What a best contractor list cannot tell you

Of the 770 listings in the index, 703 include a star rating taken from a public map profile. For those 703 rated companies, the mean is 4.78 and the median is 5, on a scale from 1 to 5 whose full range appears somewhere in the dataset. Most ratings are at the top, and only a small number are lower.

The problem becomes concrete when you count the wall. More than half of the 703 rated companies, 394 of them, carry a rating of exactly 5.0. Another 143 companies have ratings between 4.8 and 4.99, which puts 537 in total at 4.8 or above. Only 37 companies hold a rating below 4.0. Any ranking that begins with stars would list 537 nearly equal numbers in order, and the gaps it shows would mostly be a byproduct of how many reviews each company happens to have collected.

This compression is a familiar pattern in public review averages for service businesses, and concrete work is no exception. For a directory, the consequence is what counts: this dataset's star rating distinguishes the bottom 37 companies from everyone else fairly well, but it does not distinguish the top 537 from one another at all. A ranked list of the top of that field would imply a precision the underlying numbers lack.

Review counts differ far more than ratings do. Across all listings, the index contains 30,003 reviews. The median company has 20 reviews. The largest single count is 1,656. That difference from the middle of the field to the top is eighty-fold, and it represents real variation, not rounding.

At the lower end, the spread is wide too. 145 companies have five reviews or fewer, while 69 have 100 or more. A company with four reviews and a 5.0 average is not comparable to one with 400 reviews and a 4.8 average, even though a naive sort would place the first one higher. The four-review average has almost no statistical weight behind it.

A review count does not measure quality either. It reflects how long a business has been collecting public feedback, how consistently it asks customers to leave reviews, and the volume of business it handles. A small crew that completes 20 driveways a year and never asks for a review will stay near the median indefinitely, no matter how well it pours concrete. Sorting businesses by review count arranges them by marketing effort and business age, which is a legitimate factor to examine as long as it is called what it is.

Among the 770 listings, 67 lack both a star rating and a review count on their map profile. Those records still include a business name and a category, usually an area, and, in most cases, a phone number. They have no public feedback signal of any kind.

A missing rating is not a low rating. A newly registered business, a company that receives all work by referral, and a company whose customers never submit reviews all produce the same empty field. A ranking based only on stars would need to either drop those 67 records or push them to the bottom, and either choice distorts data that actually says nothing. The score handles these cases differently: it replaces the missing rating with the market baseline in the review pillar and scores the rest of the record's contents.

The remaining fields show the same meaningful unevenness for the same reason. Of the 770 records, 757 have a phone number and 701 have a website; however, only 671 have a street address and 727 have map coordinates. No record among the 770 includes an email address. A directory can either report those gaps or hide them behind a ranked list that appears complete.

The mixed nature of the field is clear from the label Google Maps gives each business, which is separate from the categories this site publishes. Concrete contractor is the label on 254 of the listings. Waterproofing service is the label on 115, construction company on 108, flooring contractor on 86, paving contractor on 36, masonry contractor on 31, general contractor on 31, plain contractor on 20, landscaper on 15 and floor refinishing service on 14.

These companies are not bidding for the same work. A waterproofing service that injects basement cracks and a flooring contractor that grinds and coats garage slabs both appear in a concrete directory. Both may have a 5.0 rating, but neither one is an alternative to the other for a homeowner replacing a driveway. If you rank one against the other, the list simply puts first the businesses with the most reviews in whichever sub-trade accumulates reviews fastest.

The site's eight internal categories show the same problem. Concrete Contractors contains 559 of the 770 records, which makes it more a default than a category. Foundations and Underpinning has 155, Concrete Patios and Walkways 154, Concrete Driveways 140, Polished Concrete and Epoxy Floors 134, Concrete Repair and Leveling 93, Stamped and Decorative Concrete 29 and Garage Pads and Slabs 10. A business can appear in several of these groups at once, so a ranked list of "concrete contractors" would mix specialists and generalists without telling them apart.

The index contains the facts each record actually holds: name, area, category set, rating and review count when a public profile has one, and contact details when available. Beyond those fields, the index includes exactly one derived number, the ConcreteFinder Score, and nothing else: no badge, no tier, no editorial recommendation, and no paid position.

The score is a deterministic number out of 100, calculated from three components. Forty-five points are awarded for reviews and reputation, based on a Bayesian-adjusted rating rather than the raw star average, plus review volume. Thirty points reflect what a company states on its own website, and each point includes the sentence from which it was read. Twenty-five points assess the condition of the record: whether the website answered when it was checked, whether photographs, a phone number and hours are on file, and whether the company states its own service areas. The formula and every constant in it appear on the methodology page.

That split is why a star rating by itself cannot decide the rank here, but the score can. The rating contributes no more than 30 of the 100 points, and the system pulls it toward the dataset mean before counting it, so a 5.0 from four reviews does not place above a 4.8 from three hundred. The remaining 70 points come from evidence that is not squeezed against a ceiling. The 67 records with no rating are not removed either: for them, the review pillar falls back to the market baseline, and the other two pillars score whatever the website and the record actually show.

The middle pillar presents only what a company has said about itself. No registry was contacted, so when a company says it is licensed and insured, that statement is recorded as a claim, not as a verification. Before signing, confirm licence, insurance, and WSIB coverage directly with any contractor.

Geography follows the same publication format. Across 41 neighbourhoods, 11 cities and seven districts, the site covers 59 areas. Mississauga has 28 companies, with all 28 rated at an average of 4.87; Agincourt has 28, with 25 rated at 4.73; and Mimico has 24, with all 24 rated at 4.95. Ten of the 59 areas hold no company at all, among them Cabbagetown, Kensington Market, Riverdale and The Annex. It is more useful to report an empty area as empty than to fill it with the closest business and call that local coverage.

The site names the source next to each figure instead of publishing one house estimate. According to Nice Guy Concrete's 2026 Ontario guide, a broom-finish concrete driveway costs $9 to $15 per square foot installed, all in. Toronto Concrete Repairs lists a standard broom-finish double driveway at $6,000 to $7,500 and exposed aggregate at $8,000 to $10,000. The 2026 Canadian guide from Green Building Canada sets the GTA broom-finish range at $12 to $20 per square foot. The three sources conflict, and the driveway cost page reports that conflict directly instead of averaging the numbers.

The index records neither the number of jobs completed by any company, nor how long that company has been in business, nor its employee count, nor what it charges. It contains no warranty terms, no licence status and no insurance status. No one contacted a company to verify any of its contents, and no one interviewed anyone for any part of it.

A "best contractor" list can be credible only if it includes at least some of that. Unless you have completion data, you cannot tell whether a 5.0 rating covers three jobs or three hundred. To know which company can legally pull a permit and which cannot, you need licence status. You need warranty terms to compare what happens when a slab cracks in its second winter. A list made only from stars and review counts would look as though it had that research behind it when it did not, and a reader could not tell the difference.