All the results
Percentages use the number shown in each row as the base. Rows marked * are based on the 1,598 homepages whose HTML was small enough for us to read in full, because those checks look at the footer and the end of the page.
| Homepage loaded (HTTP 2xx, HTML) | 2,965 of 3,777 fetched (78.5%) |
|---|---|
| No mobile viewport tag | 131 of 2,965 (4.4%) |
| No valid HTTPS certificate | 106 of 2,965 (3.6%) |
| No http→https redirect | 325 of 2,965 (11.0%) |
| No page title | 81 of 2,965 (2.7%) |
| No meta description | 908 of 2,965 (30.6%) |
| Footer copyright year shown | 823 of 1,598 (51.5%)* |
| Copyright year 2023 or earlier (of those showing a year) | 162 of 823 (19.7%)* |
| Copyright year 2026 (of those showing a year) | 577 of 823 (70.1%)* |
| No tap-to-call (tel:) link | 904 of 1,598 (56.6%)* |
| No Google review link | 1,571 of 1,598 (98.3%)* |
| No LocalBusiness-type schema | 1,419 of 1,598 (88.8%)* |
| Median homepage HTML size | 130.4 KB |
| Median response time (page fully received) | 1,534 ms |
| Slowest tenth took longer than | 3,840 ms |
| Took over 3 seconds | 497 of 2,965 (16.8%) |
The same figures are in a CSV of aggregate figures (totals only).
What the sites are built with
From the page's generator tag and well-known file paths. "Other / unknown" includes custom sites and platforms we did not look for.
| Platform | Homepages | Share |
|---|---|---|
| WordPress | 1,154 | 38.9% |
| Other / unknown | 942 | 31.8% |
| Wix | 275 | 9.3% |
| Squarespace | 221 | 7.5% |
| Shopify | 172 | 5.8% |
| GoDaddy | 51 | 1.7% |
| Weebly | 40 | 1.3% |
| Duda | 34 | 1.1% |
| Webflow | 30 | 1.0% |
| Drupal | 16 | 0.5% |
| Joomla | 15 | 0.5% |
| Jimdo | 9 | 0.3% |
| Framer | 6 | 0.2% |
By area
| Area | Loaded | No mobile viewport | HTTPS or redirect problem | No meta description |
|---|---|---|---|---|
| West Sussex | 1,198 | 4.3% | 12.7% | 30.2% |
| East Sussex | 1,033 | 5.2% | 11.6% | 32.9% |
| Brighton & Hove | 734 | 3.4% | 11.9% | 28.1% |
Method and limits
- Business list. On 12 October 2026 we asked OpenStreetMap, through the Overpass API, for every node, way and relation in West Sussex, East Sussex and Brighton and Hove that has a
websiteorcontact:websitetag, carries a shop, craft, office, tourism or amenity tag, and has nobrandorbrand:wikidatatag. We left out schools, places of worship, town halls, police, hospitals and libraries, and counted each website domain once. That gave 3,818 sites. - Fetching. We read each site's robots.txt first and skipped 41 sites whose rules disallow us or all crawlers. For the rest we made one request for the homepage, no more than five at a time, with a 15 second timeout, no retries and no links followed beyond redirects, plus one HEAD request to the http:// address to see whether it redirects to https://. Our user agent was
MillwardsStudyBot/1.0 (+https://millwards.shop/sussex-website-study-2026.html#method; hello@millwards.shop)(we have not confirmed that this address forwards to a monitored inbox, so please use the enquiry form in the opt-out section instead). Correction, 12 October 2026: a first run was interrupted by a fault on our side and restarted, and an accidental second copy of the crawler then ran alongside the first. As a result some homepages were fetched more than once, and across the two runs up to 10 requests were in flight at once (five per run, each to a different host). The published figures use the first completed run only; the repeat fetches were not counted. - Result. 2,965 homepages loaded as an HTML page. 812 did not (errors, timeouts, non-HTML or non-success responses) and are not in any percentage.
- Limits. Homepages only. OpenStreetMap coverage is incomplete and made by volunteers, so this is a sample of businesses that happen to be listed with a website, not every Sussex business. Some sites show different content to automated requests, some check the viewport or schema with scripts we do not run, and a business's website tag can point to a social page or a directory. The copyright year, tap-to-call, review-link and schema checks read the HTML only and may miss content added by JavaScript. Platform detection is a best guess. Response time was measured from one UK connection at one moment. Treat the numbers as a snapshot, not a ranking of any business.
- Who ran it. Millwards Digital, a one-person web design business in Sussex. We sell websites, so read the findings with that in mind; the checks are listed above and you can repeat them yourself.
Fix-it checklist
Each item is a check from the study. Most can be done in a website builder or WordPress dashboard without a developer.
- Add the mobile viewport tag. A page needs
<meta name="viewport" content="width=device-width, initial-scale=1">in its head and a layout that fits a phone. On a website builder, choose a mobile-ready template. Then use the free ten-point website checklist on your own phone. - Get a valid HTTPS certificate and redirect. Most hosts offer a free certificate. Once it is on, make sure http:// addresses redirect to https://, so visitors never see "Not secure".
- Write a title and a meta description. One plain sentence on what you do and where, for each important page. It is what people see in search results.
- Update the footer year, or remove it. An old copyright year suggests nobody looks after the site. Either update it automatically or take it out.
- Make the phone number tap-to-call. Link it with
tel:so a customer on a phone can call in one tap. - Link to your Google reviews. Very few homepages did. Make a free Google review link and QR code and put the link on your site, receipts or window.
- Add LocalBusiness schema. It tells search engines your business name, area and contact details in a machine-readable form. Keep it identical to what the page says.
- Check speed. Large uncompressed images are the usual cause of slow pages.
Not sure what a fix or a rebuild would cost? Try the free website cost calculator, read the UK website costs guide, or see web design in Sussex.
Data credit and opt-out
Business list © OpenStreetMap contributors, available under the Open Database Licence. The results on this page are our own aggregate counts.
Want your site left out of future runs? Use the enquiry form and write “Study opt-out” with your website address. The form emails the Millwards Digital admin inbox, which is how we confirmed it is monitored. Updated 12 Oct 2026: methodology note corrected.