Most planning arguments are decided by whoever sounds most certain.

They don’t have to be. A great many planning questions that get settled by assertion are, underneath, empirical, and the record needed to answer them already exists. Collaboration here has one purpose: to test assumptions about how the planning system actually behaves. If you are working on a question like that, I would like to hear about it.

What data can actually settle

Planning generates an extraordinary volume of evidence and then largely declines to use it. Every application produces a decision, a date, a reason, and a location. Multiply that across a city and you have a record of how policy behaves once it meets practice, which is a different thing from what the policy says and usually more useful.

The questions this can reach are broader than they first appear. Not because data answers everything, but because so many live disputes turn on a factual claim nobody has checked.

Which is why this is not consultancy, and not commissioned research either. It is closer to shared inquiry. Some questions need local knowledge, some need domain expertise, and some need structured evidence. The useful work almost always needs all three, and I only have one of them.

Does this instrument work?
Design codes, thresholds, density standards and permitted development rights all assume a behavioural response. Where the instrument already exists somewhere, the response is measurable rather than hypothetical.
Is this borough unusual?
Comparison is the whole point of a common method. An approval rate on its own says very little; the same rate set against thirty-two comparable authorities says a great deal.
Where is delivery actually lost?
Between allocation and consent, sites fall out for reasons that are recorded one decision at a time and rarely aggregated. The pattern is usually narrower, and more fixable, than the anecdote suggests.
What changed after the change?
Policy interventions have dates. Where there is a before and an after, and comparable authorities that did not intervene, the effect can be estimated honestly, including when the honest answer is that nothing happened.
Is the assumption in the evidence base true?
Viability models, trajectories and needs assessments all carry assumptions about consenting behaviour. Those assumptions can be checked against what was actually consented.
Something else entirely
This list is not meant to be exhaustive, and the most interesting questions have tended to arrive from people who were not thinking in terms of data at all.

Questions already on the table

These are the ones I think matter, and several are already partly answered in the published research. None of them are settled. If you are working on one of them, we are working on the same thing.

  • Do policy thresholds change what applicants propose, or only what they are allowed to build?
  • How consistent are planning outcomes between boroughs, once you account for what is being applied for?
  • What actually causes refusal, as against what everyone assumes causes refusal?
  • Do design codes change outcomes, and how long would an effect take to show up?
  • What does conservation area status cost in delivery, and where does that cost fall?
  • What changes after a policy changes, in the authorities that changed it and the ones that did not?
  • What does the time a decision takes tell you about the decision itself?

How I would like to work

There’s no defined customer here, and no product to sell you. Perfect Scale is a dataset and a method, and it is more useful in other people’s hands than in mine alone. None of what follows needs a budget to start a conversation.

Local authorities
Evidence for local plans, monitoring reports and examination. Including the uncomfortable version, where the record does not support the position you were preparing to defend.
Design codes and guidance
Understanding how a rule influences what gets proposed and what gets approved, before it is adopted and afterwards.
Researchers and universities
Access to structured planning evidence, and to the method behind it, for work that is going to be peer reviewed rather than published as a claim.
Housing organisations and think tanks
Questions about delivery, viability and whether a policy is doing what it was introduced to do.
Journalists
Interrogating a planning claim against the record. I will also tell you when the data does not support the story, which is the more common outcome.
Practices and people building adjacent things
Architects, planners and anyone constructing something the evidence base could sit underneath. The most interesting requests have come from here.

Three things I hold to, so they are worth saying up front. The method doesn’t change to suit the answer, and it is identical in every borough, so the result is what it is. Limits are published alongside findings, including when the data cannot support the claim someone was hoping for. And any commercial or authorship relationship is disclosed on the work it touches.

If a collaboration would only work by relaxing one of those, it’s better we find that out early.

The planning system produces an enormous amount of evidence about itself, and most of it stays fragmented across thirty-three registers. Turning that record into something people can actually reason with is not a thing anyone does alone. Good evidence rarely is.

Start a conversation

No form-filling ritual. Tell me roughly what you are trying to work out.

Goes straight to abre@perfect-scale.co.uk. No list, no newsletter, no automated follow-up sequence. Your address is used to reply to you and nothing else. See the privacy notice.