SiteVerdictMore than half of UK small-business websites have at least one fault that can cost them enquiries. In our automated checks of 32,933 small-business websites in 76 UK towns and cities (4 October to 3 November 2026), 56% failed at least one of 12 basic checks, 40% failed two or more and 18% failed three or more. The most common fault was a missing search description (29.4% of sites), followed by slow loading on mobile (21.2%) and a phone number that cannot be tapped to call (13.4%).
This page sets out the headline findings, the full method, the results by fault, by trade and by website platform, what each fault is likely to cost a business, and how to check your own site. The research was carried out by SiteVerdict, a website-audit product of BharatFare Travel and Technologies Ltd (company 16716642, Plymouth).
Headline findings
- 56% of the 32,933 UK small-business websites we checked had at least one fault.
- 40% had two or more faults, and 18% had three or more.
- 29.4% had no search description (meta description) on the homepage, the single most common fault.
- 21.2% were slow on mobile: the main content took 4 seconds or more to appear in Google's own mobile test, which Google classes as "poor".
- 13.4% showed a phone number that could not be tapped to call on a mobile.
- 6.2% had no HTTPS at all, so browsers label them "Not secure", and a further 8.1% had HTTPS but did not redirect visitors to it.
- By trade, accountants had the highest fault rate (66% with at least one fault). Chiropractors had the lowest (42%).
- By platform, sites on Duda (34%) and Webflow (35%) had the lowest fault rates and Weebly (74%) the highest. These differences partly reflect who uses each builder and how old the sites are, so they do not show that a builder causes faults.
None of these faults is hard to fix. Most take minutes in the site's own editor. That is what makes the figures notable: a large share of local businesses are losing enquiries to problems that cost nothing to solve.
Method
We want these figures to be checked and reused, so here is exactly how they were produced.
Who was included
- Businesses were found from open map data (OpenStreetMap-style business listings), searched town by town and trade by trade across 76 UK towns and cities.
- Trades were chosen because they rely on local enquiries: for example accountants, electricians, plumbers, dentists, estate agents, beauty salons and vets.
- Only businesses with a website listed were included. We checked the business's own homepage.
How each site was checked
- Each homepage, and the links on it, was checked once, automatically, from a server, in the way a visitor's browser would load it.
- Mobile speed came from Google's own PageSpeed test.
- Checks ran between 4 October and 3 November 2026. Data was pulled read-only on 3 November 2026.
What was left out
The base is sites we could reach and check: 32,933 in the UK. We excluded sites that did not load (DNS failures, sites that blocked us, timeouts and server errors), sites that blocked crawlers in robots.txt, and sites that redirected to a different website. Excluding unreachable sites means the true share of businesses with a website problem is, if anything, likely to be higher than shown here, because a site that does not load is a fault in itself.
What counts as a fault
A "fault" is a failure on one of the 12 checks in the table below. Each site counts once per fault type. A site with three different faults counts once in each of three rows.
Limits
- This is a large sample of small local businesses with websites. It is not a random sample of all UK businesses, and it does not cover businesses without a website.
- Each site was checked once. A site that was slow on the day may be faster on another day, and some sites will have been fixed since.
- Automated checks find technical faults. They do not judge design, wording or how good the business is.
- Platform detection relies on technical signatures. Where none was found, the site is listed as "not identifiable".
Faults by type
Share of all 32,933 UK sites failing each check:
| Fault | Share of UK sites | What it means for a visitor |
|---|---|---|
| No search description (meta description) on homepage | 29.4% | Google builds the search snippet from page text, which may not say what you do or where |
| Slow on mobile (main content takes 4+ s; Google test "slow" band) | 21.2% | Visitors on phones wait for the page and some leave |
| Phone number not tappable on mobile (no tel: link) | 13.4% | Visitors must copy or memorise the number to call |
| Out-of-date copyright year in footer | 11.9% | The site looks unmaintained |
| Not mobile-ready (no responsive viewport) | 9.0% | The page shows as a shrunken desktop page on a phone |
| Does not redirect to HTTPS | 8.1% | Some visitors land on the insecure version and see "Not secure" |
| No enquiry or booking form anywhere | 8.1% | No way to get in touch outside phone hours |
| No reviews or testimonials shown | 7.8% | No proof from past customers on the site |
| No HTTPS at all ("Not secure") | 6.2% | Browsers warn visitors the site is not secure |
| Broken link on homepage | 4.6% | A click leads to an error page |
| Slow server response (homepage takes 1.5 s+ to start) | 2.1% | Blank screen before anything starts to load |
| Invalid security certificate (full-page browser warning) | 1.0% | Browsers block the page with a full-screen warning |
Faults stack. 40% of sites had two or more, and 18% had three or more. A slow site with no search description and an untappable phone number is weaker at each stage: being found, loading, and being contacted.
