What changed, and when.
The dataset is refreshed quarterly, which means the numbers on this site move. This page records what moved and why, including the corrections. An evidence base that quietly revises itself is not one you can check, so the revisions are published too.
Every decided residential application of one to nine units across all 33 London planning authorities, coded to a single method. Decisions run from January 2023.
Changed in this release
- Refusal-reason corpus repaired. A regeneration on 15 July 2026 had cut the coded refusal corpus to roughly a third of its size. It has been rebuilt to 13,520 coded reason paragraphs across 4,500 refused schemes. The repair was a merge rather than a rollback, so applications harvested since June are retained.
- Coverage note added for Haringey. Its refusal-reason extraction sits outside the repaired set and is not directly comparable with the other 32 authorities.
Known limits in this release
- The most recent months are incomplete. Decisions reach the London Datahub with a lag, so monthly counts fall away after March 2026. Recent quarters will fill in rather than stay as published.
- Refusal-reason classification is being rebuilt. The design share quoted elsewhere on this site, 35.5% of first-listed reasons, comes from the June 2026 classification of the corpus. That classification did not survive the July regeneration and is being regenerated. Until it is, treat the design share as provisional.
- A rebuild reconciliation is outstanding. The current rebuild counts 15,534 applications and 12,493 decided, against the 15,533 and 12,458 published across this site. The difference is small and its cause is known. Rather than restate figures piecemeal, the site, the white paper and the claims register are re-stamped together at a refresh.
Earlier refreshes were not itemised, so this record starts here rather than pretending to a history it cannot evidence. Two corrections from that period are worth stating, because both changed published numbers.
- Refusal reasons were reclassified. The original classifier defaulted to design, which inflated it as a refusal category. Reclassifying the underlying reason text moved design from roughly 55% of primary reasons to 35.5%, and revealed policy and transport as much larger categories than previously reported.
- Non-development applications were removed from the cohort. Condition discharges, non-material amendments and lawful development certificates had leaked into the small-site cohort. Excluding them changed several borough approval rates.
Each borough is re-pulled from the Planning London Datahub, re-harvested for newly refused applications, classified, and regenerated. Published figures are checked against the rebuilt data before anything is restated, and where a published number cannot be reproduced it is flagged here rather than quietly changed to match.
If you find something on this site that the record does not support, tell me. Corrections appear in the next release note with the finding attributed.
