Contractor Review & Reputation Statistics for 2026

Reputation statistics are the worst-sourced numbers in local marketing. The same handful of figures circulate for years, each blog citing the last, until the original study is three links back and nobody has checked whether it studied restaurants, retailers or contractors. This page lists what the research actually establishes, names the population each study measured, and flags where the number stops being transferable.

A one-star increase in Yelp rating produced a 5-9% increase in revenue for independent restaurants. That comes from Michael Luca's Harvard Business School working paper, which matched Yelp data against Washington State Department of Revenue filings and used Yelp's rounding thresholds as a regression discontinuity — making it causal rather than correlational.

Because ratings round and consumers filter. Anderson and Magruder found an extra half-star caused restaurants to sell out 19 percentage points more often — a 49% relative increase — with the effect strongest where other information was scarce. Ratings do not work as a smooth slope; they work as a set of cliffs at the display thresholds.

Fewer than most people assume, because the returns fall off a cliff. Northwestern's Spiegel Research Center found purchase likelihood rose 270% between zero and five reviews, then the marginal value of each additional review dropped sharply. The first five reviews are the ones that change behaviour; the fiftieth mostly changes nothing.

No. Spiegel found purchase likelihood peaks between 4.0 and 4.7 and then declines as ratings approach 5.0, because a flawless record reads as filtered rather than earned. A contractor sitting at 4.8 with two hundred reviews is in a stronger position than one at 5.0 with nine.

BrightLocal's 2026 Local Consumer Review Survey of 1,002 US adults found 68% will not consider a business rated below four stars, and 31% require 4.5 or better. That is a hard filter applied before your marketing gets a chance to work — below four stars, most of your ad spend is buying impressions from people who have already excluded you.

For most local searches, yes. 74% of consumers prioritise reviews from the last three months, which means a hundred reviews from 2023 carry less weight than fifteen from this quarter. Reputation is a flow rather than a stock — a business that stops asking starts decaying regardless of its total.

It changes a great deal in both directions. 80% of consumers are more likely to use a business that responds to all its reviews, and 42% are unlikely to use one that ignores them. But 50% view generic templated responses negatively, so a copy-pasted thank-you performs measurably worse than silence in half the audience.

On Google Local Services Ads they feed the ranking directly, which makes reputation a media-efficiency lever rather than a branding exercise. SearchLight's benchmark across 888 contractors and $6.72M of tracked LSA spend recorded a 43.9% blended book rate and $233 cost per paying customer on a channel where review profile carries weight.

Since 21 October 2024, federal ones with teeth. The FTC's rule on consumer reviews and testimonials bans buying positive reviews, writing fake ones, suppressing negative ones, and undisclosed insider reviews — and authorises civil penalties for knowing violations. Platform policy used to be the ceiling; now it is the floor.

Enough to keep a rolling three-month window populated, which for most contractors means somewhere between four and twelve a month depending on job volume. The arithmetic of moving an existing average is unforgiving — a business at 4.2 with 80 reviews needs far more five-star reviews to reach 4.5 than most owners expect.

The two causal studies, without reservation, for direction. The survey figures for consumer expectations, because that is what surveys measure well. And none of the magnitudes as forecasts for your business — every one of them was measured on restaurants or retail products, and no published dataset has run this experiment on trades.