How we count
Every number, and what it rests on
We publish counts and rates from councils' own planning registers. Here is where they come from, what we count and leave out, the least data we will publish a figure on, and what we never do.
Where it comes from
We read the registers ourselves
Applications and decisions
Every application, decision and status on this site was read by us from the council's own public planning register. We do not buy it, and we do not take it from another company or an aggregator. Each application links back to the council's own record, which is the final word. Councils we do not read yet are listed as such on the coverage page.
Designations at a postcode
Conservation areas, listed buildings, green belt, flood zones and similar come from the government's open planning data (planning.data.gov.uk), under the Open Government Licence v3.0. When that service does not answer, we say so and never report “none found”.
Our figures describe the applications we have read, not a council's full archive. We always show how many decisions a figure rests on and the dates it covers.
What counts
A decision on a scheme, and nothing else
An approval rate is a statement about how a council treats schemes. So it counts decisions on schemes that were approved or refused, over the last two years. Withdrawn applications and outcomes we cannot read as approved or refused are left out. Filings that are not a decision on a scheme are still stored and still shown, but never move a rate:
Tree works and notices
Felling, pruning and tree preservation notices.
Discharge of conditions
Sign-offs of conditions on a permission already granted.
Non-material amendments
Small changes to a permission already granted, and variations of conditions.
Lawful development certificates
A certificate that something is already lawful, not a decision on a scheme.
Prior approvals and notifications
Permitted-development checks and notifications.
Advertisement consent
Signs and hoardings.
Pre-application advice
Advice before an application is made.
Hazardous substances, screening and scoping opinions
Specialist consents and opinions.
The council's own “application type” is the strongest signal, and the proposal wording backs it up. A scheme that merely mentions a tree (“erection of a dwelling, removing one tree”) stays a scheme. One set of rules decides this for every figure we publish. A rate of 97% or more on 30 or more decisions is held back for checking and not featured, because a rate that high usually means filings that are not decisions have crept in.
The minimum sample
We would rather show nothing than a figure on too little
| The figure | Needs at least | How it behaves |
|---|---|---|
| A council's approval rate | 10 decided schemes | Below that the council shows its history as "too little" and borrows no figure. |
| A rate for one kind of scheme at a council | 10 of that kind | Each kind has to reach the minimum on its own. |
| The approval figure around a postcode | 10 decided schemes | The circle starts at 500 m and widens to 1 km, then 2 km, only as far as it needs to. It says how far it went. Under 10 within 2 km we show no figure. |
| A firm's page | 5 decided schemes | A firm below that has no page. A rate for a firm at one council needs 5 there. |
| The typical days to a decision | 10 decisions with usable dates | Days run from validation (or receipt) to decision; a decision on the same day is treated as a placeholder, not a speed. |
| A reason-for-refusal pattern | 5 decision notices read | We say how many notices we have read. |
What we never do
Lines we do not cross
We do not predict.
A figure describes how past decisions went. It is not a prediction for any application, and nothing here gives planning or legal advice.
No one can pay to rank.
Pages for planning firms are built from decided applications. A firm cannot buy a better place on them.
We do not guess at gaps.
Where we hold too little, we say so. Where a service did not answer, we say that, not "nothing found".
AI is checked against the figures.
Plain-English summaries and short cards may be worded by AI, but a line is thrown away if it uses a number that is not one of its own figures, praises or blames a council or firm, or predicts an outcome.
Think a figure is wrong?
Tell us the page and what you expected. We check it against the council's own register and correct it.