The Evidence Loop: What 170,000 Appeals Reveal About Policy

David Robinson, Co-Founder, Planning Appeals

“When the policy citations, inspector reasoning, and outcomes from over 170,000 of those decisions are structured and analysed systematically, something significant emerges: the planning system’s own feedback loop becomes visible.”

England’s planning system issued over 18,500 appeal decisions in 2025, volumes not seen since 2019. Each one is a determination of how planning policy is interpreted in practice: how material considerations are weighed, how local and national policy interact, and where the balance falls. Collectively, these decisions represent one of the richest sources of information about how planning policy actually works on the ground. But until recently, that information has been siloed inside individual decision letters, accessible one case at a time, never at scale.

When the policy citations, inspector reasoning, and outcomes from over 170,000 of those decisions are structured and analysed systematically, something significant emerges: the planning system’s own feedback loop becomes visible.

Green belt: a policy shift, measured in real time

The revised National Planning Policy Framework, published in December 2024, introduced the concept of “grey belt” and signalled a recalibration of how parts of former green belt land is assessed. A question for the sector was whether this would translate into different outcomes on the ground, and how quickly.

The data is unambiguous. In 2025, the allow rate for green belt appeals in England rose to 33%, up from a consistent 25% across 2022, 2023, and 2024. That eight-percentage point increase is the largest single shift in the dataset, and it held across all four quarters of 2025. Over 830 appeal decisions now reference “grey belt” directly, with a non-enforcement allow rate of nearly 36%.

What this tells us is significant for anyone involved in planning policy, whether writing it, applying it, or considering changes. National policy changes are translating into measurably different local outcomes, and at pace. For local authorities reviewing local plans or preparing spatial development strategies, planning appeals represent live evidence of how the relationship between national and local policy is shifting. For practitioners advising clients, the ground has moved.

Procedure matters more than most practitioners realise

A second pattern in the data deserves wider attention. The procedure through which a planning appeal is determined (written representations, hearing, or inquiry) correlates with dramatically different outcomes.

Across all non-enforcement appeals in England: written representations produce a 29% allow rate, hearings 43%, and inquiries 53%. For planning appeals specifically, inquiries reach 55%. The gap is also widening. In 2025, inquiry allow rates hit 72%, the highest figure in the dataset.

Context matters here. The Planning Inspectorate determines procedure based on the complexity and public interest of the case. Householder appeals go to written representations in virtually every instance. Inquiries are reserved for larger, more complex matters, and cases that reach inquiry tend to involve more issues, more detailed reasoning, and often stronger grounds are presented. The data bears this out: decisions from hearings and inquiries are typically significantly longer and more detailed than written representations.

So the higher allow rates of an inquiry procedure are partly a reflection of case selection. But the scale of the gap, and the fact that it is widening year on year, raises questions worth examining. Are there cases going through written representations that would benefit from oral procedure? Is the system allocating its resources in the way that best serves consistent, well-reasoned decision-making?

What the feedback loop already shows

Every appeal determination is, in effect, a stress test of the policies it considers. The green belt and procedure findings above are just two examples. The same structured analysis can be applied to individual local and national plan policies across every authority in England.

In one district council, a development plan policy governing live/work units has been cited in over 100 inspector arguments, with an allow rate above 75%. Inspectors are consistently interpreting the policy more permissively than the authority that wrote it. That interpretation gap is precisely the kind of signal that should inform the next local plan review, and it is only visible when policy citations are structured across decisions rather than read one letter at a time.

This is not information for one side of the planning industry. It serves local authority officers seeking to understand why refusals are being overturned, consultants advising on the strength of a case, and policymakers evaluating whether reforms are delivering their intended outcomes. It is intelligence for the entire planning lifecycle, from plan-making through to application, refusal, appeal, and back again.

The planning profession has always generated this evidence. The difference now is that it can be read and analysed at scale, and the patterns being revealed are already changing how professionals approach planning decisions.

Methodology: All statistics are derived from structured analysis of over 170,000 decided Planning Inspectorate appeal decisions for England, using the Planning Appeals platform. Allow rates are calculated as allowed / (allowed + dismissed), excluding enforcement cases. Green belt and grey belt figures are based on full-text analysis of decision letters. Local policy allow rates are based on structured citation data linking individual policies to inspector reasoning and outcomes. Data current as of April 2026.

David Robinson is ARB registered, RIBA-chartered architect and co-founder of Planning Appeals (planningappeals.co.uk), a planning appeal intelligence platform. The platform structures inspector reasoning, NPPF and local policy citations, and procedural outcomes, with tools for exploring LPA performance, procedure trends, and policy effectiveness.