Our approach to collecting and analysing planning decision data across England, Scotland, Wales and Northern Ireland.
PlanningLens indexes 309 planning authorities, covering 3,824,659 planning decisions in total. Together these authorities cover more than 83% of the UK population, including every planning authority in Northern Ireland. That makes it one of the largest structured planning decision databases available to homeowners in the UK. We add new councils regularly and update existing data on an ongoing basis. The raw index is what the postcode checker, the API and evidence packs draw on; aggregate statistics apply two further quality gates on top of it, so the figures quoted on different pages describe deliberately different tiers of the same dataset:
Raw index — 309 planning authorities, 302 with a published council page. The 309 comprise 291 council entries covering 295 local planning authorities (4 shared planning services are listed once), 5 predecessor authorities abolished in 2023 whose historic records are retained, and 13 national-park and development-corporation authorities. 7 entries are held below the 300-decision publication threshold and have no page: Bedford, Cairngorms NP, Cherwell, South Hams, Teignbridge, West Devon, Wychavon. Measured against the 361 ordinary UK planning authorities on 1 September 2026: 304 held, 57 not yet indexed. 255 published authorities have data within 30 days of this build; each council page states the date its data runs to.
Quality tier — 302 planning authorities, 3,823,197 decisions. A council enters aggregate statistics once its portal data shows an approval rate of at least 60% and at least 300 decided applications. These thresholds screen out councils whose decision wording we cannot yet classify reliably, or whose sample is too small to be meaningful, so that a portal quirk never masquerades as a genuinely strict council. National approval rates cluster in the high eighties, so a portal reading far below that range almost always signals decision wording our classifier cannot yet read, not a council refusing half of everything it sees; the threshold sits well below the range real councils produce.
Ranked tier — 296 planning authorities, 3,740,318 decisions. 6 councils pass the numeric thresholds but are excluded from the ranked league table and national insight pages because of known portal-data distortions such as bulk imports and portal-migration artefacts: Blackburn with Darwen, Buckinghamshire, Cheshire West & Chester, Chesterfield, Medway, Stroud. Their individual council pages remain live, with every figure traceable to the council's own portal record.
Our automated systems read publicly available planning decision records from council portals. For each application, we extract the reference number, address, ward, proposal description, application type, and decision outcome. Northern Ireland records combine the regional planning portal with the decision lists published by the Department for Infrastructure. All data comes from publicly available council records; we do not access any private or restricted information.
Scheduled collection jobs run on a rolling schedule, each covering one family of council planning portals. Each job walks a council's newly determined applications and stops when it reaches decisions already on record, so the dataset grows incrementally rather than being rebuilt. Most new decisions appear on PlanningLens within a couple of days of the council publishing them; a small number of portal families update on a slower cycle, and any council whose feed goes quiet is flagged by an automated health check.
When a collection gap is known, it is listed here rather than left for readers to find. Collection has stopped for 16 councils on scrape evidence, checked 10 September 2026: Brighton and Hove, Somerset, Scarborough, Exmoor National Park, Craven, North East Derbyshire, Telford and Wrekin, Leicester, Barnsley, Bath and North East Somerset, Kirklees, Rotherham, Burnley, Liverpool, Brent and Birmingham. Their council pages remain live and show the records held and the date of the last decision on record. A further 57 council stores carry no collection timestamp, so their currency cannot yet be verified; their pages show the last decision date instead. This notice is typed and dated. It will be replaced by a list derived from each store's own timestamps.
Each planning decision is classified based on the decision wording used by the council:
Approved, includes decisions worded as granted, approved, permitted, consent given, prior approval given, and similar positive outcomes.
Refused, includes decisions worded as refused, rejected, declined, dismissed, and similar negative outcomes.
Excluded, withdrawn applications, applications with no recorded decision, and procedural outcomes. These are not included in approval rate calculations.
Approval rate = approved applications divided by total decided applications, multiplied by 100. Only applications with a clear approved or refused outcome are included.
Rates are calculated at three levels: council-wide, by ward, and by application type. The approval likelihood score shown on our homepage uses a weighted blend of these three factors, with application type weighted most heavily.
