Tony Wang7 min readOne in Three Rated US Businesses Shows a Perfect 5.0. Bus Stops Do Better.
We measured every star rating in 20M US business listings. The perfect score is the single most common rating, and the reason isn't what you'd expect.
The shape of the distribution
Ratings do not fall on a bell curve. They pile up at the top, with a smaller second spike at the very bottom.
| Rating | Listings | Share of rated |
|---|---|---|
| 5.0 | 4,425,421 | 33.9% |
| 4.5–4.9 | 3,743,102 | 28.7% |
| 4.0–4.4 | 2,439,421 | 18.7% |
| 3.5–3.9 | 990,321 | 7.6% |
| 3.0–3.4 | 710,809 | 5.5% |
| 2.5–2.9 | 214,585 | 1.6% |
| 2.0–2.4 | 180,754 | 1.4% |
| 1.5–1.9 | 49,150 | 0.4% |
| 1.0–1.4 | 282,312 | 2.2% |
That bottom reversal is the tell. A business rated between 1.5 and 1.9 needs a genuine spread of bad reviews. A business rated exactly 1.0 needs one review, and that person was angry. There are 282,312 of them — five times as many as the band just above.
The same logic runs at the top, and it is why the 5.0 pile is so large. A perfect score is easiest to hold when almost nobody has rated you.
Only one of the two spikes is an artifact
If both spikes were purely a low-review-count effect, both should melt away as you require more reviews. Only one does.
| Minimum reviews | Listings | Display 5.0 | Display 1.0–1.4 |
|---|---|---|---|
| 1+ | 12,769,478 | 33.6% | 2.1% |
| 3+ | 9,403,516 | 26.8% | 0.5% |
| 5+ | 8,152,123 | 24.0% | 0.3% |
| 10+ | 6,431,383 | 20.1% | 0.2% |
| 25+ | 4,197,484 | 13.7% | 0.1% |
| 50+ | 3,084,307 | 10.3% | 0.0% |
| 100+ | 2,127,650 | 7.6% | 0.0% |
The one-star spike is fully explained by tiny samples: by 50 reviews it is gone. The five-star spike is not. It falls from 33.6% to 7.6%, but 7.6% of heavily-reviewed businesses is still 162,180 companies holding a perfect public score across 100 or more reviews.
Who holds a perfect score at scale
The categories are the answer, and they are strikingly consistent.
| Category | Perfect 5.0 with 100+ reviews |
|---|---|
| dentist | 5,278 |
| chiropractor | 3,543 |
| insurance_agency | 3,403 |
| roofing_contractor | 2,721 |
| car_detailing_service | 2,689 |
| real_estate_agent | 2,642 |
| medical_spa | 2,386 |
| real_estate_agency | 2,314 |
| hair_salon | 2,154 |
| barber_shop | 2,016 |
There is not a restaurant, hotel, bar or shop in the list. Every entry is a service business with a one-to-one customer relationship and a natural moment to ask — the dentist at checkout, the roofer on completion, the agent at closing. These are the businesses best positioned to request a review from someone who just had a good experience, and to request it individually rather than hoping a stranger volunteers.
Contrast that with restaurants and hotels, where reviews arrive unprompted and disproportionately from the disappointed. That is the same asymmetry we found when booking structure beat price at predicting a business's rating, and it points the same way: what a star rating mostly measures is who was asked.
Bus stops outperform the American economy
How we did this, and the caveats
| Decision | What we did |
|---|---|
| Source | Crawlora's Google Business index, 20.0M US listings, queried directly rather than through paginated search |
| Population | Listings carrying a rating above zero — 13.0M of the 20.0M |
| Rating | The star field as displayed on the listing |
| Thresholds | Review-count floors applied to the same population, so each row is directly comparable |
A rating of 0 means never rated, not rated zero. All 7.0M unrated listings are excluded. Treating them as zeros would drag every average in this study downward and is the single easiest way to get this dataset wrong.
"5.0" means displayed 5.0, not a flawless average. The star field is mixed precision: most
records carry Google's rounded one-decimal display value — there are 825,702 listings at exactly
4.8 and 805,021 at exactly 4.9 — while a minority carry a computed mean to several decimals
(4.8333, 4.8571). So a listing shown here as 5.0 may have a true average anywhere from 4.95 up. We
report the number a customer actually sees, which is the number that influences their decision,
but it is not a claim that every review was five stars.
130,319 listings show a 5.0 with a review count of zero. They are concentrated in farms, parks, family practice physicians and real estate agents. This is an internal inconsistency in the data, not a finding, and every review-count-floored figure in this post excludes them.
Category counts are of listings, not companies. A business appearing under several categories is counted in each. The category table ranks where perfect scores concentrate, not how many distinct firms hold one.
This measures displayed ratings, not quality or honesty. Nothing here identifies fake reviews. A dentist with 300 five-star reviews may be excellent and diligent about asking. The finding is that the rating distribution is shaped by who solicits reviews, not that anyone is cheating.
Check the ratings in your own market
The Google Business dataset behind this analysis carries rating, review count, category and geography on 132M+ listings — enough to test how much a star rating is worth in any industry you care about. 2,000 free credits a month, no card.
Related reading
Frequently asked questions
How common is a perfect 5.0 rating on Google?
Among 13.0M rated US business listings it is the single most common rating: 4,425,421 listings, or 33.9%, display a perfect 5.0. That is more than every rating below 4.0 combined. The share falls as review volume rises — 20.1% among businesses with 10 or more reviews and 7.6% among those with 100 or more — but never disappears.
Does a 1.0 rating mean a business is terrible?
Usually it means one person left one angry review. Listings rated 1.0 to 1.4 are five times more common than those rated 1.5 to 1.9, which is backwards for a real distribution and is the signature of a single-review artifact. Requiring 50 or more reviews makes the 1.0 spike vanish entirely, while 10.3% of that same group still display a perfect 5.0.
Which businesses hold a perfect 5.0 on lots of reviews?
162,180 US businesses display a 5.0 across 100 or more reviews, and the categories are consistent: dentists, chiropractors, insurance agencies, roofing contractors, car detailers, real estate agents, med spas, hair salons and barber shops. There is not a restaurant, hotel or shop among them. All are service businesses with a one-to-one customer relationship and a natural moment to ask for a review.
Why are bus stops rated so highly?
Bus stops display a perfect 5.0 in 49.0% of rated cases against a 33.9% base rate, with ATMs at 30.0% and parking lots at 29.9%. Nobody evaluates a bench. Such listings collect occasional cheerful or ironic five-star reviews and essentially never a considered negative one, because there is no service to be disappointed by. It is the clearest case of a high rating carrying no information.
Does a 5.0 in this data mean every review was five stars?
No, and this matters. The star field is mixed precision: most records carry Google's rounded one-decimal display value, with 825,702 listings at exactly 4.8 and 805,021 at exactly 4.9, while a minority carry a computed mean to several decimal places. A listing shown as 5.0 may have a true average anywhere from 4.95 up. We report the figure a customer actually sees rather than a claim about the underlying reviews.