We checked again eight days later. Nothing had moved.

On 2026-09-03 we ran a free check on 145 small businesses and counted 545 open problems. On 2026-09-11 we ran every one of those checks again on the same addresses. Of the 141 we could read both times, 545 problems were still open. Not one had been fixed.

Free. No signup. Takes about a minute.

Why we looked twice

A measurement taken once is a photograph, not a fact.

The first run told us what was open on one morning. It could not tell us whether these were problems businesses were quietly working through, or problems nobody had ever mentioned to them. The only way to find out was to wait and look again, so we did, with the same scanner and the same checks.

What we compared

Every check that failed for a business on 2026-09-03, checked again on 2026-09-11 for that same business. The comparison is one check at a time, not two totals held up against each other, so nothing here can be an accident of two numbers happening to match.

What we left out

4 of the 145 addresses did not answer at all the second time, so they are excluded from every count on this page rather than counted as failures. We cannot tell a business that has closed from a server having a bad afternoon, and we will not guess.

The count

545 problems open in September. The same 545 eight days on.

Not one business out of 141 had closed a single one. This is not a story about lazy businesses. It is what happens to a problem nobody has told you about: it stays exactly where it is, because nothing in a normal week brings it to your attention.

Each of these is free to fix and most take an afternoon. Three of them are a setting at the company that sells you your web address, changed once and then true forever. None of them requires buying anything from us.

What this does not prove

Eight days is a short time.

We are not saying these businesses will never fix these things, and we are not saying eight days is long enough for anybody to act. It is not. What the second reading shows is narrower and, we think, more useful: nothing here is drifting shut on its own. Every one of these stays open until a person decides to close it, and almost nobody knows there is anything to decide.

We will read this cohort again. When we do, this page gets the new count, including the count that makes us wrong.

You cannot fix what nobody has told you about. This is us telling you.

The same check, on your own address, in about a minute. It is free, it needs no signup, and the result is a private page that is not listed anywhere.

Free. No signup. Takes about a minute.