Key findings
- Only 12% of small business homepages scored “good” (90+) in Google’s mobile PageSpeed test. 22% scored “poor” (under 50), and the median score was 64.
- In the same mobile test, the main content took a median 7.2 seconds to appear. Only 12% met Google’s 2.5-second “good” mark, and 72% took longer than 4 seconds.
- 1 in 3 homepages (32%) had no tap-to-call phone link. Among hair salons, it was more than half (53%).
- 42% of homepages weighed more than 3 MB, and 23% more than 5 MB.
- 59% had no local business structured data (schema), and 23% had no meta description.
- Only 13.5% of homepages had enough visitors for Google to publish real-user speed data, so for most, Google’s lab test is the only speed measure available.
How does your own website compare? Run the same checks on your site with the free website check. It takes about 30 seconds and shows where you stand against these 1,020 sites.
On 12 November 2026, I tested 1,020 small business websites from the United States, the United Kingdom, Canada and Australia: dentists, hair salons, plumbers and builders. Each homepage went through Google’s own mobile speed test and a check of the basics that turn a visitor into an enquiry. Below are the results, how the study was done and what it means if you run a small business.
Mobile speed: most sites are slower than Google recommends
Google PageSpeed Insights scores a page from 0 to 100 on a simulated mid-range phone on a mobile connection. Google calls 90+ good, 50–89 needs improvement and under 50 poor.
| Mobile PageSpeed score | Share of sites |
|---|---|
| Good (90–100) | 12.3% |
| Needs improvement (50–89) | 65.4% |
| Poor (0–49) | 22.4% |
The biggest problem is Largest Contentful Paint (LCP): how long the main image or headline takes to appear. Google’s threshold for “good” is 2.5 seconds. In the mobile test, the median was 7.2 seconds, only 12.3% of sites stayed under 2.5 seconds, and 72.5% took longer than 4 seconds. Layout stability was much better: 82.8% passed Google’s layout shift (CLS) threshold.
Heavy pages are a common cause. The median homepage weighed 2.5 MB, 41.8% were over 3 MB and 22.8% were over 5 MB. Large images, videos and sliders are common causes.
Enquiry basics: the easy wins many sites miss
| Check | Sites that had it |
|---|---|
| HTTPS (secure padlock) | 96.8% |
| Mobile viewport tag (page set up for phones) | 96.4% |
| Page title | 97.6% |
| Meta description (the summary Google may show) | 76.9% |
| Tap-to-call phone link | 67.5% |
| Local business structured data (schema) | 40.9% |
The basics of security and mobile layout are nearly universal now. The gaps are in the details that help a visitor take the next step: a phone number you can tap, and the information that helps Google understand and describe the business.
Results by industry
| Dentists | Hair salons | Plumbers | Builders | |
|---|---|---|---|---|
| Sites tested | 262 | 239 | 257 | 262 |
| Median mobile score | 64 | 64 | 65 | 64 |
| Good (90+) | 13.0% | 9.2% | 16.3% | 10.3% |
| Poor (under 50) | 26.0% | 18.8% | 22.2% | 22.1% |
| LCP under 2.5 s | 12.2% | 11.3% | 14.4% | 11.1% |
| Median page weight | 2.1 MB | 2.5 MB | 2.4 MB | 2.9 MB |
| Tap-to-call link | 81.3% | 46.9% | 81.3% | 58.8% |
| Meta description | 78.6% | 64.9% | 87.2% | 76.0% |
| Local business schema | 47.7% | 29.7% | 51.0% | 34.4% |
Speed was similar across all four industries. The differences are in the enquiry basics: hair salons were the least likely to have a tap-to-call link, a meta description or schema. Some salons rely on online booking apps instead, but a phone link still helps customers who’d rather call. Builders had the heaviest pages (median 2.9 MB).
Results by country
| United States | United Kingdom | Canada | Australia | |
|---|---|---|---|---|
| Sites tested | 429 | 256 | 180 | 155 |
| Median mobile score | 63 | 66 | 66 | 61 |
| Good (90+) | 14.5% | 13.3% | 10.6% | 6.5% |
| Poor (under 50) | 21.9% | 18.0% | 20.6% | 32.9% |
| Local business schema | 49.9% | 31.2% | 37.2% | 36.1% |
Australian sites in the sample were the slowest: one in three scored “poor”. US sites were the most likely to use local business schema.
