which issue it is.
Star rating vs. reputation score vs. CSAT and NPS
| Metric | What it measures | Who sets it | Best for |
|---|---|---|---|
| Star rating | Average public rating on one platform | Guests, with the platform's rules | How new guests see you before visiting |
| Reputation score | Composite of rating, volume, recency, responses and sentiment | You or your software vendor | One summary across platforms and locations |
| CSAT | Satisfaction with a visit, from your own survey | You | Spotting problems at a touchpoint |
| NPS | Likelihood to recommend, from your own survey | You | Loyalty trends |
The star rating is public and platform-controlled, which makes it the number guests actually see. A reputation score adds signals such as review velocity and review recency, while CSAT and NPS capture private feedback from guests who never post a review.
What is a good star rating for a restaurant?
A good star rating depends on the platform, your category and the restaurants guests compare you with. Platforms differ in who reviews and which reviews count, so a 4.2 on Yelp and a 4.2 on Google describe different things.
The most reliable benchmarks are internal and local: your own trend, the spread between your locations on the same platform, and nearby competitors in your category. Hash House A Go Go, for example, improved its Google ratings to 4.9 stars across all locations.
A location-level view of reputation usually reveals that a brand average hides a few locations pulling it down.
How to track star ratings across locations
Tracking works best when every location's ratings from every platform arrive in one place. A unified inbox gathers new reviews as they arrive, and review aggregation lets you report on them together.
A weekly report per location typically shows the current average per platform, the change since last month, the rating distribution and the number of new reviews. An operator report puts those figures in front of each general manager.
Recording your exact average, before rounding, makes small trends visible weeks before they change the number guests see.
Limitations of the star rating
- Self-selection. Guests with very good or very bad experiences are the most likely to post, which produces the J-shaped pattern.
- Different rules on each platform. Filtering, recency weighting and rounding make cross-platform comparisons approximate.
- Small samples. A location with 20 ratings can move by more than a tenth of a star with one review, so trends need context.
- Lost detail. An average hides whether problems are about food, speed, accuracy or staff.
- Manipulation. Fake and incentivized reviews distort averages, which is why the FTC's 2024 rule on fake reviews bans buying them and suppressing negative ones.
How to improve your star rating
A lasting improvement comes from fixing what low ratings describe and from hearing from more of your satisfied guests. Both parts matter, because a fix without new reviews takes a long time to show in an established average.
- Finding the themes behind low ratings. Grouping 1- and 2-star reviews by topic shows which issue affects the most guests, and operational insights can tie those themes to specific locations and dayparts.
- Responding to every negative review. A specific, calm reply, which an AI review manager can draft in your brand's voice, and a recovery offer can bring a guest back and show future readers you listen. Your review response rate is the measure to watch.
- Inviting every guest to review. Fair review solicitation through receipts, QR codes and follow-up messages keeps new ratings flowing. Review boosting automates the invitations.
- Asking everyone the same way. Sending only happy guests to public sites is review gating, which Google's policies prohibit.
Results at scale are possible: Baskin Robbins Malaysia tripled its Google ratings and achieved 10x ROI in 8 weeks, and Island Wing Co. used Momos to boost customer ratings by 35%.
Best star rating tools for restaurants
For a restaurant brand, a good star rating tool covers every platform your guests use, shows ratings and distributions by location, and helps your team reply quickly. Reports that link low ratings to specific topics make the data actionable.
| Tool | Best known for |
|---|---|
 MomosBest pick | Guest AI for multi-location restaurants. Reviews from Google, reservation and delivery platforms arrive in one Unified Inbox, AI categorization shows which topics drive low ratings, and the Alfie assistant drafts and sends replies, with reporting across every location. |
 Birdeye | Reviews, listings and customer messaging for multi-location businesses |
 Reputation | Enterprise reputation management across reviews, listings and social media (formerly Reputation.com) |
 Chatmeter | Local brand reputation management with review and sentiment analytics |
 ReviewTrackers | Review monitoring and reporting across many review sites |
 Yext | Business listings and information management across search engines, maps and directories |
Each tool takes a different approach, so it helps to compare them against your own platforms and number of locations.
Common star rating mistakes
- Comparing ratings across platforms. A 4.4 on Google and a 4.0 on Yelp may describe the same guest experience.
- Watching only the rounded number. The exact average shows movement weeks earlier.
- Ignoring the distribution. A steady average can hide a growing share of 1-star ratings.
- Reacting to small samples. A single review at a low-volume location is a signal to read, and a trend needs more data.
- Chasing ratings with incentives. Paying for or gating reviews breaks platform rules and, in the US, the FTC's rule.
- Overlooking delivery app ratings. They often reflect packaging, accuracy and temperature problems that your team can fix.
History of the star rating
Star scales for hotels and restaurants appeared in printed travel guides long before the internet. Online, consumer review sites made the star rating a crowd-sourced number: Tripadvisor (founded 2000) uses a five-point "bubble" scale, and Yelp (founded 2004) rates local businesses from one to five stars.
Google's local reviews, now part of Google Business Profile (renamed from Google My Business in 2021), placed star ratings directly in Search and Maps. Delivery apps such as Uber Eats and DoorDash added ratings of their own, collected after each order.
Research soon showed that these averages affect restaurant demand. In a Harvard Business School working paper, Reviews, Reputation, and Revenue: The Case of Yelp.com (2011), Michael Luca found that a one-star increase in Yelp rating led to a 5 to 9% increase in revenue for independent restaurants, with no measurable effect for chain-affiliated restaurants.
Display formats have also evolved. Yelp added numerical ratings rounded to one decimal place next to its star icons, and in early 2025 Tripadvisor began showing ratings to the nearest tenth of a bubble.
Frequently asked questions
How is a star rating calculated?
Most platforms add up the stars from every rating they count and divide by the number of those ratings. Google describes its review score as the average of all ratings published for a place. Some platforms then adjust the result, for example Uber Eats gives more weight to recent ratings, and Yelp counts only recommended reviews.
Why is my rating different on Google and Yelp?
Each platform has its own guests, its own filters and its own rules. Yelp's recommendation software leaves some reviews out of the average, delivery apps weight recent orders, and the people who review on each site differ. A gap between platforms is normal and is best tracked over time.
Why hasn't my Google rating changed after a new review?
Google's Business Profile Help says it can take up to two weeks for a review score to update after a new review is posted. With many existing reviews, one new rating also moves the average only slightly, so the rounded number you see may stay the same.
What is a good star rating for a restaurant?
No single benchmark applies across platforms, cities and cuisines. The most useful comparisons are your own trend, the gap between your locations on the same platform, and the ratings of nearby restaurants guests compare you with. The share of 1- and 2-star ratings is also worth watching.
How many reviews does it take to raise my star rating?
It depends on how many reviews you already have and how far you want to move. The calculator on our online reputation management page estimates how many 5-star reviews a target rating would take.
Can I remove a bad rating to improve my average?
Platforms remove reviews that break their policies, such as fake or off-topic ones, and you can flag those. Honest negative ratings stay. A calm, specific reply to the negative review and a fix for the underlying problem do more for your average over time.
Is a 5.0 rating better than a 4.8?
Not always in a guest's eyes. A perfect score built on a handful of reviews can look less convincing than a 4.8 backed by hundreds of recent ones. Volume and recency give the average its credibility.