How we track the numbers

We track explicit numerical commitments: promises that contain a specific figure a member of the public could later check against official data. A pledge to reduce unemployment to a stated percentage, to build a stated number of houses, or to invest a stated amount is trackable. A general aspiration with no number is not recorded here.

We do not decide whether a government succeeded or failed. There are no scores, grades, ratings, or pass/fail marks anywhere on this site, and no red or green performance colours. We show the promised number, the number at the start of the term, the latest official number, and the change between them. What those numbers mean is for you to judge.

This is the latest comparable official measurement available at the beginning of the relevant government term. It is a baseline for reading the change, not a target and not a claim about cause. Where a commitment is cumulative (for example, a number of houses to be built over a term), the starting point is zero at the beginning of the term.

This is the most recent official measurement we have matched to the commitment. Every latest figure carries its data period, so you can see how current it is. Numbers are revised by the bodies that publish them; when a revised figure is released, we record the update.

Not all official numbers are equal. Where more than one official figure exists, we prefer, in order: independent statutory statistics (such as the national statistics agency), then official financial and audit records, then government administrative data, then other primary official sources. Every observation names the source it came from.

Some numbers are produced by an independent statistics or audit body; others are reported by the department responsible for the commitment. Both can be official. We label which is which so you can weigh a self-reported figure differently from an independently measured one if you choose.

A promise is matched to the official indicator that measures it. Most matches are direct. Some are proxies or components, because no single official series measures the exact thing promised. Where a match is not exact, that is disclosed, and where no reliable official dataset exists, the promise is shown with its wording but marked “official tracking data unavailable” rather than filled with a guess.

DidThey collects, extracts and matches data using automated tools. Automation makes broad, timely coverage possible, but it can introduce errors in extraction or matching. Treat the figures here as a starting point and confirm anything important against the linked primary source.

When we find an error, or one is reported to us, we correct it and record that a change was made. The record is meant to be auditable: observations are kept rather than quietly overwritten, so the history of what was shown remains visible.

DidThey is nonpartisan. We track commitments from any party that makes them, apply the same treatment to all, and do not campaign for or against anyone. The absence of a verdict is deliberate: our role is to lay out promises, numbers and sources cleanly, not to tell you what to conclude.

We show raw official numbers rather than derived indexes or composite “delivery” scores. A composite score hides the choices that went into it. A raw number, next to what was promised and where it started, with its source attached, lets you do your own reasoning without a layer of our opinion in between.