How many Google reviews does a roofer need to reach the map pack top three?

Roofing owners hear one story about the map pack. The storm and replacement companies have piled up hundreds or thousands of Google reviews, so a roofer with a few dozen cannot reach the top three, and the number to chase is whatever the leader in the city carries.

We tested that belief, and one of our own. Eight days ago we measured electricians and offered a rule of thumb: read the twenty review counts in your map pack, take the middle one, multiply by three, and that is roughly where the top three begins. A rule like that is only useful if it travels. So on October 2, 2026 we captured the Google Maps top 20 for “roofing contractor” in 22 US cities, 11 large metros and 11 small cities: 440 results, 422 roofing contractors after screening, 22 searches.

The short answer: the leader’s total is not the entry price, and the multiply-by-three rule does not survive the trip to roofing intact. It splits by market size.

The number to write down first

The median roofing contractor in the map pack has 80 reviews and a 4.9 rating. The median business in the top three has 182.

Maps position	Listings	Median reviews	Median rating
1 to 3	66	182	4.9
4 to 7	86	106	4.9
8 to 12	110	64	4.9
13 to 20	160	51	4.9

That pooled ladder is two different ladders stacked together, and almost nobody lives in the average of them.

	Large metros	Small cities
Median listing	153 reviews	33 reviews
Median top three listing	246 reviews	119 reviews
Median floor (least reviewed of the top three)	134 reviews	69 reviews
Markets with a top three listing under 100 reviews	3 of 11	9 of 11
Top three, as a multiple of the market’s median listing	1.5	4.1

The gap between the groups is significant on a rank test, at 0.541 across 421 listings.

The leader is rarely the most reviewed

The business at position one held the most reviews in its city in 4 of 22 markets. In the other 18, someone with more reviews sat below it. In 12 of the 22 the most reviewed roofer was somewhere in the top three; in one large metro it sat at position 18. The highest count in the sample, 2,508 reviews, was at position 12.

The floor says the same thing from below. Across the 22 markets the least reviewed business in the top three carried a median of 98 reviews. In 12 cities one of the three had fewer than 100; in five, fewer than 50. The lowest floor was 22 reviews, at position three in a large metro.

The storm companies are a smaller share than the belief suggests: 29 brand locations, 7 percent of listings, holding 8 of 66 top three slots at a median position of 12.

Why the rule of thumb splits

Maps position	Large metros, median reviews	Small cities, median reviews
1 to 3	246	119
4 to 7	186	54
8 to 12	239	28
13 to 20	95	18

In small cities, review count is a clean staircase that rises at every step. The rank correlation between reviews and position is 0.517, strong and significant across 210 listings, and the top three carry 4.1 times their market’s median listing, 3.81 times on independent businesses alone.

In large metros the staircase has a landing. Positions 8 to 12 carry a median of 239 reviews, about the same as the top three at 246. Volume separates the bottom of the list from the rest and then stops sorting. The correlation is weaker, at 0.329 across 211 listings, and the top three carry only 1.5 times the market median, 1.34 times on independents.

So the instruction depends on your market:

• Small city: read the twenty counts in your pack, take the middle one, and multiply by four. That is where the top three sits, and in 9 of these 11 cities at least one top three roofer had fewer than 100 reviews.
• Large metro: the middle number times one and a half gets you into the conversation, not to the top. Once you hold about as many reviews as the roofers at positions 8 to 12, something other than volume is deciding the order.

Where review count stops mattering

Pooled, the correlation between review count and position is 0.35, and it holds on the 392 independents at 0.346. Splitting all listings into review count thirds shows where it lives: significant at 0.202 among listings with 1 to 39 reviews, then 0.163 for 39 to 134 and 0.092 for 141 or more, neither of which clears the bar. Going from a handful of reviews to a few dozen appears to matter. Above roughly 140, this sample cannot detect an effect of volume on position, which is not the same as saying there is none.

Eight of the 66 top three businesses carry fewer reviews than the median listing in their own market, a median of 105 against 204 for the rest of the top three. None is a brand location, and 7 of the 8 link to a website of their own.

Rating is a floor, again

Rating had no relationship to position, at 0.01. A perfect 5.0 is common at low volume: 142 listings with a median of 39 reviews, yet only 13 of the 66 top three slots. As in our dentist, auto repair and electrician studies, rating is a floor to clear and volume is the ladder; in large metro roofing even the ladder runs out.

What changed in verification

Our first pass tagged only 10 brand locations, one of them wrongly: a family company matched a brand name inside its own name. Opening the site of every listing whose host appeared in two or more of our cities, or whose name carried a region, found franchises and companies listing offices in other metros. The corrected count is 29. The independents’ top three multiple moved from 2.12 to 1.9; the pooled medians, the floors and the metro and small city multiples did not move.

The bigger catch: seven listings in four large metros link to websites built from one template, with the same headline pattern and layout, a call number, every listing open 24 hours, 71 to 118 reviews and ratings of 4.8 to 4.9. One, listed in Ohio, links to a site about roof repair in New Jersey. We could not confirm a local operator behind any of them and excluded all seven, plus one listing with the same name pattern and no website, two that point at referral sites carrying pages for many cities, and one whose site shows a call number from another state. They sat mostly at positions 13 to 20, which is why the benchmarks barely moved. A roofer reading their own pack should know a few listings beside them may not be roofers. The other seven exclusions were general contractors, gutter and siding companies and a supplier. All 18 are in the dataset with their reasons.

Limits: what this can and cannot tell you

Every business here already ranks in the top 20. This study describes what separates listings inside the map pack, not what separates them from roofers who never appear, and the floor would look different if those businesses were in the data.

We measured position, rating, review count, categories, listed hours, business name and where the website link goes. We did not measure proximity to the searcher, service area, photos, posts, review recency, responses to reviews or paid placement. In large metros, where volume stops sorting the top of the list, any of those could be doing the work.

Results were captured once, on a weekday morning, with one query per city and Google’s default location for it. Maps results move with position, device and time, and roofing review counts move with the weather: a city with a hail season this year may carry counts that reflect it. One listing showed no review count and is reported as missing, not zero. Correlation describes what co-occurs among businesses that already rank; none of it establishes cause.

The number to write down

In a small city: the median roofer in your pack has around 33 reviews, and the top three sit at about four times that. In a large metro: expect a median listing around 153 and a top three around 246, and know that the roofers at positions 8 to 12 carry nearly as many. Past that point, more reviews alone did not visibly move anyone in this sample.

The full dataset, including every listing, the per market floors and the 18 excluded results with their reasons, is below.