We checked 408 small business websites across 20 industries and 25 UK towns and cities, ran Google’s Lighthouse speed test on 391 of them, and read 621 public posts in which UK small business owners talk about Google, search engine optimisation (SEO) and getting found. This report sets out what we found, what it means for your business, and exactly how we did it.
About this report
- 408 websites checked
- 20 industries
- 25 towns and cities
- 168 owner posts analysed
Every number in this report is labelled. Our data means we collected it ourselves in October 2026, and you can download it below. Anything else links to its original source. You can download the website data as a CSV (comma-separated values) spreadsheet and the full report as a PDF (Portable Document Format).
Key findings
These are the strongest findings from our own data. Each one is written so you can quote it, with a link back to this page:
- 52%
52% of the UK small business websites we checked were missing at least one of the three things a customer needs to get in touch: a tap-to-call phone number, a full address with postcode, or opening hours.
Our data - 20%
Around 2 in 10 small business homepages had no tap-to-call phone link, so a customer on a phone has to copy the number by hand.
Our data - 47
The typical small business website scored 47 out of 100 on Google's Lighthouse mobile speed test, and 54% scored in the "poor" band below 50.
Our data - 7.2s
On a simulated mobile connection, the main content of the typical site took 7.2 seconds to appear, against the 2.5 seconds Google calls good. Only 5% of sites were within it.
Our data - 53%
53% of small business websites had no LocalBusiness structured data, the code that tells Google and artificial intelligence (AI) tools a business's name, address, phone number and hours in a form machines read reliably.
Our data - 78%
78% of websites had no pages for the areas they serve beyond their own town, such as an "areas we cover" page or a page for each nearby town.
Our data - 39%
Trades and home services were the least likely to show all three contact basics: 39% did, against 59% of professional services firms.
Our data - 52%
Online booking is common on health, beauty and pet care websites (52%) but rare on trades and home services websites (3%).
Our data - 31%
31% of websites showed no reviews or testimonials on the homepage and did not link to a reviews page.
Our data - 51%
In 168 public forum posts by UK small business owners about Google and getting found, 51% were owners trying to do SEO themselves, and 10% described a bad experience with an SEO provider, web designer or cold pitch.
Our data
Why this matters for small businesses
Small businesses are most of the UK economy. The Department for Business and Trade counted 5.7 million private sector businesses at the start of 2025, and 99.2% of them had fewer than 50 employees. About three in four had no employees beyond the owners. Source: Department for Business and Trade, October 2025
How people find those businesses is changing. Ofcom’s 2026 study of UK adults found that 54% now use AI tools such as ChatGPT, Copilot or Gemini, up from 31% a year earlier, and that 75% of online adults read AI-generated summaries in search results at least sometimes. Source: Ofcom, April 2026 Google says its AI Mode passed 1 billion monthly users in 2026 and that it sends “billions of clicks to websites every week” through AI features in Search. Source: Google, July 2026
Google says local results are based on relevance, distance and prominence, and that complete, accurate business information helps. Source: Google Business Profile Help Your website is one of the places that information comes from, which is why we looked at what small business websites actually show.
Can customers get in touch from your website?
We looked for three basics a customer needs before they call or visit. Across all 408 sites, 48% showed all three and 52% were missing at least one. Our data
Our data: 408 UK small business websites checked in October 2026.
Tap-to-call matters for anyone searching on a phone. A number written as plain text, or inside an image, means the customer has to copy it, and some will simply call the next business instead. 80% of homepages had a tap-to-call link.
Addresses tell a more mixed story. Trades and home services showed a full address on 70% of sites, against 92% for professional services. Some tradespeople work from home and choose not to publish an address, which Google allows for service-area businesses, so a missing address is not always a mistake. Opening hours are different: 40% of trades sites gave none, and customers use them to decide whether to ring now or try someone else.
Our data: 408 UK small business websites checked in October 2026.
Are small business websites fast on a phone?
