353 sites measured · generated 2026-11-04
We graded 353 local business websites. The median scores 50/100.
Not a survey of what owners say. A reading of what their sites actually serve: a weighted count of measured faults, on a wider check set than the eight checks the instant site check runs. Same page signals, more of them, and a different scale: read this median and a check score as two readings, not one number.
50/100
Median site health across the measured set.
56%
Too many assets. The most common measured fault.
39%
No tap-to-call, on the device where the search happened.
19%
Chain sites, where the local branch cannot buy the fix.
What is actually broken
The faults, by how many sites have them.
Share of the 353 measured sites
Too many assets
56% · 196 sites
More requests than a phone on a weak signal should have to make.
jQuery front end
54% · 192 sites
An older front end that is harder to make fast and rank.
Heavy markup
53% · 187 sites
Markup weight the visitor pays for before seeing anything.
No LocalBusiness schema
46% · 164 sites
No LocalBusiness structured data, so Google Maps has less to trust about who and where you are.
No tap-to-call
39% · 136 sites
A number you can read but not tap, on the device where the search happened.
Title too long
34% · 120 sites
The title is cut off in the result, so the useful part never shows.
No meta description
24% · 84 sites
Search engines have to invent the snippet under your title.
Template builder
20% · 71 sites
A template builder that makes speed and ranking an uphill fight.
The remaining 7 measured faults
- Missing H116%
- Dead contact form6%
- Slow server response5%
- No HTTPS2%
- Hidden from search2%
- No mobile viewport1%
- Missing page title0%
By city
Where the sites are, and how they score.
| City | Measured | Median score | Most common fault |
|---|---|---|---|
| Austin | 126 | 48/100 | jQuery front end |
| Dallas | 107 | 55/100 | Too many assets |
| Houston | 70 | 50/100 | No LocalBusiness schema |
| San Antonio | 26 | 50/100 | jQuery front end |
| Fort Worth | 20 | 54/100 | jQuery front end |
| Other | 4 | 62/100 | No LocalBusiness schema |
A reading, not a verdict
What this does and does not say
It says
These 353 sites, on the day they were graded, served pages with these measurable problems. The instant site check weighs its own eight checks on any URL: a different score, not this one.
It does not say
That every local site is broken, or that a fix guarantees revenue. It is not a census: sites that block scraping are absent, not healthy.
Contactability, separately
74% publish no email link and 58% no form. For a phone-first shop that is a channel choice, not a fault, but it is why cold outreach reaches fewer of these businesses than the count suggests.
Toolkit site-health 0-100, higher is better. On a local row the toolkit stores an opportunity score (higher = worse); this report inverts it once, here. The median is also published over independent (non-chain) rows only. Not a census. These are the businesses the toolkit sourced, graded and could verify as live; a site that blocks scraping is absent, not healthy. Percentages are shares of the measured set, not of the city. The filter is siteLiveness.verdict = live; the generation date is below. No business name, address, email or phone is published here; only aggregates, generated 2026-11-04.
The median is 50/100 across every measured site and 48/100 across the 287 independent, non-chain sites. Chains are 19% of the set and cannot buy the fix locally, so both are published rather than counting them as peers.
Where does your site sit?
Run the instant site check on your own URL. If it finds faults worth fixing, the teardown shows the one fix worth doing first.