Methodology

How each figure on this site is produced, where it comes from, and when we refuse to publish one.

Where the data comes from

Every borough figure is computed from planning applications published on each London borough’s statutory planning register. We collect applications and decisions continuously, normalise them into a single structure, and compute a monthly statistics snapshot. Each page shows the snapshot period and the date it was computed.

Metric definitions

• Approval rate: approvals as a share of all applications with a published decision in the trailing 12 months (refusals, withdrawals and other outcomes count in the denominator).
• Decision speed: the median calendar days from a valid application to a decision, per application tier. The statutory targets are 56 days for householder and minor applications and 91 days for majors.
• Householder applications: extensions, loft conversions, outbuildings and similar applications by homeowners, as classified by the deciding borough.
• Major schemes: applications classified as major development (10+ dwellings or equivalent floorspace thresholds), read from the development type each borough publishes. We show a borough’s count only when at least 80% of its decided applications carry a development type; many boroughs publish none, so there is no major-schemes league table.
• Live caseload (“waiting for a decision”): applications received in the last 24 months that have no published decision.
• Band: 55% approval rate plus 45% the share of decisions made within the statutory time (weighted by volume across the tiers with enough decisions), then each borough is placed in fifths across London: A is the top fifth, E the bottom fifth. A band describes a borough’s decisions as a whole, never how a particular scheme will fare.

What we refuse to publish

A ranking is only honest if its denominator can support it. Any metric computed from fewer than 20 decisions in the window is shown as “too few decisions” and left out of rankings. A borough whose published data shows fewer than 20 decisions in the window has none of its decision figures published and appears in no ranking.

A borough whose published data records no refusals at all across 100 or more decisions is not a council that refuses nothing: its register words decisions in a way we don’t yet read. Its approval and refusal rates are withheld and it is left out of the approval rankings until we can read them; its decision counts and decision times still stand.

When the newest application in a borough’s published data is more than 60 days old, its feed has stopped updating: we mark its application counts and live caseload as not current rather than publish them.

How we measure coverage

The homepage says how much of London’s planning record the Reading Room can search. Each figure is read live from our own API, through the same operations any API customer calls, and dated with the day it was measured. If a measurement is unavailable, its sentence is left out rather than guessed.

Applications we hold documents for

London planning applications with a valid date (or, failing that, a received date) in the last ten years that have at least one document downloaded from the council’s register. From the get_capture_depth operation, London, 10-year window, measured daily.

Documents the Reading Room can search

Of the documents downloaded for those applications, the ones whose text has been extracted and indexed for search. Drawings and photographs are never indexed, and newer documents wait in a queue, so this is roughly half of what we hold. From the get_indexing_backlog operation, London, 10-year window, measured daily.

The past week

London applications registered in the last seven days, and how many of them we already hold documents for. It shows whether the archive is keeping up. From get_capture_depth, London, 7-day window.

We don’t publish a completeness percentage on the homepage: the register we would divide by counts applications differently from the councils’ own portals, so the ratio can exceed 100% and would mislead.

Caveats

• Boroughs occasionally backfill or amend decisions; a snapshot reflects the register as captured at computation time.
• Classification (householder, minor, major) follows each borough’s own published categorisation.
• Figures describe what has been decided. They are no substitute for pre-application advice; always check with the borough.

Questions about the methodology? See about us or the questions and answers.