There's a particular kind of frustration that belongs specifically to property people. You drive past a plot of land you've had your eye on for months, only to notice a weathered planning notice zip-tied to a lamppost — a notice that's been there for three weeks, announcing a decision you needed to know about the day it was filed, not the day you happened to drive past. UK planning data has always been technically public. It has never, in any meaningful sense, been visible.
Planning Radar AI was built directly into that gap, at a moment when the underlying data was finally starting to change. Individual UK councils had begun publishing genuinely structured open data — machine-readable formats, geocoded records, the kind of information that used to live only in scanned PDFs and searchable-in-theory-only portals. A wave of independent analysis had even managed to compile millions of planning decisions across years of records, surfacing something genuinely useful: real, measurable variance between how different councils actually behave, some approving well over a fifth more applications than others, some taking dramatically longer to decide. That variance is valuable information. It was also, until now, buried behind hundreds of separate council websites, each with its own quirks, none of them talking to each other.
The instinct behind building this actor came from recognizing what the existing tools in this space were actually offering, once you looked closely. A search of the marketplace turned up exactly one meaningful competitor — and it had essentially no users at all. Not because nobody needed planning intelligence. Because what existed was raw extraction: a structured version of a planning search, technically more convenient than a council website, but not meaningfully more useful. It answered "what does this record say" without ever answering the question a property investor or developer actually has, which is closer to "should I care about this, and why."
That distinction shaped the entire build. Rather than treating planning data as a search problem, the actor was built around a different question entirely: what's changing, near a place I actually care about, and does it matter? Point it at a postcode and a radius, and it doesn't return a list of applications sorted by date. It geocodes every record, calculates real distance from the point you're watching, and — critically — resolves developer identities across applications, so the same development company doesn't quietly show up as three slightly different name variants across three separate filings. That resolution turns out to be the single most valuable piece of the whole system. A developer who has submitted five applications in a specific area over the past year is telling you something meaningful about where they believe growth is heading. A generic planning search will never surface that pattern, because it was never designed to look for it.
The genuinely differentiating feature, though, is the plain-English layer sitting on top of all that structured data. Rather than returning a raw status change — approved, refused, under review — the actor generates the kind of summary a knowledgeable colleague might send you: three new residential applications appeared within a mile of a specific postcode this week, and one of them comes from a developer who has filed five applications in the region already. That sentence does more work for a busy investor than a spreadsheet of fifteen fields ever could, because it tells you not just what happened, but why it might be worth ten minutes of your attention.
The people who reach for this tool are exactly the ones who used to rely on a lamppost notice or a lucky drive-by — property investors watching a specific area for early signals, developers trying to understand a competitor's expansion pattern before it becomes obvious to everyone, architects and planning consultants who need to track activity across multiple council areas without manually checking each one every week. For all of them, the value isn't just faster access to public records. It's the difference between planning data as an archive you search when you remember to, and planning data as a living signal that reaches you the moment something near you actually changes.
There's something quietly fitting about a tool built to watch permission itself — the earliest, most upstream signal in the entire property lifecycle, months or years before a listing ever appears on a property portal. Long before a house is built, before it's marketed, before anyone lists it for sale, someone filed a piece of paper with a council saying what they intended to do with a piece of land. That paper has always mattered enormously to the right person at the right moment. This actor exists so that person finally finds out about it when it's filed, not when they happen to notice the notice on their way past.