How these figures are produced

Every number on this site traces back to a document a public body itself published. This page says where each one comes from, what it does not cover, and what has been got wrong.

Where the data comes from

Councils are required to publish every payment they make above a threshold, usually £500. Those files are the raw material. Each council's own published files are taken as they are, read mechanically, and no figure is retyped or re-keyed by hand.

Budget and outturn figures come from the returns councils submit to central government. Every page names the source it used and links to it.

What spending data does not show

Payments above the threshold are not the whole of a council's budget. Staff pay, most small purchases and internal transfers are outside the published files. A total on this site is a total of what was published, and every page showing one says so beside it. It should never be read as what the council spends.

Some councils redact supplier names, usually where the supplier is an individual. Every such redaction is mirrored rather than defeated. Payments to individuals are counted in totals and never listed, and no individual is named anywhere on this site.

How the data is processed

Published files arrive in different shapes: different column names, different date formats, some as spreadsheets and some as PDFs. Processing maps each council's columns onto one common set of fields. The mapping is checked mechanically by reconciling the processed totals back to the council's own published outturn return, or to its published budget where the outturn return is not yet due, so a mis-read column shows up as a number that does not add up rather than as a plausible wrong figure.

A council year that does not reconcile shows no total. That is not a styling choice: the published summary for such a council carries no headline block at all, so there is nothing to print even by mistake. Its page states what is held and why no figure is shown.

Two things this site will not do

It does not estimate. If a figure is not in a source document it is left out, and a section with no figure behind it is absent rather than filled with a placeholder.

It does not rank councils against each other by what they spend. The periods held differ from council to council, so a table ordered by total would be ordering by how much data happens to be held. Councils are compared only within a type and only when every council in the comparison has a complete, reconciled year, which is a comparison this site does not yet publish.

Known limitations

  • Some councils' published files contain the same rows twice. Where that is unresolved, estate-wide transaction counts are shown as an upper bound rather than a total.
  • Councils that publish only as PDFs are harder to process reliably, so some are recorded as held but not yet published.
  • Coverage is uneven by the design of the disclosure regime, not by choice. Spending data is currently held for 238 of the 317 principal councils in England, and 251 of the 333 bodies with a page here have a published total. The coverage page lists every body and what is held for it.

Corrections

No corrections have been issued. When a figure on this site is found to be wrong it is corrected quickly and recorded here, with the original figure, the corrected figure and the date. Corrections are not quietly edited away.

If a figure here and a council's own published file disagree, the council's file is right and this site is wrong. Tell us at info@aidoge.co.uk and it will be corrected and recorded.

Editorial policy, right of reply, complaints and data protection