Finding
Whito scored 112 UK business websites on 19 November 2026 and the average was 71.5 out of 100; price (31) and reviews (33) were the weakest of 14 areas, and phones and speed (91 each) the strongest.
Source: Whito, whito.co.uk/research/uk-business-website-quality-index, 2026-11-19. Free to quote with a link to this page.
Last Updated on November 19, 2026
Read a short summary
General summary
On 19 November 2026 Whito ran the same checks on the homepages of UK business websites, six from each of 20 trades. 112 could be scored. The average was 71.5 out of 100. Nearly every site worked on a phone, loaded quickly and could be read by search engines. The low scores were for showing a price (31 out of 100) and showing reviews (33). Plumbers, removal firms and accountants scored highest. Furniture makers and roofers scored lowest.
Bullet points
- 120 websites tested on 19 November 2026. 112 scored.
- Average 71.5 out of 100. Middle site 72. Range 42.6 to 95.
- Strongest areas: phones 91, speed 91, search engines 90.
- Weakest areas: price 31, reviews 33, business identity 60.
- National brands averaged 75.6. Local firms averaged 70.4.
- The full file is free to download.
Basic explainer
We opened lots of business websites and asked the same questions of each one. Does it work on a phone? Is it quick? Does it show a price, a phone number and who runs the business? Each site got a mark out of 100. Most sites get the technical parts right. Many leave out the things a customer wants to know before they call.
These summaries were written with AI from this page and checked by Whito.
Short answer: Whito scored 112 UK business websites on 19 November 2026 and the average was 71.5 out of 100; price (31) and reviews (33) were the weakest of 14 areas, and phones and speed (91 each) the strongest. The technical basics are close to universal. What pulls scores down is information a customer wants before they get in touch: what it costs, what other customers say, and who the business legally is.
This is wave 2 of the UK Business Website Census. Wave 1 counted 22 basic checks on 100 sites. This wave reads 120 fields per site, groups them into 14 areas and gives each site a score.
What we tested
We took the 20 trades with the most websites in the Whito directory and drew six websites at random from each, leaving out sites tested in wave 1. That gave 120 addresses. We opened each homepage in an ordinary Chrome browser on 19 November 2026 and ran one script that asked the same questions of every site.
| Outcome | Websites |
|---|---|
| Loaded and scored | 112 |
| Our test tool was not allowed to run on the site | 5 |
| Security certificate error, site would not open | 1 |
| Address now shows a parked domain page | 1 |
| Site overwrote the test before it finished | 1 |
| Total tested | 120 |
The five sites in the second row were blocked by a rule in our own browser tool, not by the websites. We did not work round it. Of the 112 scored, 87 are local firms and 25 are national brands.
The average UK business website scored 71.5 out of 100
The middle site scored 72. A quarter scored under 64 and a quarter scored over 80.7. The best site scored 95 and the worst 42.6.
| Score | Websites |
|---|---|
| 80 or more | 29 |
| 70 to 79 | 35 |
| 60 to 69 | 27 |
| 50 to 59 | 15 |
| Under 50 | 6 |
Our reading: the parts a web designer or website builder sets up by default are nearly always there. The parts the owner has to decide to publish are where the points go.
Where UK business websites lose points
| Problem | Websites (of 112) | Share |
|---|---|---|
| No price in pounds on the homepage or a pricing page | 67 | 60% |
| No review platform named and no rating shown | 71 | 63% |
| No local business details in machine-readable form | 66 | 59% |
| A heading level skipped (for example H2 straight to H4) | 58 | 52% |
| No company number on the home, contact, privacy or terms page | 50 | 45% |
| A form field with no label | 52 | 46% |
| No terms page linked | 47 | 42% |
| None or several main headings (H1) instead of one | 36 | 32% |
| No tap-to-call phone number | 31 | 28% |
| Main text smaller than 16 pixels | 29 | 26% |
| At least one image with no alt text attribute | 21 | 19% |
| No privacy page linked | 20 | 18% |
A label tells a screen reader what a form box is for. Alt text does the same job for a picture. Machine-readable details (often called schema or structured data) are a short block of code that tells Google and AI tools the business name, address and type.
