National Review Benchmarks Misrank A Third Of Contractors

The Same Review Count, Two Opposite Answers

The median HVAC contractor in California has 26 reviews. In Arkansas, the median is 4.

A shop with 12 reviews is above California's P25 of 9 but below its median of 26 and comfortably above the median in Arkansas. Both of those shops read the same national benchmark, and it tells them the same thing. At least one of them is being badly misled.

We measured how often that happens. Across roughly 105,000 operating trade listings, we compared reported national and state percentiles for listings in states with at least 100 listings in that trade. HVAC has 44 published states. The national-rank eligibility pool is not fully documented here, so treat the table as the reported historical calculation, not an independently reproduced current score:

Across the published-state eligibility pool, between three and four contractors in ten have national and state ranks differing by more than ten percentile points. Not slightly off: more than ten percentile points off.

Review Volume Varies by Market

Possible explanations include transaction volume, company age, customer mix, review-request practices and platform use. This directory snapshot does not identify their effects or prove that rural shops accrue reviews more slowly. Compare your own completed-job count and review-request history before attributing a rank difference to geography.

National and local percentile ranks answer different comparison questions. Neither directly measures service quality. A geographic difference in ranks is a reason to choose a relevant comparison set, not proof that geography causes the difference.

A contractor's local comparison set includes the shops a customer sees alongside it in search results. That local search comparison set is useful for customer acquisition. National ranks can still describe the broader directory; a state boundary may also be too broad for a local service area.

What "Enough Reviews" Actually Means

There is no national answer to how many reviews a contractor needs, which is why a national target alone does not answer a local positioning question.

Nationally, the median HVAC shop has 9 reviews, the bottom quartile has 4 or fewer, and the top 10% has 65 or more. Those national values are descriptive context, not a local score. The median differs by state, and a state can also have a wide internal range: California's 9-to-76 IQR is wider than the 4-to-26 state-median span cited here.

The useful version is local. In California, HVAC breaks at 4 reviews for the bottom decile, 9 for the bottom quartile, 26 at the median, 76 for the top quartile and 187 for the top 10%. In Arkansas the same five breakpoints are 2, 3, 4, 8 and 17.

Clearing a state median is one descriptive comparison, not proof of enough reviews to win local searches. Compare nearby services, review recency and customer quality signals as well. For most contractors that is a much smaller number than the national top-decile figures imply, and the time to get there depends on completed jobs, customer response and your actual review-accrual history; this directory does not establish a quarter-long improvement.

The Part Nobody Selling You Reputation Software Will Say

A review count is not a revenue forecast for a contractor of your size.

Across the whole market, review count and revenue do move together, and that correlation gets quoted a great deal. It is driven by very large firms. Inside the under-$20M range, where most contractors operate, the earlier analysis reported a weak association of r=0.11. The matched revenue source, sample count and join method are not documented here, so that coefficient remains an unreproduced reference, not evidence that adding reviews has no effect.

Star ratings cluster tightly in this directory snapshot; a rating-to-revenue comparison is not published here. Ratings cluster tightly across every trade we examined, with most sitting near a 4.0 median, so rating is a poor differentiator for getting into a conversation at all.

The honest reading of both facts together:

Test how reviews influence residential inquiries. A shop with two reviews reads as a risk no matter how good both of them are. That is a plausible credibility mechanism to test using local inquiry and booking outcomes, not a causal lift measured by this directory.

Commercial account economics need their own evidence. If your work is commercial, review count alone does not establish your revenue opportunity. Check which buildings you serve and what those buildings are actually worth, which is a different question with a different answer.

Commercial procurement can depend on relationships, scope, capacity and buyer requirements. The unreproduced within-ICP correlation does not identify why those outcomes differ.

What To Do

Rank yourself against your state, not the country. The review percentile calculator does this for five trades across 164 published state curves, and it will tell you when your state's sample is too thin to publish rather than quietly falling back to a national figure.

Test a review request after completed work. Compare request timing, response and subsequent inquiries in your own records. Compute your gap to the relevant published state or local comparison; it can be much larger than a handful. A rank difference does not establish the fastest way to gain reviews or clients.

Compare reviews with other operating priorities. Reviews may influence residential inquiries; test that against your own lead and booking records. For commercial work, customer and job profitability may be another useful decision surface. The directory does not establish which activity produces more revenue.

How This Was Measured

Review counts for operating listings only, drawn from a scrape of roughly 110,000 US trade contractor locations from public business listings, April 2026. Closed businesses are excluded: a closed shop's review count describes a business nobody can hire.

Every listing is assigned to a single primary trade in a fixed priority order to avoid double counting. Percentiles are nearest-rank over observed counts, so every breakpoint is a real review count rather than an interpolated one. A state is published only where it carries at least 100 listings for that trade, which is why HVAC has 44 published states and plumbing has 26.

The misranking rate is computed directly rather than estimated: for every shop in a published state, we take its percentile within its state and its percentile nationally, and count the share where those differ by more than 10 points.

This is directory data, and we grade it as such. It is suitable for market-structure analysis, not for underwriting an individual business. Aggregates only, and no individual listing is identified.

The full distributions, including star ratings and listing claim rates by trade, are published in the contractor reputation benchmarks, free to cite with attribution.