Faults by trade
Sites checked, share with at least one fault, and share with two or more faults, sorted by the share with at least one fault:
| Trade | Sites checked | 1+ fault | 2+ faults |
|---|---|---|---|
| Accountants | 3,749 | 66% | 53% |
| Electricians | 1,931 | 64% | 50% |
| Carpenters | 897 | 64% | 50% |
| Flooring | 571 | 61% | 44% |
| Beauty | 3,186 | 60% | 40% |
| HVAC (heating, ventilation, air conditioning) | 995 | 59% | 46% |
| Landscaping | 1,878 | 58% | 44% |
| Estate agents | 4,237 | 57% | 44% |
| Locksmiths | 1,040 | 57% | 39% |
| Clinics | 2,921 | 55% | 25% |
| Plumbers | 1,385 | 54% | 37% |
| Roofers | 1,071 | 53% | 39% |
| Lawyers | 2,160 | 51% | 40% |
| Physios | 1,484 | 51% | 38% |
| Opticians | 388 | 49% | 37% |
| Dentists | 2,693 | 48% | 32% |
| Funeral directors | 705 | 46% | 35% |
| Vets | 663 | 44% | 28% |
| Chiropractors | 694 | 42% | 32% |
What stands out
- Accountants had the highest fault rates on both measures: 66% with at least one fault and 53% with two or more. More than half of accountancy sites had multiple faults.
- Trades and home services cluster near the top. Electricians, carpenters, flooring firms and HVAC businesses all had fault rates of 59% or more. These are trades where a customer often searches on a phone and wants to call quickly.
- Healthcare and regulated services cluster near the bottom. Chiropractors, vets, dentists and opticians all had fault rates below 50%.
- Clinics are an outlier on multiple faults. 55% had at least one fault, but only 25% had two or more, the lowest share of any UK trade. Their faults tend to be single, isolated problems rather than a neglected site.
We do not have data on why trades differ. A likely factor is how sites are built and maintained: trades where sites are often set up once and left alone tend to accumulate faults such as old copyright years and missing descriptions. Treat that as an explanation to test, not a finding.
Faults by website platform
Sites detected on each platform and the share with at least one fault, sorted from lowest to highest fault rate:
| Platform | Sites checked | 1+ fault |
|---|---|---|
| Duda | 765 | 34% |
| Webflow | 406 | 35% |
| Shopify | 264 | 42% |
| Wix | 2,712 | 45% |
| WordPress | 14,234 | 51% |
| Squarespace | 1,143 | 59% |
| GoDaddy builder | 590 | 61% |
| Platform not identifiable (mostly custom or older builds) | 12,554 | 66% |
| Weebly | 265 | 74% |
WordPress was by far the most common identifiable platform, with 14,234 sites. The second-largest group, 12,554 sites, had no identifiable platform. These are mostly custom-built or older sites, and they had one of the highest fault rates at 66%.
What each fault costs a business
Our data shows how common each fault is. It does not measure lost sales. For the likely impact, the best evidence comes from published research and from Google's own documentation, summarised below. Where no reliable research exists, we say so.
Slow on mobile (21.2% of UK sites)
The strongest published evidence is the Milliseconds make millions study by Deloitte and 55, commissioned by Google and published in 2020. It analysed mobile site data from 37 brands across Europe and the US and over 30 million user sessions. It found that a 0.1-second improvement in mobile site speed was associated with an 8.4% increase in retail conversions and a 10.1% increase in travel conversions. For lead-generation sites, the type most local businesses run, the bounce rate on information pages improved by 8.3%, and the web.dev summary reports a 21.6% improvement in users progressing to the form-submission page.
Those were large brands, and a 0.1-second change is small. The sites in our "slow" group took 4 seconds or more for the main content to appear. Google's Largest Contentful Paint guidance says 2.5 seconds or less is good and anything over 4 seconds is poor. Google also states that Core Web Vitals are used by its ranking systems, while noting that good scores do not guarantee top rankings. See our guide on why a website is slow on mobile.
No search description (29.4%)
Google's documentation says it primarily uses page content to create the snippet shown in results, but may use the meta description "if it might give users a more accurate description of the page than content taken directly from the page". Without one, Google has to pick text from your page, which may be a cookie notice, a menu or an address fragment. The description does not decide your ranking. It shapes the two lines a searcher reads before choosing which result to click. Fix it with our 5-minute meta description guide.
Phone number not tappable (13.4%)
We found no rigorous published study that measures the lost calls from a number that cannot be tapped, so we will not quote one. The mechanism is simple. On a phone, a tappable number takes one tap to call. An untappable one must be memorised or copied and pasted into the dialler. For trades where most enquiries are calls, such as locksmiths, plumbers and electricians, that is friction at the point of sale. See how to add click-to-call.