Almost every site was set up to fit a phone screen (95% had a mobile viewport setting) and 99.5% used a secure connection, known as HTTPS (Hypertext Transfer Protocol Secure). Our data
We ran Lighthouse, the open-source test behind Google’s PageSpeed Insights, on each homepage with its standard mobile settings. The middle score was 47 out of 100. Lighthouse calls 90 to 100 good, 50 to 89 in need of improvement and below 50 poor. Source: Chrome for Developers: Lighthouse performance scoring
- Poor (0 to 49): 54%
- Needs improvement (50 to 89): 44%
- Good (90 to 100): 2%
Our data: Lighthouse mobile test of 391 websites, October 2026.
The largest contentful paint (LCP) is how long the main content of a page takes to appear. Google counts 2.5 seconds or less as good. Source: web.dev: Largest Contentful Paint On a simulated mobile connection, the typical homepage in our sample took 7.2 seconds, and 5% of homepages were within 2.5 seconds. Our data
We also recorded which website builder each site used, where the page showed it. Only builders with at least 20 sites in the sample are shown, because smaller groups can swing a long way on a few sites:
Our data: Lighthouse mobile test of 391 websites, October 2026.
Do websites tell Google where they work?
Most sites named their town in the homepage title (90%), which is the easy part. The deeper signals were much rarer. Our data
Our data: 408 UK small business websites checked in October 2026.
Structured data is a short piece of code in the page that states facts such as your business type, address, phone number and hours in a format search engines read reliably. Google uses LocalBusiness structured data to understand a business’s details. Source: Google Search Central: Local business structured data 47% of the sites we checked had it.
Area pages were most common among estate agents (48%) and locksmiths (45%), and least common among hair salons (4%) and vets (7%).
Reviews and booking on the website
69% of sites showed reviews or testimonials on the homepage or linked to a reviews page. The range by industry was wide, from 100% for dentists to 39% for hair salons. Our data
Reviews weigh heavily with customers. In BrightLocal’s 2026 survey of 1,002 US adults, 97% read reviews for local businesses, 68% would only use a business rated 4 stars or more, and 74% look for reviews written in the last three months. Source: BrightLocal Local Consumer Review Survey 2026 (US) We found no UK survey of the same size and date, so treat these as a guide to UK behaviour rather than a UK measure.
Online booking split the sample sharply. 16% of sites let customers book online, but the share ran from 78% for hair salons and 77% for physiotherapists down to none at all for 10 of the 20 industries we checked. Our data
Our data: 408 UK small business websites checked in October 2026.
What small business owners say about Google
To hear owners in their own words, we read 621 public posts on UK Business Forums, a public forum for UK small business owners, from January 2022 to October 2026. We kept only posts in which a business owner talks about their own business and Google, SEO, their website or getting found, which left 168 posts. We then tagged each one by theme. A post can have more than one theme. Our data
Our data: 168 posts by UK small business owners on UK Business Forums, 2022 to 2026. Posts can carry more than one theme.
The themes were clearest in the posts themselves. These are paraphrased so no one can be identified, each with a link to the public thread:
- Doing it themselves: A handyman closed an earlier website after getting no results despite keyword SEO, and wants help with keywords for the new one (thread). A small six-page website gets no organic traffic and they ask whether backlinks are essential (thread). A former plumbing business owner found local SEO easy and asks how to structure many landing pages for a new site (thread).
- Bad experiences: Hired an SEO expert who changed the site’s content and tags; after launch, sales disappeared (thread). Has paid an SEO company and website builder monthly for years with hardly any reports on the work done (thread). Their four-year-old website is very slow, and they feel conned by the person who built it (thread).
- Google Business Profile problems: Changed their address on their Google Business Profile, it stayed pending for three months, support could not help, and then the listing was suspended (thread). Could not verify their Google Business Profile after moving their registered office, because Google wanted a video of premises they don’t trade from (thread).
- Reviews: Spent an hour and a half emailing 40 customers for Google reviews and got none from actual customers (thread). A B2B firm says every Google review it has received is fake, from people who never dealt with them (thread).
