Multi-location review dashboard: 7 metrics to track
Rating, volume, velocity, response rate and time, rating distribution, themes: the review metrics to track per location, how to read them and what to do.
- multi-location review dashboard
- google review metrics
- restaurant review kpis
- review reporting per location
- multi-location review management software
- track google rating across locations
Written with the help of AI, then checked before publication: sources, links and prices. Our editorial policy
A multi-location review dashboard should answer one question in a minute: which restaurant needs attention this week, and why. Seven metrics are enough, tracked location by location: displayed rating and period rating, volume, velocity, response rate, response time, rating distribution and recurring themes. Each one is read against the location's own trend, then by combining metrics with each other.
This article explains how to calculate each metric, how to read it and what action to take. Network organization (profiles, roles, charter, collection) is covered in our guide to managing Google reviews for multiple restaurants.
The 7 metrics and how to calculate them
A useful metric has a fixed definition, identical across all locations; without that, comparisons mean nothing.
| Metric | Calculation | Suggested period | What it measures |
|---|---|---|---|
| Displayed rating | Average of all reviews on the profile | Cumulative | What guests see |
| Period rating | Average of reviews received in the period | Rolling 30 or 90 days | Current experience |
| Volume | Total number of reviews | Cumulative | Profile credibility |
| Velocity | Reviews received per month | Month, compared with the same month last year | Health of review collection |
| Response rate | Reviews replied to / reviews received x 100 | Rolling 30 days | Team discipline |
| Response time | Median time between review and reply | Rolling 30 days | Responsiveness |
| Rating distribution | Share of 1, 2, 3, 4 and 5 star reviews | 90 days | Consistency of the experience |
| Themes | Mentions per topic, positive and negative | Month | Root causes |
The table has eight rows for seven metrics: rating is tracked in two forms, and that is the most important nuance.
Displayed rating and period rating
The displayed rating sums up years of reviews, so it moves very slowly; the period rating shows what guests experience today. A restaurant displayed at 4.5 whose reviews from the last 90 days hover around 4.1 is declining, even if nothing shows on Google yet. Track the gap between the two: it is the earliest warning you have.
Volume and velocity
Volume tells you whether the profile inspires trust; velocity tells you whether collection still works. A location can have many cumulative reviews and receive almost none now: its rating then depends on old reviews. Compare velocity with the same month last year rather than with the previous month, to neutralize seasonal effects (summer terrace, annual closure, holidays).
If velocity drops for good, the problem is often practical: QR code removed, reminder turned off, team no longer asking. Our guide to getting Google reviews explains how to restart collection within Google's rules.
Response rate and response time
Response rate measures whether the team replies; response time measures whether it replies on time. Use the median rather than the average: one review forgotten for two months distorts the whole average. Measure from when the review is posted to when the reply is sent, keeping in mind that Google then reviews each reply, usually within 10 minutes but sometimes up to 30 days, according to Google's help on managing reviews.
Rating distribution
Distribution reveals what the average hides. Two restaurants at 4.2 can be very different: one mostly receives 4 stars, the other alternates between 5s and 1s. A "U-shaped" distribution, with many extreme ratings, signals an inconsistent experience: dinner service different from lunch, a less trained weekend team, a kitchen that cannot keep up on busy days.
Recurring themes
Themes turn a number into a cause. Sort comments by topic (waiting time, welcome, food quality, price, cleanliness, noise) and count positive and negative mentions per location. In a group where every restaurant serves the same menu, a negative theme rising in only one location points to a local problem, not a concept problem.
Reading the dashboard: combinations that matter
A single metric is often misleading; two metrics moving together tell a story. Here are typical combinations and the action that follows.
| What you see | Likely reading | Action |
|---|---|---|
| Period rating falling, velocity stable, "waiting time" theme rising | Service under strain | Review staffing on the time slots mentioned in reviews |
| Velocity collapsing, rating stable | Collection has stopped | Check QR codes and whether the team still asks |
| Response rate falling at a single location | Local organization problem | Temporarily reassign reply approval |
| Median response time growing, rate stable | Replies done in catch-up mode | Set a daily approval slot |
| U-shaped distribution | Experience varies by service | Cross-check negative reviews with days and times |
| Sudden spike of 1-star reviews with no text | Possible fake review attack | Document and report suspicious reviews |
| Sudden spike of 5-star reviews with no text | Solicitation to check | Make sure no team offers anything in return |
The last case deserves attention. The Google Maps policy on prohibited content bans offering incentives for reviews and selectively soliciting positive reviews: an unexplained spike at one restaurant should be checked before Google notices it.
