Reputation score
What is a reputation score: reputation score explained
A reputation score is a single number summarising a business's public review performance, usually built from star average, review volume, recency, response rate and how the reviews are spread across sites. There is no standard formula. Each platform and each software supplier weights the inputs differently, so two scores for the same store are not comparable.
You will also see reputation index, review score, rating score, or a vendor's branded name for its own composite. The related term is review management, which is the practice this number tries to measure.
What does a reputation score mean?
It means one supplier's opinion of your review profile, compressed. That is not a criticism, because the compression is genuinely useful for spotting movement across a group of stores. It becomes a problem the moment the number is treated as an audited fact.
Two things are worth separating:
- The star average. Published by the review platform itself, visible to shoppers, and calculated the same boring way everywhere.
- The composite score. Calculated by a dashboard, visible only to you, and weighted according to choices someone made in a product meeting.
How is a reputation score calculated?
The star average has one formula: total stars awarded divided by number of ratings.
A composite typically normalises several components to a common scale and applies weights, for example star average at 40 percent, review volume at 20 percent, recency at 20 percent, response rate at 10 percent, and spread across sites at 10 percent. Those weights are illustrative rather than standard, which is precisely the point. Ask any supplier quoting you a score to show the weighting, and ask what happens to it when a competitor gains 200 reviews.
Reputation score explained: a worked example
A rooftop has 120 reviews averaging 4.2 stars. That is 504 total stars. The general manager wants 4.5 by year end.
Solve for the number of new five star reviews needed. 504 plus 5n, divided by 120 plus n, equals 4.5. That gives 0.5n equals 36, so n equals 72. Seventy-two consecutive perfect reviews, with not one dissatisfied customer among them, to move 0.3 of a star.
Run the same maths for a store with 600 reviews and the number becomes absurd. This is the single most useful thing to understand about the metric: a mature review profile is heavily damped, so the score lags real change in the operation by months, and any score that jumps quickly deserves a look at how it moved.
How a reputation score differs from CSI
- Who runs it. CSI is typically a manufacturer survey of your own customers. A reputation score is built from public reviews anyone can read.
- Who sees it. CSI goes to you and the OEM. A star average sits next to your price on a listing page in front of every shopper.
- Who answers. CSI samples the customers the survey reaches. Reviews skew towards people motivated enough to post, at both ends.
- What it pays. CSI is often tied to manufacturer bonus money. Reputation is tied to whether someone clicks your listing at all.
They can move in opposite directions for entirely legitimate reasons, and a store chasing both with the same tactic usually ends up gaming one of them.
What the score cannot tell you
- Which department is the problem. One number over sales and service drive hides the fact that they often perform nothing alike.
- Whether it was earned. A score built on filtered requests reads identically to a score built honestly, right up until a platform acts on it.
- What to fix. The text does that. A drop of 0.2 tells you to go and read, not what to change.
- How you compare. Only against stores measured with the same weighting and the same source sites, which in practice means only against your own history.
Using the number without being used by it
Treat the composite as a trend line for your own store and nothing more, and put the star average and the review text in front of the people who cause them. A target expressed as "4.5 by December" pushes a manager towards the request list, because that is the only lever that moves fast.
A target expressed as "every closed order gets asked, every review gets answered within two working days, and the top complaint theme changes by December" pushes them towards the operation. The score follows from the second one, slowly, which is the honest speed at which reputation moves.