On security, all 112 sites used https. 53 sent the header that tells browsers to always use the secure address (HSTS), and 27 set a content security policy. 2 loaded insecure files on a secure page.
Scores by type of business
| Group | Websites | Average | Lowest | Highest |
|---|---|---|---|---|
| Trades and home services | 45 | 74.1 | 42.6 | 93.2 |
| Professional services | 29 | 73.8 | 50.5 | 95.0 |
| Health and care | 14 | 70.7 | 44.5 | 89.3 |
| Industrial and supply | 24 | 64.6 | 45.6 | 82.7 |
Industrial and supply firms scored 8 for reviews and 15 for price. Many sell to other firms on quotation, so this is partly how the trade works. It still means a new buyer, or an AI tool reading the site for one, finds less to go on.
| Trade | Websites | Average score |
|---|---|---|
| Plumbers | 6 | 83.9 |
| Removals | 5 | 83.2 |
| Accountants | 6 | 81.8 |
| Solicitors | 6 | 81.0 |
| Dentists | 5 | 80.8 |
| Architects | 5 | 74.5 |
| Windows and doors | 6 | 74.2 |
| Builders | 6 | 72.6 |
| Garages | 5 | 72.5 |
| Electricians | 6 | 72.2 |
| Cleaners | 6 | 71.5 |
| IT consultants | 6 | 67.8 |
| Engineering companies | 6 | 67.5 |
| Manufacturers | 6 | 66.7 |
| Wholesalers and distributors | 6 | 66.6 |
| Vets | 5 | 65.8 |
| Home care | 4 | 64.2 |
| Interior designers | 6 | 63.9 |
| Roofers | 5 | 61.5 |
| Furniture makers | 6 | 57.6 |
Each trade has four to six sites, so one unusual site moves a trade several places. Read the trade table as a first look. The four groups above are the safer comparison.
Local firms and national brands
National brands averaged 75.6 and local firms 70.4. The gap comes from what brands publish: 16 of 25 showed a price on the homepage or a pricing page against 29 of 87 local firms, and 15 of 25 named a review platform against 14 of 87. Local firms were ahead on being easy to ring: 67 of 87 had a tap-to-call number against 14 of 25 brands.
Of the 53 sites built on WordPress the average was 71.6. The 7 Wix sites averaged 68.1. Seven sites is too few to rank one builder above another.
Why there is no regional table
The brief for this index asked for regional comparisons. We have left them out. 33 of the 112 sites are in Nottingham and Derby postcodes and 10 are in Leeds, because that is where the Whito directory has most listings. A regional table from this sample would describe our directory more than the regions. Wave 3 will draw a set number of sites per region.
AI tools: a third of sites have an llms.txt file, mostly put there by the website builder
An llms.txt file is a plain text page that describes a site for AI tools. 37 of 112 sites had one, close to the 27 of 90 in wave 1. All 7 Wix sites had one and so did 20 of the 53 WordPress sites. We opened three by hand. One said it was generated by the Yoast SEO plugin and one came from Wix. So most of these files look like a default from the software, not a choice by the owner.
12 sites named an AI crawler in robots.txt, the file that tells automated visitors where they may go. 6 blocked at least one. None blocked every visitor.
Speed on one ordinary connection
On one desktop computer on UK home broadband, the middle site sent its first byte in 0.25 seconds and finished loading in 1.17 seconds. 5 of 112 took more than 4 seconds and 15 took more than 0.8 seconds to send the first byte. This is one visit on one day from one place. A phone on mobile data will be slower.
How this compares with wave 1
The two waves use different draws, so differences are not a trend. On the checks that are the same, the picture holds: every measured site used https in both waves, 98% and 100% had a phone-ready page, 91% and 89% had a meta description, and 30% and 33% had an llms.txt file. A price on the homepage was rarer in this wave (25% against 37%) because it holds fewer national brands.