- Cost: A private dental practice in London grew through Google Ads but finds them very expensive for cosmetic treatments and wants to try Instagram (thread). Pays £20 a month for a directory listing service for SEO and wonders if it is money wasted (thread).
- AI search: Has noticed Google’s AI answers listing recommended companies and wants to know how to appear (thread). Approached by a company selling AI engine optimisation and unsure whether it is real (thread).
Two things stood out. First, owners want to do this themselves: 51% of posts were owners asking how to handle SEO, their Google listing or their website on their own. Second, AI search barely featured. Only 6 of the 168 posts (4%) raised AI search or AI tools, even though Ofcom found most UK adults now use them. Owners are still talking about the basics, which is where most of the gaps in our website data are too.
Results by industry
The table shows each check by industry. With 15 to 26 sites per industry, treat small gaps between industries with caution. Our data
On a phone, swipe the table sideways to see every column.
| Industry | Sites | Tap-to-call | Address | Hours | Area pages | Structured data | Online booking | Reviews shown | Speed score |
|---|---|---|---|---|---|---|---|---|---|
| Accountants | 19 | 84% | 100% | 79% | 21% | 53% | 5% | 74% | 49 |
| Builders | 19 | 84% | 37% | 47% | 21% | 37% | 0% | 74% | 59 |
| Cleaning companies | 22 | 59% | 73% | 46% | 23% | 46% | 0% | 82% | 57 |
| Dentists | 23 | 91% | 96% | 96% | 13% | 78% | 13% | 100% | 36 |
| Dog groomers | 22 | 46% | 73% | 55% | 9% | 46% | 36% | 59% | 53 |
| Electricians | 18 | 89% | 61% | 39% | 11% | 22% | 0% | 67% | 46 |
| Estate agents | 23 | 70% | 83% | 78% | 48% | 39% | 0% | 61% | 37 |
| Florists | 23 | 74% | 74% | 70% | 9% | 57% | 0% | 52% | 46 |
| Garages | 16 | 100% | 100% | 75% | 13% | 44% | 19% | 75% | 39 |
| Gardeners and landscapers | 15 | 67% | 67% | 60% | 20% | 53% | 0% | 67% | 53 |
| Hair salons | 23 | 78% | 87% | 83% | 4% | 39% | 78% | 39% | 46 |
| Heating engineers | 19 | 95% | 68% | 63% | 32% | 68% | 5% | 68% | 50 |
| Locksmiths | 20 | 95% | 70% | 90% | 45% | 50% | 0% | 80% | 61 |
| Pest control | 21 | 86% | 76% | 52% | 38% | 38% | 5% | 76% | 40 |
| Physiotherapists | 22 | 68% | 86% | 64% | 9% | 27% | 77% | 82% | 45 |
| Plumbers | 21 | 95% | 76% | 71% | 24% | 48% | 14% | 76% | 54 |
| Removal companies | 26 | 89% | 85% | 58% | 31% | 54% | 0% | 69% | 42 |
| Roofers | 22 | 91% | 73% | 68% | 23% | 50% | 0% | 73% | 54 |
| Solicitors | 19 | 79% | 95% | 79% | 32% | 53% | 0% | 68% | 45 |
| Vets | 15 | 67% | 93% | 67% | 7% | 40% | 60% | 40% | 41 |
Speed score is the middle Lighthouse mobile performance score for that group, out of 100. Shading shows 75% or more, 50% to 74% and under 50%, and the number is always shown.
Results by town and city
This table shows the same checks for each of the 25 towns and cities. Each town has 13 to 19 sites from a mix of industries, so it is a snapshot, not a ranking of towns. Our data
On a phone, swipe the table sideways to see every column.
