Assessing Decision-Stage Evidence in Major Planning Applications: A Reproducible Cross-Authority Method

Abstract

This paper presents a reproducible method for assessing decision-stage evidence in major planning applications. The method focuses on a narrow procedural boundary: what evidence can legitimately be said to have existed before the decision-maker at the moment of determination, and what is instead constructed or completed later.

The approach consists of: (i) deterministic evidence capture from public planning registers; (ii) separation of text-bearing decision materials from image-only plans and scanned representations; (iii) extraction of decision-stage signals (including deferral language such as reserved matters, indicative, and by condition); and (iv) classification of findings into governance risk families. Across cases, similar patterns are observed: outline-stage decision gates in core infrastructure topics are repeatedly treated as capable of later completion, creating governance and decision-exposure risk even where substantial evidence exists in the wider file.

The paper does not assess planning merits or outcomes and does not state whether any decision was lawful or unlawful. It provides a practical, auditable methodology for evidential boundary analysis in major planning determinations.

---

Introduction

Major planning applications routinely generate large evidence files. In such environments, a recurrent governance question arises: is the decision demonstrably grounded in decision-stage evidence, or does reasoning rely on later completion, clarification, or reconstruction?

This paper addresses a procedural problem rather than a merits dispute. It fixes the boundary between evidence that existed at decision-stage and material that emerges after challenge, redesign, or further work.

The paper proceeds by defining decision-stage evidence, explaining why hindsight cannot cure evidential absence, and setting out a reproducible capture-and-assessment method. The method is then applied across multiple authorities to evaluate whether similar deferral mechanisms recur in core topics such as highways, drainage, utilities capacity, and ecology.

The paper is written for a professional audience concerned with governance integrity, auditability, and defensible decision recording.

---

Definitions and Scope

Decision-stage evidence means the evidence that was available to the decision-maker at the moment of determination (committee, delegated decision, or equivalent). It is not a retrospective narrative of what might have been understood.

Post-decision material means material created or uploaded after the decision moment, including explanatory notes, further technical work, or later revisions. It may be useful in practice but cannot be treated as if it existed earlier.

Major applications are used because they provide the richest decision file and the strongest stress-test for evidential boundary analysis: large document volumes, multiple technical topics, iterative revisions, and extended consultation.

---

Methodology

The method is designed to be reproducible across authorities using public planning registers and structured evidence capture. It focuses on isolating the decision-stage evidential boundary and identifying procedural risk signals that recur across major applications.

Step 1 — Deterministic evidence capture. For each authority, the full document set is downloaded directly from the planning register using a stable, auditable retrieval process. Evidence manifests are produced with file hashes and capture timestamps.

Step 2 — Document-type separation. Documents are separated into:

• Text-bearing documents (assessments, reports, decision notices, consultation responses).
• Image-dominant documents (plans/drawings) and scanned representations, which are excluded from text-based evidential reasoning unless OCR is undertaken.

Step 3 — Decision-stage signal extraction. Text-bearing documents are searched for signals indicating deferral of decision gates, including:

• reserved matters reliance for matters framed as outline-stage acceptability issues
• indicative or illustrative framing for core infrastructure topics
• by condition language where the condition would effectively complete the missing assessment

Step 4 — Classification. Findings are classified into cross-cutting governance exposure and topic-specific evidence/deferral patterns (highways, drainage, utilities, ecology, noise/air, climate).

---

Case Application and Sampling

The method is applied to multiple major applications across different planning authorities. Cases are selected to maximise comparability: major schemes with substantial document volumes and active technical consultation.

For each case, the full evidence file is captured from the planning register, processed into a text-bearing corpus, and analysed for decision-stage deferral signals and topic-specific evidence gaps.

The sample is intentionally small at this stage and is presented as an initial demonstration of method and observed recurrence, not as a statistical generalisation across all authorities.

---

Findings

Across cases, similar procedural patterns are observed. The most consistent signals arise in topics that operate as practical decision gates: highways/transport, drainage/flood risk, and infrastructure/utilities capacity.

Common recurring features include:

• Repeated reliance on reserved matters to develop matters framed as fundamental to acceptability.
• Use of indicative or illustrative framing in decision-critical topics, creating evidential ambiguity at decision-stage.
• Conditions used to defer completion of assessments rather than secure implementation of already-established acceptability.

These findings do not state that decisions were unlawful. They identify recurring governance risk patterns that increase decision exposure and reduce auditability.

---

Discussion and Implications

The observed recurrence across authorities is consistent with structural drivers: shared evidence workflows, decision-making under document-volume constraints, and routine reliance on legally available deferral mechanisms (conditions and reserved matters) to manage complexity and delivery risk.

This creates a recurring governance tension:

• Major applications require extensive technical assurance.
• Decision-stage determination compresses this into summary reasoning.
• Deferral language becomes a common mechanism for bridging unresolved acceptability questions.

The method provides a way to make this boundary explicit and to test, case by case, whether decision-stage evidential sufficiency is clearly demonstrated.

---

Limitations and Conclusion

Limitations. The sample is small and is presented as an initial demonstration of method and observed recurrence. The approach focuses on text-bearing decision materials; image-only documents (plans/drawings) and scanned representations are excluded unless OCR is undertaken. The paper does not assess merits or outcomes and does not determine lawfulness.

Conclusion. This paper sets out a reproducible method for assessing decision-stage evidence in major planning applications across authorities. Applied across cases, the method identifies recurring decision-stage evidential risk patterns, especially where outline-stage decision gates in core infrastructure topics are repeatedly framed as capable of later completion by reserved matters, indicative schemes, or conditions.

In many cases, deferral is justified on the basis that unresolved matters can be completed later through reserved matters, detailed design, or conditions. However, these mechanisms depend on future deliverability.

Accordingly, the method treats deferral language as a decision-exposure signal: it is not enough that a solution might exist in theory; the decision record should demonstrate, at the point of determination, that the proposal is acceptable and that any required measures are realistically deliverable within the consent framework.

Future work will expand the case set and refine the classification taxonomy for recurring deferral mechanisms and decision-gate topics.

---

Supporting Reports and Replication Materials

The following reports were generated by applying the methodology described in this paper to major planning applications across different authorities. They are provided to support transparency, auditability, and replication.

Each report is derived from a full public evidence capture, followed by separation of text-bearing decision materials and structured decision-stage analysis. The reports do not assess planning merits or determine lawfulness; they document decision-stage evidential structure and identified governance risk signals.

---