Your stores get inspected every day.
Those 'inspections' come from your customers. And a surprising proportion of them post their evaluations of your stores on public review sites.
Your customers have been auditing your stores for 20 years. No one has read the file properly.
A focus on star ratings often misleads and hides the real insight.
Everyone who touches this data does the same thing with it. Total the stars, rank the stores, reply to the angry ones. That ranking is close to meaningless, because most stores do not have enough reviews for their average to mean anything.
So a tool built on that ranking can tell you a store is unhappy. It can't tell you what's really important: which store, what motivated the rating change, or when it changed.
What we do instead
Our motivating question
Where does your operation diverge from its own norm in a statistically significant way, and what is driving it?
Look at every store, not just the loud ones
We pull the full public record for every location you run, going back years. Not a sample. All of it, including listings you may not know people have been writing about.
Take the brand out, then follow your signature items
Most of what a customer says is about your brand overall and would have been said at any of your sites. We hold that constant, and what's left is the site itself. Then we find every review that names any of your signature items and read what it says: made right, made wrong, or missing. It's the same item and the same spec at every store, so any difference is the store.
Find what moved, when, and why
A store that has always run average is not news. A store that changed in March is noteworthy. We date the change, describe the change, tell you what shifted with it, and tell you whether the same pattern holds across enough of your sites to be statistically significant rather than noise.
Measured · Public record only
What it finds
The findings are usually uncomfortable. That is the point of an outside read on public review data.
Words that do not reach the threshold appear too rarely to separate from chance.
A finding is not a diagnosis
The list above is what customers wrote. It is not yet what is wrong. Complaints that read as serious staff behavior problems very often resolve into simple retraining, or a policy change the store could easily make.
They don't take American currency apparently. Tried paying with cash and they told me they couldn't take it. Employees gave me an attitude and turned me away.
Absolutely horrible place. Never a cashier available, and long lines to boot. They are incredibly rude and make you feel like you are there to serve them.
6 reviews in this theme
The site had gone cashless. The clerk had no way to complete the sale and no script for explaining it.
Post the policy change at the door and at the pump, and give staff one sentence to say.
Hey, just so you know... it's illegal to sell air. Have a nice day.
Terrible business practices. You should demand a receipt before buying anything. Bought 2 drinks on sale and a Reese's Peanut Butter Cup, questioned total but couldn't get a paper receipt.
6 reviews in this theme
State statute requires free air at fuel sites, and the pump was set to charge. Receipts were not being issued by default.
Reset the pump, and check every site in that state for the same setting.
Your worker Lucy at your Harrison location was harassing me because I put 6 sugars in my coffee and accused me of stealing.
The old man behind the counter started shouting at me that I can't have a backpack in the beer cave when there's no signage.
5 reviews in this theme
No posted policy on either, so staff were improvising enforcement rules with customers in front of them.
Publish the rule and post it on in-store signs, so nobody has to invent one at the counter.
Always get a reciept. They will scam you by scanning an extra time or not applying the discounts that are advertised.
I started to notice more than once I wasn't getting my full change back. Today they shorted me 25 cents.
16 reviews in this theme
Advertised promotions were not configured at the register.
Audit promo setup against what is currently advertised, chain wide.
What lands on your desk
Measured · Public record only
A 60-site fuel and convenience operator.
| What they knew | What the read showed |
|---|---|
| Foodservice was underperforming somewhere in the chain, and the star ratings did not say where. | 4 stores were drifting away from the chain average on foodservice performance. All 4 began drifting inside the same 8-week window. |
| Six sites had adopted the same branded kitchen program that quarter. All 6 were assumed to be performing alike. | The language that described the change was about waiting, not about the food. It appeared at 2 of the 6. |
| The assumption was a food or supplier problem. | The 4 sites that were fine had staged the rollout over a quarter. The 2 that drifted had opened the program in a hurry, inside of 2 weeks, in the same window their wait complaints started. |
The Insight: this was a rollout-pace problem, not a food problem. Two sites took on a complex kitchen program too fast and never closed the training gap, and their customers described it as experiencing long wait times rather than as bad food. The fix was a training intervention at 2 stores, copied from the other 4, who rolled out after a full training program.
None of this was visible by looking at the star ratings alone. Based on star ratings, only 3 of the 60 stores could be told apart from the overall chain at all.
What we can and cannot see
Caveat. Silence here is not evidence that nothing happened.
We don't need anything to run it. We need somewhere to send it.
After we find it in this analysis, we evaluate its impact in the P&L.
This read tells you which stores are off and what your customers say changed. It does not tell you what the impact to the bottom line would be of correcting any of the identified issues. That requires your own numbers to evaluate.
A Calibration Read needs 3 to 5 changes you've already made, and measures what each one did to sales, store by store, on your POS export. We propose the candidates rather than asking you to pick blind. The Store-to-Store Read is how we know which ones to propose. Movement in the public record is a signal that something happened that is large enough to measure in your sales data.
The fee credits in full toward a Lever Board engagement.
Where this sits within the Seurat analytical journey
We can get further without your data than you would expect. The dollar column is where that stops.
| You give us | You get | |
|---|---|---|
| Store-to-Store Read | Nothing | Which of your stores are underperforming, and what your customers say is wrong with them |
| Calibration Read | One data export | What the changes you've already made actually contributed to in-store sales |
| Lever Board | 2 years of sales data | Every proven lever ranked and suggested for each of your stores, with its lift contribution evaluated |
| Remodel Sequence Read | 2 years of sales data and your capital plan | Which sites on your capital plan should get the money first, which shouldn't get it at all, and what would have to change for the rest |
| Monitoring | Nothing new | Verdicts as the reads settle, plus where a lever is not showing up in the data the way it should, which usually means it is not being executed the way it was specified |