| Town or city | Sites | Tap-to-call | Address | Hours | Area pages | Structured data | Online booking | Reviews shown | Speed score |
|---|---|---|---|---|---|---|---|---|---|
| AberdeenScotland | 17 | 77% | 82% | 71% | 18% | 53% | 18% | 41% | 43 |
| BelfastNorthern Ireland | 18 | 78% | 89% | 56% | 11% | 44% | 6% | 67% | 42 |
| BirminghamWest Midlands | 14 | 86% | 71% | 64% | 43% | 64% | 14% | 79% | 46 |
| BrightonSouth East | 19 | 90% | 63% | 63% | 21% | 32% | 16% | 63% | 53 |
| BristolSouth West | 17 | 82% | 94% | 94% | 41% | 53% | 12% | 71% | 39 |
| CambridgeEast of England | 13 | 69% | 77% | 39% | 8% | 39% | 31% | 69% | 36 |
| CardiffWales | 17 | 82% | 71% | 82% | 35% | 47% | 12% | 82% | 33 |
| CroydonLondon | 16 | 88% | 81% | 81% | 25% | 50% | 19% | 44% | 49 |
| EdinburghScotland | 16 | 94% | 88% | 88% | 25% | 69% | 13% | 94% | 36 |
| ExeterSouth West | 15 | 67% | 73% | 53% | 7% | 27% | 20% | 73% | 42 |
| GlasgowScotland | 18 | 89% | 89% | 78% | 17% | 44% | 11% | 83% | 44 |
| LeedsYorkshire and the Humber | 16 | 75% | 69% | 50% | 25% | 25% | 19% | 75% | 40 |
| LeicesterEast Midlands | 18 | 72% | 72% | 78% | 17% | 50% | 6% | 72% | 36 |
| LincolnEast Midlands | 15 | 67% | 80% | 60% | 20% | 40% | 13% | 60% | 67 |
| ManchesterNorth West | 19 | 74% | 95% | 68% | 16% | 58% | 11% | 84% | 59 |
| Newcastle upon TyneNorth East | 16 | 69% | 63% | 75% | 31% | 38% | 19% | 69% | 56 |
| NorwichEast of England | 16 | 69% | 63% | 69% | 19% | 63% | 6% | 75% | 64 |
| NottinghamEast Midlands | 14 | 71% | 86% | 71% | 21% | 57% | 21% | 86% | 56 |
| PlymouthSouth West | 13 | 92% | 85% | 85% | 8% | 54% | 8% | 77% | 51 |
| PrestonNorth West | 15 | 73% | 80% | 60% | 20% | 47% | 33% | 53% | 45 |
| ReadingSouth East | 18 | 94% | 72% | 61% | 33% | 50% | 11% | 78% | 49 |
| SheffieldYorkshire and the Humber | 16 | 69% | 88% | 69% | 13% | 69% | 25% | 81% | 47 |
| SouthamptonSouth East | 18 | 94% | 83% | 61% | 17% | 44% | 22% | 61% | 48 |
| SwanseaWales | 17 | 77% | 65% | 41% | 12% | 35% | 29% | 41% | 49 |
| YorkYorkshire and the Humber | 17 | 100% | 88% | 59% | 41% | 35% | 6% | 59% | 56 |
What this means for your business
Most of the gaps in our data can be fixed without a big budget. These are the fixes, in the order we would make them:
- Make your phone number tap-to-call: Put your number at the top of every page as a tap-to-call link, and check it works on your own phone. It is one of the quickest fixes in this report.
- Show your hours and, if you have premises, your address: Add them to your homepage or contact page in plain text, matching your Google Business Profile exactly. Our Google Business Profile health check shows whether your profile and website agree.
- Speed up your homepage on a phone: Compress large images and remove sliders and plugins you do not need. Our guide to page speed SEO covers the usual causes, and Core Web Vitals explains what Google measures.
- Add LocalBusiness structured data: One short block of code on your homepage states your business type, address, phone number and hours. Our guide to schema markup shows what to include.
- Give each main area its own page: If you serve several towns, a useful page for each one, with real local detail, gives Google a page to show in each place. See SEO for multiple cities for how to do it without thin copies.
- Show your reviews on your website: If customers praise you on Google, show those reviews where website visitors can see them too. The review booster kit helps you get more of them, and our guide on how to get more Google reviews covers what is allowed.
- Make sure AI tools can describe you: AI answers draw on the same facts as Google: your listing, your website and other sites that mention you. The AI visibility check shows whether ChatGPT and Gemini name your business.