How the score works
Each of the 14 areas is scored from 0 to 100 by averaging the checks inside it. The site score is the plain average of the 14 areas. Nothing is weighted by opinion, except that the AI area gives the llms.txt file a fifth of the marks because the file is new and its value is not proven.
| Area | Checks |
|---|---|
| Phones | Page is set up for phone screens; page is no wider than the window; main text is 16 pixels or larger |
| Speed | Full load within 2.5 seconds (half marks up to 4); first byte within 0.8 seconds; largest item on screen within 2.5 seconds |
| Access for disabled users | Page language is set; every image has an alt attribute; every form field has a label; no heading level skipped |
| Security | https; HSTS header; no-sniff header; no insecure files on the page |
| Contact details | Tap-to-call number; email link or enquiry form; postcode on the home or contact page |
| Price | A price in pounds on the home or pricing page; a price with a unit (per hour, plus VAT) or in machine-readable form |
| Trust pages | Privacy link; terms link; about link; a named accreditation or regulator |
| Reviews | Names a review platform or shows a rating; has a review widget or rating data |
| Getting in touch to buy | Enquiry form; quote or booking link; tap-to-call number |
| Machine-readable details | Any structured data; business or organisation type; local business, service or product type |
| Search engines | Meta description; exactly one H1; canonical tag; not hidden from search; sitemap; robots.txt |
| AI tools | robots.txt blocks no AI crawler (40%); words are in the page itself, not only added by scripts (40%); llms.txt file (20%) |
| Business identity | Company number; postcode; registered-in wording or VAT number |
| Services explained | Services or products link; 300 or more words on the homepage; two or more H2 headings |
Every score was worked out twice, once in the browser and once from the downloaded file with separate code. The two agreed on all 112 sites.
Limits
- The sites come from the Whito directory. It is not a sample of all UK business websites and it leans towards the East Midlands and Leeds.
- Six sites per trade is small. Group results are firmer than trade results.
- We read the homepage, the files robots.txt, llms.txt and sitemap.xml, and four linked pages (pricing, contact, privacy, terms). Details kept elsewhere on a site are not counted.
- Checks are automatic. A price check looks for a pound sign followed by a number. It cannot tell a clear price list from a single from-price.
- Speed is one visit from one desktop. The phone checks were run in a desktop-width window, so they show whether a site is set up for phones, not how it looks on one.
- Colour contrast and keyboard use were not tested. The access score covers four basics only.
- Sole traders and partnerships have no company number, so a missing number is not a fault for them. See the Trust Index for the count among registered companies only.
- A missing item is not a sign of a bad business. It means a customer cannot see it on the site.
Download the data
Download the wave 2 file (CSV, 124 rows). One row per website tested, with every field, the 14 area scores and the site score. Business names, web addresses and company numbers are left out. Free to use with a link to this page (CC BY 4.0). The method and scoring code are open on request: data and sources.
Two more studies use the same file: the UK Business Trust Index, on what a customer can check, and Can AI agents buy from UK businesses?
What a UK business owner can do with this
Start with the two cheapest fixes on the list. Put a price, or a from-price with what it covers, where a visitor can see it. Then name where your reviews are and link to them. Those two areas are where most sites in this study lost a third of their marks. To see how your own homepage does on the basics, run the free website check.
Common questions
What is a good score on the UK Business Website Quality Index?
The average of the 112 UK business websites scored on 19 November 2026 was 71.5 out of 100 and the middle site scored 72. 29 sites scored 80 or more. 21 scored under 60. The highest was 95 and the lowest 42.6.
What do UK business websites get wrong most often?
Price and reviews. 67 of 112 sites showed no price in pounds on the homepage or a pricing page, and 71 named no review platform and showed no rating. On the technical side, 58 skipped a heading level and 52 had a form field with no label.
Is this a score for every UK business website?
No. It covers 112 websites drawn from the Whito directory, six from each of 20 trades. A third of them are in the Nottingham and Derby area. It shows what these sites do. It is not an estimate for all UK business websites.
Cite this research: Whito (2026), UK Business Website Quality Index 2027, wave 2 of the UK Business Website Census, collected 19 November 2026. https://whito.co.uk/research/uk-business-website-quality-index/
If your business is named here and you think we have a fact wrong, email hello@whito.co.uk and we will check it and correct it. If you want to respond, send us a statement and we will publish it alongside. Full detail: Corrections, right of reply and removals.


