What's really wrong with small business websites

Six things the data says

• 9

  The median website had nine problems worth fixing. We logged 4,043 issues across 439 sites. Only 11 sites (2.5%) were rated strong with little to change.

• 75%

  Three in four sites had missing or broken schema markup (304 of the 405 sites where it was checked), the most common problem in the study.

• 70%

  Seven in ten lacked proper service pages: services crammed onto one page, missing, or thin. It was the fix most often called top priority, on 192 sites.

• 7%

  Mobile is solved. Structure is not. Only 7% had mobile layout problems, while 63% had title tag and meta description issues and 61% had heading problems.

• 73%

  Money, health and legal sites lacked trust signals more often (73% against 56%), and the gap was flagged as high priority more than twice as often.

• 68%

  Health and medical sites were the most likely to need significant work, against 39% of local service businesses and 39% of marketing agencies.

• Key statistics
• About this research
• The problems, ranked
• Service pages
• Schema markup
• Titles and headings
• Location pages
• Trust
• Conversion
• By business type and country
• The fix order
• Questions
• Downloads and data
• Method and limitations

The numbers, ready to quote

Each statistic links to itself. Quote freely with a link back to this page.

• The median small business website we reviewed had 9 distinct SEO and conversion problems (439 websites, AI Ranking, 2026).Copy link
• Only 2.5% of small business websites (11 of 439) were rated strong with little to change (AI Ranking, 2026).Copy link
• 46% of small business websites (202 of 439) needed significant work, in the reviewer's overall assessment (AI Ranking, 2026).Copy link
• 75% of small business websites had missing, wrong or invalid schema markup (304 of 405 sites checked, AI Ranking, 2026).Copy link
• 70% of small business websites had weak, missing or merged service pages (273 of 392 checked), the problem most often called the top priority fix (AI Ranking, 2026).Copy link
• 63% of small business websites had title tag or meta description problems (228 of 361 checked, AI Ranking, 2026).Copy link
• 61% of small business websites had H1 or heading problems, most often a slogan used where the keyword should be (194 of 318 checked, AI Ranking, 2026).Copy link
• 62% of local business websites had missing or near-duplicate location pages (160 of 258 where local search was reviewed, AI Ranking, 2026).Copy link
• 62% of small business websites were missing trust signals such as reviews, sources for claims or a named person (224 of 362 checked, AI Ranking, 2026).Copy link
• Health, money, legal and safety websites lacked trust signals 73% of the time against 56% for other sites, and the gap was called high priority 44% against 19% (AI Ranking, 2026).Copy link
• 44% of small business websites showed no visible owner, team or author (160 of 362 checked, AI Ranking, 2026).Copy link
• 57% of small business websites had calls to action that were buried, broken or cluttered (205 of 363 checked, AI Ranking, 2026).Copy link
• 48% of small business websites failed the first impression test: a visitor could not tell within seconds what the business does, where, and what to do next (196 of 406 checked, AI Ranking, 2026).Copy link
• 54% of small business websites loaded slowly enough to be flagged (174 of 322 checked, AI Ranking, 2026).Copy link
• Only 7% of small business websites had mobile layout problems (27 of 406 checked): the remaining problems are structural, not technical (AI Ranking, 2026).Copy link
• 68% of health and medical websites needed significant work, against 39% of local service businesses and 39% of marketing agencies (AI Ranking, 2026).Copy link
• 95% of e-commerce websites had schema problems (37 of 39 checked, AI Ranking, 2026).Copy link
• 64% of the websites checked for indexing had indexing or technical errors (105 of 163, AI Ranking, 2026).Copy link

What we reviewed, why, and where

What each review covered

Share of the 439 reviews in which each area was examined. Every rate in this report is calculated only over the sites where that area was actually checked.

When

Reviews recorded per month. Volume grew as the community did.

How often each problem showed up

The chart shows the share of sites with each problem, among the sites where that area was checked. The most common problems are not exotic technical failures. They are basic decisions about structure, labels and clarity that a website builder cannot make for you.

Areas checked on fewer than 150 sites are left out of the ranking because they were mostly examined when something already looked wrong: Google Business Profile gaps (45 of 59 checked) and weak backlink profiles (32 of 50). Full figures are in the data appendix.

