A record that reads public registers is only as good as its last check. When a published claim turns out wrong, it is logged here with the date, and the page is changed in the open.
2026-10-29 · Coverage · /about
It said. That community-association management is licensed in eight jurisdictions, and that nobody reads the registers together.
What we found. This site holds no source for a count of eight, and a claim that nobody does something cannot be tested from here. Two claims of that kind (Nevada and Connecticut) had already been disproved by one request each.
Now. The page says a handful of states, and no longer states a count or an absence.
2026-10-29 · Interpretation · /who-is-entering-community-management
It said. That 2026's turn down was "a collection artefact rather than a fall in entry", and that the climb was "steady rather than a single step", which looked like "sustained demand rather than a one-off regulatory event".
What we found. Counted month by month from first-issue dates, 2025 was weaker than 2024 from January to May and far stronger from June, and over the same dates 2026 is behind 2025. The register gives no cause, so the "regulatory event" remark was unsupported in either direction.
Now. The page compares the same dates in each year and shows every month, with each figure computed from the data.
2026-10-29 · Method · /methodology
It said. That the record is built from two state registers, and that entry and exit rates are derived by comparing weekly snapshots.
What we found. The site reads Florida and Virginia in full and counts Connecticut and Nevada in aggregate. Snapshots show entrants only: the Florida file lists almost only managers marked active, so a manager who leaves drops out of the file and nothing records when or why.
Now. The methodology page names all four registers and no longer claims exit rates.
2026-10-29 · Regime · /comparing-manager-licences-across-states
It said. That Connecticut licenses both the person and the business under one credential.
What we found. Connecticut General Statutes chapter 400b defines a community association manager as a natural person, and Public Act 19-177 redefined the term. The register still holds older business records, nearly all inactive.
Now. The page states that Connecticut licenses the person, reports the business records as older ones, and says what the register's reason field does and does not show.
2026-10-29 · Figures · /llms.txt
It said. Counts that lagged the data: 2,001 firms against 2,062 on the register that day, and several others.
What we found. The file was typed by hand and went stale twice in one day.
Now. llms.txt is generated from the data files on every build, so a count in it cannot lag the data.
2026-10-28 · Coverage · /comparing-manager-licences-across-states
It said. That only Florida and Virginia publish their registers as bulk files, and that Nevada offers a lookup and no download.
What we found. Nevada's Real Estate Division publishes a spreadsheet of active community association managers, found with one request.
Now. Nevada is read and counted in aggregate, and graded as counted, not read in full, because the list covers active licences only.
2026-10-28 · Coverage · /comparing-manager-licences-across-states
It said. The same sentence, which left out Connecticut: that only Florida and Virginia publish bulk files.
What we found. Connecticut's Department of Consumer Protection publishes its licence records as open data on data.ct.gov, found with one request.
Now. Connecticut is read and counted in aggregate, with its individual and business records reported separately.
Found something wrong? Write to corrections@dwellory.org with the page and the figure.