Reviewing a consultant's draft before it is submitted

The draft arrives as a PDF two days before the submission deadline, and the temptation is to check the client name is right and send it on. Resist that. A specialist report that contradicts the drawings, cites a superseded policy edition or describes a site with a hedge that was removed in 2023 does more harm than no report at all, because the case officer stops trusting the rest of the package.

Half an hour of structured reading catches nearly all of it.

Has the brief been answered

Read the scope of works first, then the report, in that order. Most reviews go wrong because the reviewer reads the report on its own terms and is persuaded by it.

Check that the report addresses the application it was commissioned for, names the right authority, and deals with the local plan policies you identified. Check that the standard you named in the brief is the one cited in the methodology. Check that every deliverable in the scope is present, including figures, appendices and the non-technical summary.

Then look for what is missing. A report that answers six of the eight questions you asked, and says nothing about the other two, has not answered the brief. Silence is not agreement.

Does the site description match the site

This is the fastest way to detect a recycled report, and it catches more than you would expect.

Walk through the site description with the survey drawing and recent photographs open. Check the boundaries, the levels, the existing buildings, the trees, the access arrangements and the neighbouring uses. Check the site area against the red line. Check the direction of the slope.

Pay attention to the date of the site visit and the weather. An ecology report based on a February walkover describing the grassland as species poor is doing something different from one based on a June survey, and the council will notice.

Where the report describes something that is no longer there, or misses something that is, go back before the design team builds an argument on it.

Are the policy references current and correctly cited

The Framework was re-issued on 17 August 2026 with policy codes replacing paragraph numbers, having already been renumbered in December 2024. A report quoting a paragraph number from the December 2023 edition is at least two editions out of date, and a report quoting the December 2024 numbering is one.

Check three things. That the edition is stated wherever a reference is given. That local plan policies are cited by their current reference and not by the number they carried in a superseded plan. And that any supplementary document relied on is still adopted, because councils withdraw them.

Where the report cites an appeal decision, check that it is described as indicative rather than as binding precedent, and that the reference is complete enough for an officer to find it.

Can every figure be traced to a source

Take every number in the conclusions and ask where it came from. Not most of them. Every one.

A trip rate should name the database and the site categories used. A noise level should name the measurement position, the date, the duration and the equipment. A carbon figure should name the quantities source and the emission factors. A cost should name the basis. A flood level should name the model or the published dataset and its version.

Numbers that appear only in a conclusion, with no derivation anywhere in the report, are the ones that fall over at consultation. Ask for the working. A consultant who cannot produce it has a problem you want to discover now.

Check the arithmetic in at least one calculation. It is surprising how often a total does not add up, and finding one error changes how carefully everything else gets checked.

Do the conclusions hedge without explaining why

Specialist reports hedge for two reasons. Either the evidence supports only a qualified conclusion, which is honest and should be explained, or the author is protecting themselves, which is not useful to anybody.

Look for phrases like unlikely to be materially affected, or considered acceptable subject to mitigation, and ask what the mitigation is, who delivers it and whether it is in the scheme. A conclusion that depends on unspecified mitigation is an invitation for the council to impose a condition nobody has costed.

Also look for the opposite problem: a conclusion stated more firmly than the evidence supports. If the survey covered one season and the conclusion speaks for the whole year, that gap belongs in the report rather than in a later objection.

Ask the author directly what they are least confident about. Good consultants answer that question well, and the answer tells you where the application is exposed.

Do the report and the drawings say the same thing

Cross check the report figures against the submission drawings.

That last row causes more validation problems than any other single issue. A report assessing revision C while revision E is submitted will be spotted, and it undermines everything else in the document.

Returning comments the consultant can act on

Write comments the author can work from without a phone call.

1. Number each comment and reference the page and paragraph.
2. Say what is wrong and what you want instead, rather than only flagging discomfort.
3. Separate factual corrections from matters of judgement, and say which is which.
4. Mark anything that changes the conclusion, so it gets attention first.
5. Give a deadline and confirm whether this is the first or the second of the review rounds in the appointment.
6. Copy anyone whose input is needed, particularly the design team member who owns the drawing being referenced.

Avoid the trap of editing the author's professional judgement. You can challenge a conclusion, ask for the reasoning, or ask for an alternative to be considered. You cannot instruct a specialist to reach a conclusion their evidence does not support, and an independent reviewer will spot a report that has been leaned on. The wider issue of independence in planning evidence is covered in independence and conflicts of interest in planning evidence, and the method for checking a report you had no part in commissioning is set out in reviewing a specialist report you did not commission.

Where an independent check is wanted before submission, the professional review route is one way to get it. For the narrative documents in the pack, PlanScope drafts to a policy structure and flags missing site facts rather than inventing them. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. Browse the report types to see what a full submission usually contains.