How to build a low rated businesses list for an area, and read it fairly
Set a rating band, require enough reviews, and compare with the norm for each kind of business. A low rated businesses list that points to real problems, not noise.
Part of the guide Location intelligence automation
A useful low rated businesses list is not every place under 4 stars. Pick one kind of business, find all of it in your area, keep those whose rating sits in a band you choose (for example 2.5 to 3.7), drop the ones rated by only a handful of people, and judge the band against what is normal for that kind of business. What is left are places where many customers have said, consistently, that something is not working.
Consultants, agencies, equipment and training suppliers, franchise development teams and investors use such lists to find businesses that may want help, or that may be for sale. This piece explains how to set the band, what a low score can and cannot tell you, a worked example, the traps, and an AI employee that refreshes the list on its own.
Low compared with what?
Ratings are not spread the same way across trades. In xMap's places database for November 2026, dentists near the centre of Toronto average 4.7 out of 5, while hotels within 10 km of the city's centre average 3.8. In London, supermarkets nearest the centre average 3.6 and beauty salons 4.7. A dentist rated 4.0 is well below its neighbours; a supermarket rated 4.0 is above most of them.
| Kind of place | City | Average rating | Rated 4.5 or more |
|---|---|---|---|
| Supermarkets | London | 3.6 | 38 of 423 rated |
| Pharmacies | London | 3.9 | 127 of 413 rated |
| Hotels | Toronto | 3.8 | 47 of 167 rated |
| Dentists | Toronto | 4.7 | 313 of 366 rated |
xMap's places database, November 2026. London supermarkets and pharmacies and Toronto dentists: of the 437, 438 and 460 places nearest each centre; Toronto hotels: all 312 within 10 km of the centre.
So set the band per trade. A rule of thumb that works well: start the top of the band about half a star below the trade's average in that city, and read a sample before you trust it.
What a low rating does and does not say
A rating is an average of what people chose to post. Read it for what it is.
- It does say that a meaningful number of customers had a worse experience than at similar places nearby, if the review count is high enough.
- It does not say why. It could be staff, cleanliness, waiting times, prices, parking or one bad month that many people wrote about.
- It does not say the business is failing. A busy budget hotel can be full every night at 3.4.
- It is not fixed. A new owner or a refit can move a rating within months, which is why a list needs refreshing.
The places data gives the rating and the number of reviews, not the text of the reviews. To learn the reason, someone still has to read them.
Setting it up in xMap Flows
Nearby places has a Minimum rating setting but no maximum, so the top of the band is set with Filter. Each Filter holds one rule; place them in a row when an item must pass several.
- Nearby places for one kind of business in your area, with Most places set high enough to cover it.
- Filter: Rating at most 3.7.
- Filter: Rating at least 2.5, which also removes the handful of places with a rating near 1 that are often closed or mislabelled.
- Filter: Reviews at least 50, so a place rated by three friends does not count.
- Sort and limit by Reviews, highest first, so the places most people complained about lead the list.
A worked example
Everything below is made up. A hospitality training firm in Cape Town wants guesthouses and small hotels that guests rate poorly, to offer a front-desk course.
- Nearby places returns 300 guesthouses and hotels within 10 km of the city centre.
- The firm sets the band at 2.5 to 3.6, because, in this made-up case, local hotels average around 4.1. 46 places fall in it.
- Requiring at least 40 reviews leaves 29.
- A salesperson reads the latest reviews of the top ten by review count: seven mention slow check-in or rude staff, three mention noise from building work.
- The seven go to the course offer; the three go on a list to revisit in six months.
The reading step is what turns a number into a reason to call.
Mistakes to avoid
- One band for every trade. 3.9 is weak for a dentist and good for a supermarket.
- No review minimum. Places with very few reviews swing wildly and fill the list with noise.
- Mixing kinds. Hotels and hostels rated together make the hostels look bad.
- Leading with the score. Telling a business owner they are rated badly opens no doors. Lead with the problem you can solve.
- Never refreshing. A list from last year names places that may have fixed things.
An AI employee that keeps the list honest
Set it up once and the list arrives on its own each month, with only the places that are new to the band. The sales prospecting guide covers how such lists feed a sales team.
- Schedule, the first day of each month.
- Area of interest with your city or districts.
- Nearby places, then the three Filters above.
- Only new, so each email holds only places that entered the band since last time.
- xMap LLM to write one line on what kind of help suits each place, from its type, rating and review count.
- Email to the team.
The template Businesses with low ratings and no website follows this shape for agencies, and Hotels guests rate lowest, for my guest experience software does it for hotels.
Questions
What rating counts as low?
It depends on the kind of business and the city. Compare with the average for that trade nearby: a dentist at 4.0 is low where dentists average 4.7, a supermarket at 4.0 is high where they average 3.6.
Why require a minimum number of reviews?
A rating from five people can swing by a whole star with one review. Requiring 30 to 50 reviews keeps the list to places where the low score reflects many customers.
Can I see why a business is rated badly?
The list gives the rating and the number of reviews, not the review text. Someone needs to read the latest reviews to learn the reason.
Are businesses with no rating included?
No. A place with no rating does not pass a rating rule in Filter, so it leaves the list. Collect those separately if you want them.
Templates that do this
Read next
More on business and places data · Every article · Templates