What was most often called the top priority

Number of sites where the reviewer treated the problem as high priority (a fix to make first).

How many problems each site had

Number of sites by count of distinct problems recorded. Most sites had between 7 and 12. The coding recorded at most 12 problems per review, so the last bar means 12 or more.

One page cannot rank for ten services

• A kitchen exhaust cleaning company in the UAE listed every service on one page.
• A law firm in Saudi Arabia grouped consultation, notary and dispute services together, so none could rank on its own.
• A business coach in Australia had service pages with no pricing, no process, no "who it's for" and no FAQs.

Most sites don't tell machines what they are

{
  "@type": "LocalBusiness",
  "name": "Brightside Plumbing"
  "telephone"    missing
  "address"      missing
  "image"        missing
  "priceRange"   missing
}

{
  "@type": "Plumber",
  "name": "Brightside Plumbing",
  "telephone": "+1 512 555 0142",
  "address": { ... Austin, TX },
  "areaServed": ["Austin", "Round Rock"],
  "aggregateRating": { 4.9, 210 }
}

• A plumber in California had LocalBusiness schema that failed validation: no name, image, phone number, price range or address.
• A digital marketing agency in Australia had only organization schema, with no service or review markup.
• A supplements store in Italy had little or no schema at all on its collection pages.

The tagline trap

• An accounting firm in Australia used its slogan as the main heading instead of the service it wanted to rank for.
• A pediatric dentist in California had several H1 tags on the homepage.
• A garage door company in Texas used its tagline as the page title rather than the service and town.

Copying a page and changing the town name doesn't work

• An invoice finance company in Texas had area pages that only swapped the place name and a few landmarks.
• A psychology practice in Australia targeted the whole country in its metadata, too broad to rank anywhere locally.
• A holiday rental site in Spain wrote its town pages as travel guides instead of listing the rentals in each town.

Health, money and legal sites have the most to prove, and prove it least

• A medical clinic in India made health claims on its blog with no source behind them.
• A life insurance broker in Canada quoted national statistics without naming where they came from.
• An orthopedic surgeon in Singapore had no patient testimonials on any service page.

Ranking is half the job

• A marketing agency in the UK had "book a consultation" buttons that did nothing when clicked.
• An engineering consultancy in the UK had no call to action on its service pages. The only button was "view case studies".
• A pet products store in Australia opened with a tagline and a discount code, leaving visitors unsure what to buy.

Who struggles most

Share of sites in each group that the reviewer judged to need significant work. Groups with at least 30 sites.

The three most common problems in each group

By country

The three largest country groups. Problems were remarkably consistent across markets: schema and service pages lead almost everywhere.

If you only do ten things, do these, in this order

The sites that came out best had a homepage that says what, where and what next in the first screen; a dedicated, well-linked page for each service and area; real people on the site; recent reviews close to the call to action; and schema that validates. None of these needs a large budget. They need decisions made before the design.

1. Pass the two-second testSay what you do, where, and what to do next in the first screen of the homepage.
2. Give every service its own pageA hub that links to one page per service, written for someone ready to buy.
3. Fix titles and H1sService plus location in the title and the H1. Brand last. Tagline as a subheading.
4. Add and validate schemaLocalBusiness or Organization, Service, Person, FAQ and reviews, matched to each page.
5. Put one clear call to action high on every service pageAnd click it to make sure it works.
6. Show who is behind the businessNames, photos, credentials, author bios and Person schema.
7. Source every claimLink statistics and health or money claims to the original source, next to the claim.
8. Build location pages that are genuinely differentLocal detail, local proof, a map. Never a copy with the town name swapped.
9. Make it fastConvert images to WebP, compress or remove heavy video, and test on mobile.
10. Check what Google has indexedUse Search Console, fix pages that are crawled but not indexed, and link blog posts to service pages.

Frequently asked questions

Take the report with you

Chart images for articles and slides

Free to use in articles, posts and presentations with a link back to this page.

How to cite this report

You are welcome to quote, chart and republish these findings with a link back to this page.

How the videos became data

Data appendix: every problem category