Why Is a Star Rating Almost Useless on Its Own?
Because it compresses everything you need to know into one number, and the interesting information is in what got compressed.
Take two gyms, both at 4.3 stars. The first has 800 reviews spread over six years with complaints about peak-hour crowding. The second has 60 reviews, 40 of which arrived in a single fortnight, and the recent ones are about billing disputes. Same rating. One is a busy gym, the other has a problem with your money.
We read gym reviews at volume while maintaining this directory, across tens of thousands of facilities in 162 countries, and the pattern that emerges is consistent: the average tells you almost nothing, and the shape of the review set tells you nearly everything.
How Do You Read a Review Set Properly?
Four steps, about five minutes total.
1. Sort by newest and read ten. Ignore the average entirely at this stage. A gym that was excellent in 2022 and changed ownership in 2025 has an average that describes a business which no longer exists.
2. Look for repeated complaints, not loud ones. One furious review about a staff member is noise. Three separate people mentioning the same broken machine across three months is an operations problem, and it will still be broken when you arrive.
3. Weight complaints by category. Not all negatives are equal:
| Complaint type | How much it should worry you |
|---|---|
| Billing, cancellation difficulty | High. Structural, will affect you. |
| Broken equipment, repeated | High. Reflects maintenance budget. |
| Cleanliness, changing rooms | High. Correlates with everything else. |
| Crowding at a specific hour | Only if you train at that hour. |
| Staff friendliness | Low. Highly subjective, staff churn. |
| Price | Low. You can see the price yourself. |
4. Check the positives for specifics. This is the fake-review test, and it's more reliable than any of the detection tools. A genuine five-star review mentions something concrete: a class, a coach's name, the 24-hour access, the fact the squat racks are free at 6am. A purchased one praises the atmosphere and the friendly staff in words that would fit a dentist's office equally well.
What Does a Manipulated Review Set Look Like?
Three signals, and you want at least two before drawing a conclusion.
Clustered dates. Genuine reviews arrive as an irregular trickle. Fifteen five-star reviews inside one week, especially on a business that averages two a month, is a campaign.
Uniform length and register. Real reviews vary wildly, from four words to four paragraphs. A run of reviews that are all two polished sentences is a tell.
Empty reviewer profiles. An account whose only ever review is this gym, posted the same week as eleven others like it, is not a customer.
What isnota reliable signal, despite what you'll read: a high proportion of five-star reviews. Plenty of good small gyms have genuinely delighted members, and a 4.8 from 200 reviews accumulated over four years is just a well-run gym. Judging by the ratio alone gets you the wrong answer often enough to be worse than useless.
How Does Gym Navigator Verify Its Listings?
This is the part where a directory should be specific rather than reassuring, so here is the actual process, including its limits.
Where the data comes from. Facility information comes from public sources, principally Google Places: name, address, opening hours, contact details, aggregate rating, review count and photos. We do not generate ratings ourselves and we do not edit them.
What gets checked. Listings are re-verified on a rolling schedule. Entries that stop resolving, because a gym closed, moved or rebranded, are redirected to the relevant location page or removed rather than left as dead pages. That happens continuously rather than in an annual sweep.
How locations are organised. Every facility is placed in a country, region and city hierarchy so you can browse from where you are rather than searching by name. When a region appears twice under different spellings, the duplicates get merged and the old paths redirect, which keeps one place from having three competing pages.
What we do not do, and won't pretend to. We have not visited 32,000 gyms. Nobody has. Any directory implying a physical inspection of every listing at this scale is describing something that did not happen. What we verify is that the data is accurate against its source and that the page still describes a facility that exists.
What Never Influences a Listing?
Worth stating plainly, because it's the question every directory should answer and most don't.
- No paid placement. A gym cannot buy a listing, a higher position, or the removal of a review. There is no mechanism to do it, not a policy against a mechanism that exists.
- No editing of ratings or reviews. They come from the public source as they are.
- Commission does not touch the directory. Some outbound equipment links earn a commission, which is disclosed on every block that carries one and explained in full on ouraffiliate disclosure. None of it affects which gyms appear or in what order.
- Ordering is based on location, data completeness and relevance. A gym with hours, photos and contact details is more useful than one with a name and nothing else, and it ranks accordingly. That's the only editorial thumb on the scale, and it rewards complete data rather than anyone's payment.
What Do We Get Wrong, and How Do You Spot It?
A directory of this size has errors in it right now. Pretending otherwise would undermine everything above, so here are the specific failure modes we know about.
Stale hours. Opening hours change more often than any other field, particularly around public holidays and after ownership changes. A listing verified three months ago can be wrong today. Always phone before a first visit if the timing is tight.
Closed facilities that still resolve. A gym closes, its public listing lingers, and our page persists until the next re-check catches it. The gap between closure and correction is the single most likely error you'll hit, and it's why entries that stop resolving get redirected rather than silently deleted.
Duplicate regions. Administrative names arrive from source data in several spellings for the same place. County Dublin, Co. Dublin and Dublin can all appear as separate branches, splitting one city's gyms across three pages. We merge these and redirect the losing paths when we find them, and there are certainly more we haven't found.
Thin entries. Some facilities exist in public data as little more than a name and a pin. Those pages are honest about having little to say rather than padded with generic copy about the benefits of exercise.
If you spot any of these, the fix takes one email. What we won't do is quietly remove a page because a gym asked us to, since a listing disappearing on request is exactly the mechanism that makes directories untrustworthy.
How Should You Combine Reviews With Everything Else?
Reviews are one input, and on their own they're the weakest of the ones available.
The sequence that works: shortlist on paper, verify in person. Use the directory to filter by location and hours, read the review sets to eliminate the ones with structural problems, then visit the surviving two or three at the hour you'd actually train. Ten minutes standing on the floor at 6pm on a Tuesday beats an hour of reading.
Our60-point gym scoring frameworkturns that visit into something measurable, with the peak-hour test and a ninety-second equipment count. Reviews feed one criterion out of six in that framework, which is roughly the weight they deserve.
Here's a position I'll defend: online reviews are systematically worse for gyms than for restaurants, because the thing that determines whether a gym works for you, whether the equipment you need is free at the hour you train, is the thing reviewers are worst at describing. Someone who trains at 10am genuinely does not know what 6pm looks like. That's not dishonesty, it's a blind spot the format can't fix.
Where Do You Start?
Browse by where you are, not by brand.Gyms in Brooklyn,gyms in Liverpool,gyms in Adelaideandgyms in Montrealeach list ratings, review counts, hours and contact details so you can eliminate most candidates without leaving your sofa.
Then score the survivors with thegym scoring framework, and if the answer comes back that no local option fits, run thehome gym break-even mathbefore assuming a membership is the only route.
Found something wrong in one of our listings? Tell us atsupport@gymnavigator.com. A directory this size will always contain errors, and the honest response to that is a correction route rather than a claim of perfection.
One last thing worth saying, because it cuts against our own interest. If a gym near you has forty reviews and you know three people who train there, ask them instead. Three first-hand accounts from people whose training resembles yours beat any review set, any directory listing, and this article. Everything above exists for the situation where you do not have that, which is most of the time, in a new city, at the start.