40 contractor website statistics. Every one of them ours.

is how long the median contractor website takes to show its main content on a throttled mobile connection. Google’s threshold for a good experience is 2.5 seconds.

A separate, earlier run of 51 US contractor sites; 50 returned a Lighthouse result.

of contractor websites are slower than Google’s 2.5-second threshold. Not a tail of outliers — close to the whole sample.

A separate, earlier run of 51 US contractor sites; 50 returned a Lighthouse result.

is the median Lighthouse mobile performance score for a US contractor homepage.

A separate, earlier run of 51 US contractor sites; 50 returned a Lighthouse result.

is the median page weight of a contractor homepage.

A separate, earlier run of 51 US contractor sites; 50 returned a Lighthouse result.

of contractor websites get too little traffic for Google to have measured their real-world speed at all. Only 7 of the 50 we tested had any field data.

A separate, earlier run of 51 US contractor sites; 50 returned a Lighthouse result.

of contractors do not appear anywhere in the first fifty organic results for their trade plus their own town — the most obvious thing a homeowner could type.

236 independents, one head-term search each, depth 50, one day.

of contractors appear nowhere at all for their trade plus their own town — not in the first fifty organic results, and not in the map pack either.

124 of 236 independents, one search each, depth 50.

of trade-plus-town searches show a map pack above the organic results. It is effectively always there, sitting above everything else on the page.

A pack appeared on 231 of 236 searches.

of contractors are not in their own town’s map pack. The most valuable position on the page is nearly universal and almost universally not theirs.

236 independents, one search each, one day.

of organic top-ten slots go to local contractors, to directories and lead marketplaces, and to social platforms respectively. Everything else takes 7%.

2,359 top-ten slots across 236 searches.

distinct local contractor domains hold a top-ten slot across the towns we checked — a median of five slots in ten — and not one of the 236 searches came back without a local business on page one.

2,359 top-ten slots across 236 searches.

of contractors already ranking in their own town have no pages for the services they offer in the places they serve. Only 14% have a real matrix of any size.

1,838 independent local contractors crawled, all already in their town’s first fifty.

is the median number of Google searches the best-built quarter of contractor sites ranks for, against the worst-built quarter. Roughly seven and a half times as many.

1,838 sites, keyword counts bought the same week. Build score excludes links.

is the median number of searches a contractor site with thirty or more service-and-location pages ranks for, against one with no place pages at all. Monotonic at every step between.

1,838 sites grouped by their own page architecture.

more searches is what the better-built half of contractors is found for — and the gain holds inside every quartile of link authority (1.8×, 1.8×, 1.7×, 1.9×), so it is not simply a marker for a bigger firm.

1,742 sites with both a crawl and an authority score.

is how often the better-built site outranks a rival in the same result set, against how often the better-linked one does. The first is chance at any gap size; the second rises to 61% at the widest gaps.

6,751 and 6,648 within-SERP pairs across 218 searches.

of service-and-location pages return their own site anywhere in the top twenty for the search the page was literally built for. 17% reach the top ten, 10% the top three.

219 searches derived from the pages’ own URLs, depth 20, one day.

sites that do rank for the search their page was built for rank with that page. The other 41 rank with something else, and 19 of those are the home page. A wide interval on 55 sites, and the direction is the useful part.

55 sites ranking in the top twenty for a derived search, of 219 measured.

of the websites listed against US contractor businesses served a readable page to an ordinary request. 13% sat behind bot protection, 12% did not resolve, and the rest were parked, timed out or returned an error.

500 listed domains, requested once each with an identified user agent. The one figure on this page that uses 500.

of contractor websites publish no LocalBusiness schema, and 16% publish no structured data of any kind. For a business whose entire search case is that it is local, the machine-readable statement of where it is is missing from half the market.

334 readable sites, JSON-LD and microdata parsed from served HTML.

of contractor websites serve duplicate title tags across their pages, and 34% have at least one page with no H1 at all. These are the first two things any audit checks.

334 readable sites, up to six pages each.

of contractor websites have no sitemap — while every single site in the sample was on HTTPS — the one thing this market has universally fixed, because hosts and browsers forced it.

