How an audit is built, and why you can check it
Every audit follows the same seven-part structure and the same verification rules. The point is that a staffer who disagrees can find the same record and see for themselves.
The Question
One civic question with a verifiable answer, not a topic. Good: does this agency have a current engineering study for the limit it set? Not: are speed limits too low?
The Standard
What should exist if the system were working, quoted from the statute, regulation, or recognized best practice, with the primary citation.
The Gap
What the public record actually shows, including what was missing and what had to be requested to find it.
The Comparison
How peer jurisdictions answer the same question under the same rule. County against county, state against state.
The Counterargument
The strongest case that the current system is right, stated fairly. If the data supports the agency, the audit says so.
The Connection
Which filed bills the finding speaks to, and who sponsors them. The audit takes no position; it records what each would change.
The Ask
Answer the question. The audit publishes either way, and points the reader to the same record so they can check it.
The verification rules
No claim is asserted that is only inferred. Every figure traces to a primary source, and every source is archived the moment it is accessed, so a dead link later does not erase the evidence.
Primary sources first
Government records and statute outrank reporting, which outranks advocacy. When sources disagree, the audit shows both and names which is closer to the record.
Two independent confirmations
A surprising figure is not published on a single source. The researcher who finds a number is not the one who confirms it.
The counterargument is required
An audit that cannot state the strongest case against its own finding is not finished.
How the county census was made
The map on the home page is the first layer of this work: a single-day census of all 46 county councils, asking one question of each, can a resident find the record of its meetings?
Each county was classified only against an evidence URL, never a self-description, because official self-descriptions of transparency repeatedly turned out to be wrong. Where a county could not be verified from the outside, it was marked unknown rather than guessed. The census is internal and pending a verification pass before any of it is published.