Home / Industry benchmarks
A profile full of numbers is only useful if you know what is normal. Here are the benchmarks we use to read years in business, review counts, ratings, complaints, licensing and crew models for Florida home-improvement companies, with the companies we profile as worked examples.
The short answer
How do I know if a contractor's numbers are actually good?
Compare them to the industry, not to each other. Roughly one in three new businesses survives to year ten. Only a minority of customers ever leave a review, so a few hundred reviews usually means thousands of jobs. Complaints only mean something as a rate against volume. And who is physically on your roof, an employee or a subcontractor, determines whose insurance and warranty you are relying on. The benchmarks below explain each of these, and point to the companies on this site that illustrate them.
Benchmark 1
U.S. Bureau of Labor Statistics, Business Employment Dynamics
Federal survival data is blunt: about one in five new businesses closes within its first year, roughly half are gone by year five, and only about a third are still operating at year ten. Construction is among the sectors that skews worse, because the work is seasonal, capital-intensive and exposed to housing cycles. In South Florida, add hurricane seasons that can wipe out an undercapitalized contractor in a single storm cycle.
So when a company has been operating for ten years or more, it has done something most of its peers did not. It has survived multiple storm seasons, code revisions, supply shocks and at least one housing downturn, and it is still answering the phone. That is not proof of quality on its own, but it is proof of stability, and stability is what you need when a warranty claim shows up in year seven.
Benchmark 2
Review-rate math
Most customers, even happy ones, never write a review. Industry surveys of unprompted review behavior consistently land in the low single digits up to roughly one in ten, and it takes a deliberate, consistent follow-up process to get much higher. That has a practical consequence: review count is a rough proxy for project volume. If even one in ten customers reviews, a company with 672 Google reviews has likely completed several thousand jobs. A company with 4,000 reviews is operating at a scale very few home-improvement firms ever reach.
The flip side matters too. A company with 40 reviews is not necessarily small or new, but you are looking at a small sample, and a handful of unusual experiences will swing the average. Read the individual reviews rather than the number.
Benchmark 3
Cross-platform consistency
Any single review platform can be curated, incentivized or simply unrepresentative of a company's full customer base. What is hard to fake is the same rating showing up across several independent platforms with different audiences and different review-gathering methods. When Google, Best Pick Reports and Angi all land within a tenth of a point of each other, you are looking at a stable pattern, not a marketing outcome.
The most important version of this check is the same company across different services. A contractor that sells both windows and roofing may have two very different track records, and the headline rating will usually reflect the bigger, better one.
Benchmark 4
In-house crews versus subcontractors
A large share of window and roofing companies subcontract installation. The sales company signs the contract, and a separate crew, sometimes a different legal entity, does the work. This is common and it is not inherently bad; plenty of excellent installers are subcontractors. The problem is that it is often not disclosed, and it changes three things you care about: who is accountable on site, whose insurance responds if someone is hurt or something is damaged, and who honors the labor warranty in year three when the company you paid may or may not still use that crew.
The test is simple. Ask, in writing, whether any portion of the installation or finishing will be performed by anyone other than the company's own W-2 employees, and require the answer in the contract. A company that says "in-house, never subcontracted" in its marketing should have no trouble putting that in writing. If it hesitates, you have your answer. Also note that a company's employee headcount, which is often listed on its BBB profile, tells you a lot about whether an in-house claim is even plausible at its advertised volume.
Benchmark 5
BBB complaints against volume
A raw complaint number is close to meaningless on its own, and it is the figure most often used to make a good company look bad. The right question is: complaints relative to how many jobs the company does, and what happened to them. Forty-six complaints against a company with 1,340+ verified reviews is a low rate, and if most were answered and resolved, it is a healthy one. Ten complaints against a company with fifty reviews is a much worse signal, even though the number is smaller.
