🏗️ UK Planning Applications Scraper — every council, one feed

Get UK planning applications from every planning authority in the country in a single run: site address with postcode and GPS, the proposal text, application type and size, status, decision and dates — plus the agent firm (architect or planning consultant) behind the application.

No council-by-council setup. No login. No API key.

What does this actor do?

• 🏠 address, postcode, ward, parish, latitude, longitude — where the work is
• 📝 description — the proposal in the applicant's own words
• 🗂️ applicationType (Full, Outline, Trees, Listed Building…), applicationSize (Small / Medium / Large)
• 🚦 status, decision, decidedBy, decidedDate
• 📅 startDate, receivedDate, validatedDate, consultationStart, targetDecisionDate
• 🏢 agentCompany, agentAddress — the firm that filed it
• 🏛️ council and deep links: planitUrl, councilUrl, documentsUrl, commentUrl

Example input

{
  "authorities": ["Cornwall", "Bristol"],
  "searchText": "extension",
  "recentDays": 7,
  "requireAgent": true,
  "maxItems": 1000
}

Run that daily and you have a live feed of new work in your patch, filtered to the rows that name a firm you can contact.

Read this first: the source redacts personal names

This is the most important thing about the data, and it is measured, not assumed. On a 100-record sample:

Field	Present	Actually usable
applicant_name (source field)	90%	0% — every value is the literal string "See source"
agent_name (source field)	69%	0% — same
case_officer (source field)	66%	0% — same
agentCompany	—	38% real
agentAddress	—	59% real
applicantAddress	—	24% real

The aggregator strips personal names and keeps organisations and addresses. So this actor does not pretend to give you a named individual — it gives you the firm. Those placeholder strings are returned as null rather than as text, so the fill rate you see is the truth, and the run log tells you how many were redacted.

If you need the named applicant, the row carries councilUrl — a deep link straight to that application on the council's own portal.

Field coverage — measured

From a real 120-application run (last 7 days, all councils):

Field	Fill
reference, council, address, status, startDate, planitUrl, councilUrl, lastChanged	100%
applicationSubType 98% · description 97% · latitude/longitude 95% · validatedDate 94%	95–98%
receivedDate 92% · applicationSize 88% · postcode 85% · ward 83% · documentCount 81%	80–92%
agentAddress 59% · agentCompany 38%	the lead fields
decision 10% · decidedDate 2%	on a recent-7-days run — see below

Decision fields look empty on a recent run because recent applications have not been decided yet. Filtered to applicationState: "Permitted" over June–July, the same fields read decision 90%, decidedDate 96%, decidedBy 80%. Nothing is broken; you are choosing which half of the pipeline to look at.

dwellings is ~2% because only major housing schemes report a unit count.

Who is this for?

Construction suppliers and trades prospecting new work, architects and surveyors tracking competitors, conveyancers, property investors, and anyone who needs a national view without wiring up 400 council portals.

Honest limits

• The aggregator rate-limits. It replies with an explicit "try again in N seconds" and this actor waits exactly that long rather than hammering. A very large run will pause occasionally — that is the actor being polite, not stuck.
• Coverage follows the aggregator, which scrapes council portals on its own schedule. A brand new application may take a day to appear.
• Council fields vary. Some publish a case officer, ward, constraint count and document links; others publish the bare minimum. That is the council's choice, not a gap here.
• Personal data. Agent firms and addresses are published as part of the statutory planning record. UK GDPR still applies to what you do with them.

Pricing

What	Price
Per application returned	$0.004

A 5,000-application month costs $20. Filtered rows cost nothing — requireAgent drops non-leads before you are charged.

FAQ

Which councils are covered? All UK planning authorities the aggregator indexes — English, Welsh, Scottish and Northern Irish. Leave authorities empty to search every one at once.

Can I run it daily for new applications? Yes — set recentDays: 1 and schedule it. That is the intended use.

Can I search by postcode? Set postcode plus radiusKm for a radius search.

How do I get only big schemes? Set applicationSize: "Large".

How do I get approvals rather than the live pipeline? Set applicationState: "Permitted" with a date range — that fills the decision fields.

Why is the applicant's name missing? The source redacts personal names. You get the agent firm plus a direct link to the council's own page for the application. See the section above.

Do I need a proxy? No. Leave it empty.

Is there a limit per run? The API returns 100 per page and this actor paginates for you; maxItems: 0 means unlimited.

Is this legal? Planning applications are a statutory public register, published so the public can inspect and comment on them. Only the public aggregator API is read, and its rate limits are honoured.

Data source and fair use

Data comes from PlanIt (planit.org.uk), a public aggregator of UK planning-authority registers. No login or paywall is involved, the documented rate limits are respected, and each row links back to both the aggregator and the originating council page.