Skip to content
PlotUProperty intelligence · AI · evidence

Planning intelligence · 8 min read

Briefing the planning consultant with the reading already done

Professional fees were being spent on assembly, not judgement. The workflow moved the reading upstream, with every claim cited to its source.

Published

About this case study

This is an anonymised composite drawn from recurring engagements. No client is identified, no figures are invented, and no outcome is guaranteed. The workflow is the substance.

The situation

A developer commissioning planning advice on prospective sites noticed a pattern: the first tranche of every consultant's fee went on assembling the same planning history, policy references and local precedent that could have been gathered before the engagement started.

The details that follow are a composite of recurring engagements of this kind. No client is identified, and nothing here is a claim about a specific company's results.

How the workflow runs

  1. Assemble the planning record

    For the site and its immediate area: application history, decisions, officer reports, committee minutes where they exist, the policies cited, and appeal decisions that set the local tone.

  2. Structure before summarising

    The record is structured first — dates, references, outcomes, policies — so summaries are built from organised material rather than from a pile of PDFs in one pass.

  3. Draft the intelligence pack with citations

    AI drafts the pack: a chronology, the policies that recur in decisions, the pattern in nearby refusals and approvals, and the appeal principles that actually transfer. Every policy statement cites its paragraph.

  4. Verify, then hand over

    A human verifies each cited claim against its source, corrects what is wrong, and the verified pack becomes the briefing document for the planning consultant — whose judgement the pack is designed to inform, not pre-empt.

What changed

Consultant conversations started at judgement instead of at assembly. The developer arrived knowing what the record said; the consultant was paid to interpret it.

  • Planning history, policy context and precedent in one cited document
  • Consultant briefings start from verified material, not from a verbal download
  • The same pack feeds appraisals and funding conversations
  • The record of what was known, and when, exists before the application does

Where the professional line sits

  • The pack is research, not planning advice — acceptability in policy terms is the consultant's call
  • No policy statement travels without a source citation
  • Anything the model cannot source is flagged for checking, never presented as fact

FAQ

Common questions

Does this replace a planning consultant?
No, and it is not designed to. It does the reading so the consultant's time goes on interpretation and strategy. The professional judgement stays exactly where it was.
How is accuracy controlled?
Three ways: work only from documents actually gathered, require a source paragraph for every policy claim, and verify each claim before the pack is shared. Unsourced statements are treated as prompts to check, not findings.
Is this a real client engagement?
It is an anonymised composite of recurring engagements. Publishing real workflows without client identities or invented figures is a deliberate editorial rule of this practice.

Keep reading

More case studies

Discuss your workflowAll case studies