- Check before you hire anyone: Some owners in our forum sample had paid for work that did not help. Our guide on how to choose an SEO agency lists the questions to ask, and SEO scams covers the warning signs.
If you would rather work through it one step at a time, the 30-day local SEO plan turns these fixes into one 20-minute task a day, and the citation finder lists the UK directories worth being on for your industry.
Free SEO audit
See how your website compares
Our free SEO audit checks the same basics as this report on your own website and Google Business Profile, then shows you the first three things to fix.
Get the PDF
The full report is also available as a designed PDF to read offline, print or share with your team.
Free download
Get the full report as a PDF
The whole report, with every chart and both tables, in a designed PDF you can read offline, print or share with your team.
Here you go. Your download is ready:
Download the data
You can download every website result in this report as a spreadsheet. It has one row per site (408 rows), with the industry, town, region, website builder and each check, and no business names or web addresses.
Download the dataset (CSV, 408 rows)
You are welcome to use the data and the findings in your own articles and research. Please credit Sandcrowd and link to this page. A suggested citation:
Sandcrowd (2026). Local Search in the UK: How Small Businesses Show Up on Google, 2027. https://sandcrowd.com/research/local-search-uk-2027/
Methodology
How we chose the websites
We picked 20 industries that are typical small and local UK businesses, and 25 towns and cities spread across England, Scotland, Wales and Northern Ireland, with every English region represented. Each town was paired with 10 of the industries, for 250 searches in total, such as “plumber Leeds UK”. Where a search gave only one usable site, we ran one follow-up search with slightly different wording (4 in all).
We ran each search once through an ordinary web search tool. We did not scrape Google Search or Google Maps, which Google’s terms do not allow. From each set of results we took the first one or two websites that belonged to a small, independent business in or serving that town. We skipped directories, lead sites, marketplaces, national chains and franchises, and SEO or web agencies. That gave 497 websites, and the search, its position and any skipped results are logged for each one.
Because every site came from a search result, the sample leans towards businesses that already show up online. Businesses with no website, or with a site that does not appear in search, are not included, so the wider picture is likely to be weaker than ours, not stronger.
How we checked each site
A script loaded each homepage, and the contact page where the homepage linked to one, as a phone would (a 412 by 823 pixel screen), on 24 October 2026. It identified itself as Sandcrowd research with a link to this page, visited no more than three sites at once with a pause between requests, and followed each site’s robots.txt file. Of the 497 sites, 66 blocked automated visits, 15 could not be reached, 6 now send visitors to a different website (such as a national chain’s website) and 1 showed only a hosting placeholder. We also left out 1 site that our hand check showed belonged to a national chain, so 408 were checked. Each check was defined like this:
- Tap-to-call: The homepage has a phone link (a “tel:” link) that opens the dialler when tapped.
- Address: A full UK postcode appears in the text of the homepage or contact page.
- Opening hours: The homepage or contact page shows a day of the week with a time (such as “Monday 9am to 5pm” or “Sunday closed”), says “24/7”, or states hours in structured data. A heading or link that only says “opening hours” does not count.
- Area pages: The homepage links to an “areas we cover” style page (such as /areas-we-cover/, /locations/ or /offices/), or to a page named for a place other than the business’s own town (such as /plumber-in-horsforth/ on a Leeds plumber’s site). Service pages named for the home town, and portfolio or news pages, do not count.
- LocalBusiness structured data: The homepage or contact page includes schema.org LocalBusiness data, or any of its more specific types such as Dentist or Plumber.
- Contact form or booking: There is a contact form, or online booking as defined next.
- Online booking: The site embeds or links to a known booking system, or has a “book now” style link that leads to a booking page. A “book” button that only opens the contact form does not count.
- Reviews shown: The homepage has a review widget, review structured data or a reviews heading, or links to a reviews or testimonials page.
- Town in title: The homepage title contains the name of the town we searched.
- Website builder: Taken from markers in the page code, such as a WordPress or Wix signature. Where none was found, it is listed as other or unknown.
