The evidence — Baseline

American local government is small

Smaller than the companies building software for it seem to think.

US municipalities by population
Population	Towns	Share
Under 1,000	14,829	50%
1,000 to 5,000	8,384	28%
5,000 to 25,000	4,409	15%
25,000 and over	1,876	6%

Half of all American municipalities have fewer than a thousand people in them. And 11,499 of them, nearly four in ten, have no website at all. Almost every one of those is a town under 5,000.

Of the small towns that do have a site, the most common thing they are running is WordPress, on about 3,300 of them. Massachusetts is genuinely unusual, with one government vendor holding more towns than everything else put together. The rest of the country does not look like that.

Our survey of 29,498 US municipalities, August 2026. Platform identified from the live site.

What we found on their websites

We checked the homepages of 324 towns in August 2026. There is more about this on the accessibility page. Two things stood out.

The problems are boring. Unreadable text, links with no name, photos with no description. Nothing exotic. Nothing that took a hard decision to get wrong. Just what happens when a tool lets a busy person make an ordinary mistake.

Even the best of them fails. The strongest platform we tested still left about four broken things on the average homepage. Not a catastrophe. Also not zero, and zero is available.

Here is the part that does not flatter our sales pitch: the platforms built specifically for government were measurably better than the general website builders. We are not claiming those companies are careless. We are claiming that a tool which offers a color picker will eventually ship something nobody can read, no matter how careful everybody is.

What we have not proven

We would rather write this down than have you discover it.

Turning a whole archive of documents into readable pages
This is central to the product and nobody has tested it against a full town archive yet. Scans, multi-column layouts and complicated tables are genuinely difficult. We are working on it now and we will publish what we find, including the parts that go badly.
Everything past the homepage
Our survey covered homepages. Everything we know says the interior pages, the forms and the documents are worse, but we did not test them, so we are not going to quote you a number.
Moving every town automatically
Some hosts block us. Where that happens, moving your site takes people.
How much time this saves your Clerk
We think the website eats a couple of hours a month and we intend to cut that to under thirty minutes. Nobody has measured the before, including us. Treat it as a hypothesis.

How we will know if it worked

What we are aiming at, against what towns have today
What	Aiming at	Today
Document arrives to live on the site	Under 5 minutes, no training	Hours to weeks
Broken things on a published page	None, continuously	About four, on the best platform
Town documents readable as pages	All of them	Close to none
Clerk time on the website each month	Under 30 minutes	Nobody has measured it

Accessibility is what gets us in the door. Giving the Clerk their afternoon back is what keeps us there.