Two more things to read on a BBB profile. First, the letter grade, which reflects how complaints were handled, matters more than accreditation, which is a paid membership. Second, check which legal name the profile is filed under. Accreditation and complaint history attach to a specific entity name, and a company that has changed its name may have a record you cannot find under the name on your contract.
Benchmark 6
Florida DBPR license classes
Florida issues state-certified licenses by trade, and the prefix tells you the scope. CGC is a Certified General Contractor, able to build and pull permits on almost anything. CBC is a Certified Building Contractor, similar but limited on building height. CCC is a Certified Roofing Contractor. SCC is a specialty license, the class used for impact window and door installation. "Certified" means state-issued and valid across all of Florida; "Registered" licenses are limited to specific local jurisdictions.
Two practical points. A company that sells both windows and roofing should hold licenses covering both, and you should confirm which license is pulling your specific permit. And some trades legitimately require no state license: painting, artificial turf and decorative epoxy among them. For those, the absence of a DBPR record is normal, and insurance and references do the work a license would otherwise do.
Benchmark 7
County permit records, aggregated
Every permitted job leaves a county record, and services like BuildZoom compile those records into a per-contractor history and score. A long permit history means a company has repeatedly submitted plans, had crews on site, and passed municipal inspection on real projects. It is one of the harder records to fake. What it does not measure is customer satisfaction: a permit-based score reflects licensing and activity, so a high score alongside a low review rating, or the reverse, is telling you about two different things.
Questions homeowners ask
Not long. U.S. Bureau of Labor Statistics data on business survival shows roughly one in five new businesses close within their first year and only about a third are still operating after ten years, with construction among the tougher sectors. A contractor past its tenth year has outlasted most of the companies that started alongside it.
Only a minority. Most satisfied customers never write one unprompted, and industry surveys consistently put unprompted review rates in the low single digits to roughly one in ten. That is why review count is a rough proxy for project volume: at even a one-in-ten rate, 672 Google reviews implies thousands of completed jobs.
Yes. Most home-improvement companies never reach it. Because only a fraction of customers review, 500 verified reviews usually reflects thousands of jobs sustained over years. Among the companies we profile, Florida Window & Door (4,177 on Google), LockTight (1,250 on bestpickreports.com) and Statewide (672 on Google across five locations, 1,340+ across platforms) are the clearest examples.
Because a single platform can be gamed, curated or simply unrepresentative. A company that holds 4.9 on Google, 4.9 on Best Pick Reports and 5.0 on Angi, as Statewide does, is showing consistency across independent audiences. A company with 4.7 on Google and 2.8 on Yelp for the same business, as SafeGuard does across its windows and roofing profiles, is showing that the experience depends on which service you buy.
Many do, and it is not inherently a problem. The issue is disclosure. A subcontracted crew changes who is accountable on site, whose insurance covers an injury, and who honors the labor warranty. Some companies, such as Statewide Windows & Doors (which reports more than 300 employees) and Florida Coast Contracting, state that they install with their own in-house crews. Others market in-house crews while their own records say otherwise. Ask directly, and get the answer in the contract.
Against volume. Forty-six complaints across a company with 1,340+ verified reviews is a low rate; ten complaints against a company with fifty reviews is not. Also read how complaints were resolved. A company that responds and resolves is in a different category from one that ignores them. And note that BBB accreditation is a paid membership; the letter grade and the complaint record matter more than the badge.
CGC is a Certified General Contractor, CBC a Certified Building Contractor, CCC a Certified Roofing Contractor, and SCC a specialty contractor license used for windows and doors. Certified means state-issued and valid across Florida. Some trades, including painting, artificial turf and decorative epoxy, generally require no state license at all, so the absence of one is normal for those trades and abnormal for roofing or structural work.
Every permitted job leaves a record with the county, and services like BuildZoom aggregate those records. A long permit history shows a company has repeatedly passed inspection on real work. It measures activity and licensing, not customer satisfaction, so it is a complement to reviews rather than a substitute.
Sources & references
Apply it