Results by website platform
About half the sites (50.1%) were built on WordPress, followed by Wix (9.0%), Squarespace (6.7%), GoDaddy (2.9%), Duda (2.8%) and Webflow (2.4%). Shopify, Weebly, Drupal, Joomla and Jimdo were each 1% or less. The remaining 23.7% were custom-built or used a platform that couldn’t be identified.
| Platform | Sites | Median score | Good (90+) | Poor (under 50) | Median weight |
|---|---|---|---|---|---|
| WordPress | 511 | 63 | 10.6% | 26.2% | 3.1 MB |
| Wix | 92 | 71 | 7.6% | 12.0% | 2.5 MB |
| Squarespace | 68 | 57 | 1.5% | 22.1% | 2.4 MB |
| GoDaddy | 30 | 73 | 20.0% | 20.0% | 1.1 MB |
| Custom or other | 242 | 70 | 21.5% | 16.1% | 1.8 MB |
Treat this table with care. The platform is only one factor: images, plugins, themes and how the site was built matter more, and some groups are small. WordPress sites ranged from 19 to 100: 54 scored 90+, while 16 scored under 30. That shows how much depends on how a site is built. Only 1 of 68 Squarespace sites scored 90+, and only 1 had its main content appear within 2.5 seconds in the mobile test.
What this means for your business
- A slow mobile site loses enquiries. Many of your visitors are on a phone. If the first screen takes several seconds to appear, some will go back to Google and pick the next business.
- The fixes are often simple. Compressing and resizing images, removing heavy sliders and auto-playing videos, and cutting unused plugins usually make the biggest difference.
- Make it easy to get in touch. Put a tap-to-call phone link at the top of the page, and make your contact form or booking button easy to find.
- Help Google describe your business. Write a meta description for each main page and add local business schema with your name, services, area and contact details.
You can check your own homepage for free with the free website check, which runs Google’s mobile speed test plus the other checks in this study, or test speed alone in PageSpeed Insights (choose the Mobile tab). If you’d like help, see website redesign or small business website design, or read the website redesign checklist.
Methodology
- Sample: businesses listed on OpenStreetMap with a website, tagged as dentists, hair salons, plumbers or builders, in the US, UK, Canada and Australia. Chains (businesses with a brand tag, websites listed for three or more locations and branch pages) and social media or directory links were excluded. 1,312 websites were drawn at random with a fixed seed, about 330 per industry, split roughly 40% US, 25% UK, 20% Canada and 15% Australia. Fewer Canadian builders were listed, so builders were topped up from the other countries.
- Checks: each homepage was fetched once with a standard desktop browser user agent and tested once with Google PageSpeed Insights (Lighthouse, mobile strategy). The HTML was checked for HTTPS, the viewport tag, a
tel:link, a title, a meta description and schema.org local business types. The platform was identified from code signatures. - Results: 1,045 homepages loaded and 1,020 returned a PageSpeed score; all percentages are of these 1,020. Of the rest: 62 sites blocked automated checks, 205 didn’t load, returned an error or were parked domains after two attempts, and 25 loaded but didn’t return a score.
- Accuracy: with 1,020 sites, the overall percentages are accurate to within about ±3 percentage points (95% confidence). Industry and country figures, based on 155–429 sites each, are less precise.
- Limitations:
- Businesses listed on OpenStreetMap may not represent every small business.
- Only homepages were tested.
- PageSpeed results come from a single lab run on a simulated mid-range phone and connection, so real visitors on fast phones and Wi-Fi will often see faster loads.
- Tap-to-call links or schema added by JavaScript after the page loads may not have been detected, so those figures may slightly undercount.
- Privacy: results are reported in aggregate only. No individual business is named.
Website list © OpenStreetMap contributors, available under the Open Database License. Speed data from Google PageSpeed Insights.
Using this data
You’re welcome to quote the findings or reuse the tables in articles and presentations. Please credit “Aman Web Designs, Small Business Website Report 2026” with a link to this page. For questions about the data or the method, email info@amanwebdesigns.com.
Get a fixed price for your site
Tell me what you need and you'll get a free website plan with a fixed quote within 24 hours. No obligation.
Get my free website planBook a free call →See small business website design →