Example: a 4-restaurant network in one week
Here is what a weekly reading looks like, using a fictional example built for illustration.
| Location | Displayed rating | 90-day rating | Reviews this month (same month last year) | 30-day response rate | Median response time | Rising negative theme |
|---|---|---|---|---|---|---|
| Downtown | 4.6 | 4.6 | 38 (35) | 100% | 1 day | None |
| Station | 4.4 | 4.0 | 41 (40) | 95% | 1 day | Waiting time |
| Retail park | 4.3 | 4.3 | 9 (27) | 100% | 2 days | None |
| Seafront | 4.5 | 4.5 | 22 (24) | 48% | 9 days | None |
Reading: the Station restaurant has a service problem (period rating falling, waiting time mentioned); the Retail park has lost its collection (velocity divided by three, rating stable); the Seafront has a capacity problem for replies. Downtown needs nothing. Three different actions that a group average would have hidden.
Setting your own alert thresholds
Alert thresholds are calibrated on each location's history, not on a general standard. A restaurant receiving ten reviews a month cannot have the same velocity thresholds as one receiving a hundred.
A simple method: record each metric over the last twelve months, note its usual month-to-month variation and trigger an alert when the gap clearly exceeds that variation. Then adjust after two or three months: a threshold that fires every week gets ignored, and one that never fires is useless.
Where to get the data
Google provides part of the data, but not review metrics in a form ready to compare. Profile performance statistics (views, searches, directions, calls, clicks) can be downloaded for several profiles at once into a single spreadsheet, and search terms are updated at the start of each month, with a delay of up to 5 days, according to Google's help on profile performance. These statistics complement review metrics; they do not replace them.
For reviews, you have two options: a spreadsheet updated by hand every week, manageable up to two or three locations, or software that calculates the metrics automatically. In Ma Belle Note, each location's dashboard shows average rating, review volume, response time and response rate, and the monthly AI highlights surface strong and weak themes with a suggested action. You switch from one location to another within the same account. To compare tools on this criterion, see our comparison of restaurant review software and our method for choosing review software.
Common reading mistakes
- Looking at the group average. It smooths out exactly the gaps that call for action.
- Calculating response rate since the profile was created. A location that stopped replying last month keeps a good cumulative rate.
- Comparing velocity with the previous month. August against July says nothing; August against August says something.
- Reacting to a single review. One negative review is a case to handle, not a trend.
- Tracking metrics nobody reads. Each metric needs an owner and an associated action.
Want to see these metrics calculated on your own Google profiles? Start the 7-day free trial, no credit card required.
Frequently asked questions
Which metrics belong in a review dashboard?
Seven metrics are enough, tracked location by location: the displayed rating and the period rating, total review volume, velocity (reviews received per month), response rate, response time, the distribution of 1 to 5 star ratings and the themes that recur in comments. Metrics added beyond that are rarely read and do not change decisions.
What is the difference between the displayed Google rating and the period rating?
The displayed rating is the average of every review since the profile was created: it moves very slowly. The period rating is the average of reviews received over the last 30 or 90 days only. When it stays below the displayed rating, the displayed rating will eventually drop. It is the most useful early warning metric.
How do you calculate your review response rate?
Divide the number of reviews you replied to by the number of reviews received over the same period, for example the last 30 days, then multiply by 100. Calculate it over a rolling period rather than since the profile was created: a cumulative rate hides a location that stopped replying last month.
Should you compare your restaurants with each other?
Yes, but carefully. A downtown restaurant and a retail park restaurant have neither the same guests nor the same review volume. First compare each location with its own trend, then use cross-location comparison to spot a gap on a specific theme, such as waiting time, which points to a local problem.
How often should you check the dashboard?
Daily for sensitive reviews to handle, weekly for velocity, response rate and response time, monthly for the period rating, rating distribution and themes. Underlying metrics move slowly: checking them every day pushes you to react to noise rather than to a trend.