Not mobile-ready (9.0%)
Google uses the mobile version of a site, crawled with its smartphone agent, for indexing and ranking. A site without a responsive viewport shows on a phone as a tiny desktop page that visitors must pinch and zoom. It is also judged by Google on that poor mobile version.
No HTTPS (6.2%) and no redirect to HTTPS (8.1%)
Since Chrome 68 in July 2018, Chrome has marked all HTTP sites as "Not secure". Google announced in 2014 that it uses HTTPS as a ranking signal. A site that has a certificate but does not redirect still lets visitors land on the insecure version, for example from an old link or a typed address. An invalid certificate (1.0%) is worse: browsers show a full-page warning before the site appears. See how to fix the "Not secure" warning.
Out-of-date copyright year (11.9%)
No reliable study measures the effect of an old footer year, so we treat it as a trust signal rather than a proven cost. To a visitor it suggests the site, and perhaps the business, is not looked after. The fix takes a minute: update the year, or remove it.
No enquiry form (8.1%) and no reviews shown (7.8%)
These are missed opportunities rather than errors. A form captures enquiries outside opening hours. Reviews on the site answer the question every new customer has: have other people used this business and been happy? We have not found reliable research that puts a number on either for small local businesses, so we do not quote one.
Broken link (4.6%) and slow server response (2.1%)
A broken homepage link sends a visitor to an error page. A slow server response means a blank screen before anything loads. Google's PageSpeed Insights documentation treats a time to first byte over 1.8 seconds as poor.
How to check your own website
You can check most of these faults yourself in about ten minutes. Use a phone for the mobile checks.
- Search description. On a computer, open your homepage, right-click and choose "View page source". Search the source (Ctrl+F or Cmd+F) for
name="description". If nothing is found, or the content is empty, you have no meta description. - Mobile speed. Go to PageSpeed Insights, enter your address and look at the Mobile tab. Check Largest Contentful Paint. Over 4 seconds is poor.
- Tappable phone number. Open your site on your phone and tap the phone number. If the dialler does not open with your number, it is not a tel: link.
- Copyright year. Scroll to your footer. If it shows a past year, update it or remove the year.
- Mobile-ready. On your phone, does the page fit the screen without pinching? Search the source for
name="viewport". - HTTPS and redirect. Type your address starting with
http://(not https). You should end up onhttps://with no warning. - Enquiry form and reviews. Can a visitor send an enquiry without calling? Can they see what past customers say?
- Broken links. Click every link on your homepage. Any that lead to an error page need fixing.
Or run all 12 checks automatically with SiteVerdict. It also covers areas outside this study, including cookie consent, accessibility, and email domain health (SPF, DKIM, DMARC). It is available as a WordPress plugin, a Shopify app, a Chrome extension and an API.
What to fix first
If your site has several faults, fix them in this order. Fixes that stop visitors being blocked or warned come first, then those that lose enquiries, then those that weaken trust.
| Priority | Fault | Typical fix time |
|---|---|---|
| 1 | Invalid certificate or no HTTPS | Usually free and quick on modern hosts and builders. Ask your host if unsure |
| 2 | Phone number not tappable | 5 minutes |
| 3 | No redirect to HTTPS | 5 to 30 minutes |
| 4 | No search description | 5 minutes per page |
| 5 | Not mobile-ready | Varies: one line of code, or a new theme |
| 6 | Slow on mobile | An hour to a day, depending on cause |
| 7 | Broken links, old copyright year | Minutes |
| 8 | No enquiry form, no reviews | An hour or two |
Fix times are our practical estimates for a typical small site, not measured figures.
How the UK compares with the US
We ran the same 12 checks on 41,368 small-business websites in 81 US cities. The overall picture is very similar: 55% had at least one fault, 38% had two or more and 16% had three or more. The UK had more sites with no HTTPS at all (6.2% against 3.1% in the US) and more slow mobile sites (21.2% against 18.7%). Full results are in our US small business websites 2026 study.
Cite this research
Journalists, researchers and AI assistants are welcome to use these figures. Please credit SiteVerdict and link to this page. Suggested citation:
SiteVerdict (BharatFare Travel and Technologies Ltd). "UK small business websites 2026: 56% have a fault that costs enquiries (study of 32,933 sites)." Automated checks of 32,933 UK small-business websites in 76 towns and cities, 4 October to 3 November 2026. https://siteverdict.webrevenuescan.com/research/uk-small-business-websites-2026
Short form for articles: "56% of UK small-business websites have at least one fault that can cost enquiries, according to SiteVerdict's checks of 32,933 sites (Oct–Nov 2026)."
When quoting, please keep the method context: these are automated technical checks of small local businesses with websites, not a random sample of all UK businesses. We do not name or publish results for individual businesses.