How we measured speed
We planned to use Google’s PageSpeed Insights, but its free, keyless access was out of daily quota, so we ran Lighthouse 12.2.1, the same open-source test PageSpeed Insights uses, on our own computer with its default mobile settings (a simulated mid-range phone on a slow 4G connection). We tested 391 homepages, one at a time. Scores can differ by a few points from PageSpeed Insights and from one run to the next, so we report the middle score and broad bands rather than exact rankings.
How accurate the automated checks are
We tested the script against our own judgement twice. First we checked 20 randomly chosen sites by hand, and the script agreed with us on 163 of 180 answers (91%). That check showed where its rules were too loose, such as reading “monthly” as Monday or counting a service page named after the home town as an area page, so we tightened the rules and checked every site again. Then we checked 15 different sites by hand, also chosen at random and not used to fix the rules. The script agreed with us on 129 of 135 answers (96%), and that is the figure to judge the data by. Where we and the script disagreed, we went back to the page to settle which was right.
How we read the owner posts
Reddit’s robots.txt file asks automated tools not to visit, so we left Reddit out. We used UK Business Forums, whose robots.txt allows its thread pages. We listed every thread started from January 2024 to October 2026 on six boards (sales and marketing, websites and ecommerce, IT and internet, general business, start-ups, and website reviews), and every thread started in 2022 and 2023 on the two busiest of them for this topic, sales and marketing and websites and ecommerce. We kept threads whose titles mention Google, SEO, websites, reviews, ads, AI, leads or similar words (on the website reviews board we kept every thread), and read the opening post of each (421 posts). We also read the first 10 replies in 20 active threads about getting found (200 replies).
Each post was read and tagged by hand. We marked who was writing (a business owner, someone selling SEO or web services, or someone else) and whether it was about the owner’s own visibility on Google. We excluded 84 posts by sellers, 33 follow-ups by the person who started the thread, 119 advice posts by people who were not writing about their own business, and 217 posts on other topics. That left 168 owner posts. Many forum members run online shops rather than local service businesses, so these themes describe UK small business owners in general. We stored no usernames, and the examples above are paraphrased.
Limitations
- Sample size: With 408 websites, the overall figures are a fair picture of small businesses that show up in search, but the industry and town figures rest on 10 to 20 sites each and can move by several points with a handful of sites.
- Search results, not a census: Every site came from a search result, so businesses that are hard to find online are under-represented.
- Two pages per site: We checked the homepage and the contact page. A site that shows its hours only on another page will be counted as having none, so some figures may slightly understate what sites have.
- Automated checks: Scripts read code, not intent. A phone number shown only as an image, or reviews shown in an unusual way, can be missed.
- Blocked sites: 66 sites blocked automated visits and were left out rather than worked around. These are often sites behind security services, which may differ from the rest.
- Chains that look independent: We left out sites that send visitors to a chain’s website or that our hand check showed belong to one. Some group-owned practices keep their own name and website, though, so a few may remain in the sample.
- Forum posts: Owners who post on forums are not a random sample of all owners, and a post is more likely when something has gone wrong.
Every number in this report can be traced to a file. The site list, search log, page checks, speed tests, evidence for each check, tagged posts and the scripts that produced them are kept with the report, and we are happy to share them with journalists and researchers on request.
Sources
Figures in this report that are not our own data come from these sources, each checked on 22 or 23 October 2026:
- Department for Business and Trade: Business population estimates for the UK and regions 2025, published 2 October 2025. View source.
- Ofcom: Adults' Media Use and Attitudes report 2026, published 2 April 2026. View source.
- Google: Q2 2026 earnings call remarks, published 22 July 2026. View source.
- Google Business Profile Help: How Google determines local ranking. View source.
- Google Search Central: Local business structured data. View source.
- Chrome for Developers: Lighthouse performance scoring. View source.
- web.dev: Largest Contentful Paint (LCP). View source.
- BrightLocal: Local Consumer Review Survey 2026, US adults, published 11 February 2026. View source.