If you've ever appraised a site using the government's planning performance statistics, we have some news - and it isn't the good kind.
Every quarter, the Ministry of Housing, Communities, and Local Government publishes figures showing the English planning system in a pretty flattering light: roughly 91.5% of major applications decided on time, a number that's been creeping upwards for a while now. Read that in isolation and you'd assume planning risk is a solved problem. Model your appraisal around a 13-week determination period, add a bit of contingency, and crack on.
We'd love to tell you that's a safe way to run the numbers - but it isn't.
Here's the bit the headline figure doesn't spell out: "on time" doesn't mean what most of us assume it means. The government metric measures whether a decision was made within an agreed timeframe, not the statutory one. That agreed timeframe can be extended, quite legitimately, through an Extension of Time (EoT) agreement or a Planning Performance Agreement (PPA). Both applicant and council sign up to a new deadline, and provided the decision lands before that new date, it counts as "on time" in the official stats.
Which means a determination that took the best part of a year can be recorded exactly the same way as one that took twelve weeks. The clock didn't stop, it just got a new number painted on it.
We're not here to say EoTs and PPAs are some kind of stitch-up. They exist for sensible reasons: they save councils from triggering a deemed refusal, and they save applicants the cost and faff of an appeal. But if you're the one financing a site, the agreed deadline isn't really the number you care about. What you care about is how long it actually takes, start to finish, regardless of what anyone agreed to along the way.
That's the number LandTech's data tracks. And it tells a rather different story: an average on-time rate of 48.7% for the same local planning authorities and periods the government reports as 91.5%. A 43 percentage-point gap, on the same decisions.
Strip out the agreed-deadline adjustment altogether, and the national mean elapsed time for major planning applications across England was 34.1 weeks in 2024-25. That's not a typo; it's 2.6 times the 13-week statutory target, and it means that among LPAs where the government reports a full 100% on-time rate, the real average determination time was more than double what the statutory clock allows for.
To be fair to the system, there's a genuine improvement story buried in the quarterly data too - it's not all doom. But even at its best point in the dataset, performance was still running at 2.5 times the statutory threshold. Better, yes. Sorted, no.
Here's where it gets properly interesting, and where a lot of received wisdom falls apart. Ask most developers which region has the worst planning bottleneck in England, and London tends to come up first: complexity, political sensitivity, endless consultation. It's practically an industry in-joke.
The data doesn't back it up. London is, in fact, the fastest region in the country for major application determination. The regions actually carrying the greatest elapsed-time risk are the ones that rarely get flagged as problem areas at all, and the gap between the fastest and slowest region is close to a factor of two.
It gets more granular still once you drop down to individual council level. Neighbouring authorities, sometimes just a borough boundary apart, can differ by tens of weeks on average determination time, with no change in the type of scheme going through. And speed on its own doesn't tell the whole story either: some of the slowest authorities have grant rates north of 95%, while some of the fastest refuse more often than you'd expect. Both numbers matter to your risk position, and looking at only one of them is how appraisals get quietly optimistic.
If you're still building development appraisals around a 13-week planning assumption, this isn't really a "watch this space" issue, it's already costing you. It's not a conservative estimate. It's a significant underestimate of what's actually happening, and the further underneath the true elapsed time your base case sits, the more exposed your financing costs are to slippage nobody budgeted for.
The fix isn't complicated in principle: model planning duration as a range specific to the LPA you're dealing with, not a national average and definitely not the statutory minimum. In practice, that means knowing the actual mean elapsed time, the grant rate, and how your target authority compares with its neighbours, before heads of terms are agreed, not after.
Every figure in this post comes from the same underlying dataset. So does the data you actually need to price planning risk properly: the real mean elapsed time for any specific LPA, its grant rate, how it stacks up against its neighbours, and a ready-built three-scenario financing framework that turns "planning might overrun" into an actual number in pounds.
That's what's in the full report, not a summary of it.
Download The Real Cost of Planning Delays and get:
Your next site appraisal is only as good as the assumptions underneath it. Get the ones that actually hold up.