Comparable decisions shown on council pages and in the property report are filtered to residential planning applications only. We exclude tree works, discharge of conditions, telecoms applications, signage, and other non-residential application types to ensure relevance to homeowners.
Decision times measure the gap between the date a council validates an application and the date it issues a decision. The decision-time rankings use householder applications as defined in the 2015 Development Management Procedure Order: the extensions, loft conversions and other small works that make up just over half of the roughly 300,000 applications decided in England each year. The clock runs from validation to decision, and decisions taking over 364 days are excluded as records more likely to reflect stalled cases than live processing. No separate measure was invented for our research reports; they apply this same yardstick, which predates anything it is used to measure.
Extensions of time agreed between applicant and council do not stop the clock. That is a deliberate departure from the official performance measure, which counts a decision as on time if it met the statutory deadline or a deadline the two sides agreed to move. Both are honest answers to different questions: the official measure asks whether the council met a deadline it was allowed to renegotiate, ours asks how long the applicant waited once their application was valid. The two diverge because an agreed extension moves the deadline the official measure counts against. In the year ending March 2026, official statistics recorded 93% of householder decisions in England as made within eight weeks or the agreed time; measured on the wall clock, the typical wait ran well past eight weeks.
The headline figure is the median: it moves only when the typical application moves, and cannot be dragged by a handful of stalled cases. The share of decisions made within 8 weeks is published beside it, because that is the number the statutory target speaks to. Northern Ireland operates a 15-week statutory regime for local applications, so its councils are measured against that target rather than the English 8 weeks.
Council pages carry a second, simpler figure: the median time from validation to decision across every application the council decided on or after 1 January 2020, all application types, shown only where at least 20 decisions carry both dates. Where a council’s recorded dates are known to be unreliable, that figure is withheld until the records are corrected, and the page says so.
No dataset figure on PlanningLens is typed by hand. Every count, percentage and ranking on this site is generated from the same underlying dataset, and an automated audit checks the generated pages before any release: a build containing a stale or hand-authored figure cannot reach the live site. Published totals are floor-rounded, so a headline figure is never an overstatement of what the dataset contains. Where a page cites an external statistic for context, such as a government figure, the source and vintage are given in the sources section below.
Where possible, planning applications are geocoded using postcode data from postcodes.io. This enables distance-based analysis in the property report, showing approval rates within 500m, 1km, and 2km of your property. Not all councils publish postcodes for every application, so coverage varies.
Planning statistics are based on historical data and do not predict future outcomes. Every application is assessed on its own merits by the local planning authority. Approval rates can vary significantly between wards, application types, and time periods.
Approval rates exclude withdrawn applications. Where applicants withdraw on officer advice rather than take a refusal, a council approval rate will read higher than its underlying strictness; no portal records the advice behind a withdrawal, so no dataset built from portal records can fully correct for this.
PlanningLens provides data analysis for informational purposes only. We are not a planning consultancy and do not provide formal planning advice. If you need professional planning guidance, we recommend consulting a qualified planning consultant or contacting your local planning authority.
Our council-level planning data is available as a structured JSON file for researchers, journalists, and developers. It includes approval rates, ward-level breakdowns, and application type statistics for every council we track.
Download: councils.json, 302 planning authorities, 3,823,197 decisions. Attribution required.
Figures derived from our own dataset are traceable to the council portal records they came from. Contextual definitions and national statistics on this page draw on: the Town and Country Planning (Development Management Procedure) (England) Order 2015, for the householder application class; the MHCLG quarterly release Planning Applications in England, year ending March 2026, for national volumes, the householder share and the official in-time measure; Department for Infrastructure (Northern Ireland) statistics, for the 15 week regime and Northern Ireland decision lists; ONS mid year population estimates, for population coverage; and postcodes.io, for geocoding.
Questions about our methodology or data? Email hello@planninglens.co.uk. Corrections are welcome and reach the founder directly, not a press office. Errors are fixed on the page, with a note where the change is material.