334 readable sites.

of contractor websites still load jQuery — a library whose job was to paper over browser differences that stopped existing years ago. It is the clearest single marker of a site built once and never revisited.

334 readable sites, script tags and inline usage in served HTML.

of contractor websites carry fewer than five blog posts, and most carry none. The blog exists on a great many of them — themes ship with one — but nothing has ever been put in it.

334 readable sites, counted from their own sitemaps and internal links.

of contractor websites have no real service-and-location architecture, and half the sample — 167 sites — has not got a single such page.

334 readable sites. 64% have fewer than three; 50% have none.

of contractor websites have both halves of a search programme: a built-out service-and-location architecture and somebody still publishing into it. Fifteen sites of 334, and two of those fifteen turned out to be national brands.

334 readable sites. Thresholds deliberately low: 30+ service-and-place pages and 5+ posts inside six months.

of contractor sites that did build thirty or more service-and-location pages have published fewer than five posts. They built the machine and stopped running it: all of the cost of SEO and none of the compounding.

The 49 sites carrying 30 or more service-and-place pages; 34 of them publish nothing.

contractor sites built on a hosted website builder — Wix, GoDaddy, Squarespace, Duda, Webflow, Weebly — reach a built-out architecture with content still running. Not one. On WordPress, 3.9% do.

The 82 hosted-builder sites among 334 readable ones. Causation not established: some of the gap is operators migrating off these platforms.

of the contractor location pages we could judge were templated doorway pages — the same page with the town swapped. The lazy version of local SEO is rarer than its reputation.

The 84 sites with three or more comparable sibling pages carrying enough prose to judge.

of contractors whose website would not load at all still have a live, claimed Google Business Profile carrying that same dead domain. The website has been abandoned and the profile is still trading.

The 166 sampled businesses that did not serve readable markup, matched against listings carrying the same domain.

of contractor websites give an arriving visitor nothing to act on above the fold: no phone number in the header and no form you can see without scrolling.

244 of 334 readable sites. Detection reads served HTML, so a number inside an image counts as absent — a floor.

of contractor websites have no phone number in the header, and 19% have no tappable tel: link anywhere on the homepage — which on a phone means the number has to be read, remembered and typed.

334 readable sites, measured from served HTML.

of contractor websites show a form without scrolling. 47% have no form on the homepage at all, and 48% have one below the fold.

334 readable sites; 17 of them show a form above the fold.

of contractor websites tell a visitor when they will hear back — seven sites out of 334. The single most reassuring thing a contractor can say to somebody hovering over a form is said by almost none of them.

334 readable sites. Explicit commitments only — “fast, friendly service” does not count.

of contractor websites offer online booking and financing respectively, in trades where the job is scheduled and often four figures.

334 readable sites. Counted only where the vendor’s own code is present.

of contractor websites embed a Google map and show before-and-after imagery — in trades whose entire proposition is visible, local, finished work. Two thirds (66%) display a licence number.

334 readable sites.

of reviews on contractor Google profiles carry a reply from the owner, and the median business answers 88% of them. Review response is not the weakness in this market — an audit that leads with it leads with something the prospect is already doing.

The 50 most recent reviews for each of 179 independents whose Google profile carries their website — the better-organised third of the sample, not contractors generally.

of one- and two-star reviews on contractor Google profiles have no reply at all. The easy half gets done: among businesses carrying three or more complaints, the median answers 84% of its good reviews and 38% of its bad ones.

409 one- and two-star reviews inside the 50 most recent for each of 179 independents.

of contractors have no owner reply anywhere in their ten most recent Google reviews, and for 8% that is a lapse rather than a habit — they answered before and stopped.

The 10 most recent reviews for each of 179 independents, within their 50 most recent.

of contractor websites carry no analytics at all, so they cannot tell where a lead came from — which is the precondition for everything else: a business that cannot measure search has no reason to prefer it over paid.

334 readable sites. 62% carry some analytics; 14% carry a Meta pixel and 14% Google Ads conversion tracking.