What a reviewer checks in a drafted heritage statement

Reviewing a heritage statement draft is a structural exercise before it is a factual one. The reviewer's first question is whether significance was assessed before impact, or whether the document decided the answer and then assembled a justification. That single test disposes of most poor drafts, including most AI-assisted ones, because a document written to reach a conclusion reads differently from one written to find one.

The framework is settled. Sections 16 and 66 of the Planning (Listed Buildings and Conservation Areas) Act 1990 require special regard to the desirability of preserving a listed building, its setting and its features of special architectural or historic interest, and section 72 requires special attention to preserving or enhancing the character or appearance of a conservation area. The courts have treated those duties as requiring considerable importance and weight to be given to any harm identified. National policy then adds the significance-first approach and the harm and public benefits balance.

Whether significance is assessed before impact

Open the document at the significance section and check what it contains. A statement of significance describes what is important about the asset and why, under headings a decision maker can weigh: evidential, historical, aesthetic and communal interest, with the relative contribution of different elements identified.

What it is not is a copy of the list entry. Listing descriptions exist to identify a building, not to explain its significance, and they are frequently silent about the very features a scheme affects. A draft that reproduces the National Heritage List entry and calls it a significance assessment fails at the first hurdle, and a reviewer will say so.

The second check is whether significance is broken down by element. A proposal affecting a rear outshut needs to know how the outshut contributes, not merely that the building as a whole is Grade II. Without that, the harm judgement has nothing to attach to.

A Heritage Statement that gets this right can be short. One that gets it wrong cannot be rescued by length.

The evidence base a reviewer expects

Proportionality is the governing principle: the evidence should match the significance of the asset and the scale of the proposal. For a modest extension to a Grade II house, a reviewer expects a site visit with internal inspection, historic map regression, the local Historic Environment Record search, the list entry, and any local heritage designations. For a scheme affecting a Grade I building or a scheduled monument, considerably more.

Three gaps recur. No internal inspection where internal works are proposed, which is not survivable and is the commonest reason a reviewer refuses to sign. No historic map sequence, so phasing of the building is asserted rather than shown. And no Historic Environment Record search, which also means the archaeological question has not been asked, and archaeology may need its own Archaeological Desk-Based Assessment.

Where fabric will be lost, a reviewer will also ask whether recording is proposed, and at what level. An Historic Building Recording to the appropriate Historic England level is often the mitigation that makes a scheme acceptable, and proposing it in the statement is better than having it conditioned.

Setting assessment and the staged approach

Setting is where drafts most often go thin, because it is the part that cannot be done from a desk. Historic England's advice sets out a staged approach: identify the assets affected, assess whether and how setting contributes to significance, assess the effect of the proposal on that contribution, explore ways to avoid or reduce harm and maximise enhancement, and record the decision.

A reviewer checks that each stage appears, in order, and that stage two is done before stage three. The frequent failure is a document that lists nearby listed buildings, states that the site is not visible from them, and stops. Visibility is one component of setting, not the whole of it, and historical, functional and associative relationships can matter more than views.

The other check is scope. Setting assessment should cover assets outside the site, including those on the other side of a boundary the applicant does not control, and it should say which assets were screened out and why.

Harm characterisation and the statutory duties

The reviewer looks for three things: a clear statement of whether harm arises, a characterisation of its level, and an explanation of how that level was reached.

National policy distinguishes substantial harm or total loss from less than substantial harm, and for non-designated assets requires a balanced judgement having regard to the scale of harm and the significance of the asset. The category is not a matter of impression. It follows from what significance is affected and how much of it.

Two errors get sent back routinely. Concluding "no harm" where the scheme removes historic fabric or alters a principal elevation, which is a conclusion a reviewer cannot defend. And using "less than substantial" as though it meant minor, when the range runs from barely perceptible to something close to substantial, and where in that range the harm sits is precisely what the decision maker needs to know.

Where the statement then relies on public benefits, those benefits have to be stated as benefits to the public rather than to the applicant, and weighed against harm to which considerable importance and weight attaches. Our note on less than substantial harm and appealing a heritage refusal covers how inspectors handle that balance.

Photographs, viewpoints and their absence

A heritage statement with four photographs and no viewpoint plan is a heritage statement a reviewer cannot check.

What is expected: a numbered viewpoint plan, photographs keyed to it, dated and attributed, covering the elevations affected, the internal spaces affected, the relationship with adjoining assets, and the views in which the asset is experienced. Where a proposal affects a roofline or a rear elevation visible from a public space, the photograph from that space is not optional.

Historic photographs and drawings, where they exist, do more work than most modern shots. A pre-alteration photograph showing an original window arrangement can settle an argument about reinstatement in one image.

What a reviewer will not sign

That last row matters for AI-assisted drafting specifically. PlanScope inserts a [TO BE CONFIRMED] flag wherever a site fact is missing rather than filling the gap with something plausible, which is a safeguard rather than a defect. The flags are a task list for the human author, and a document still carrying them is not a document to submit or to sign.

Turning review comments into a submittable document

Work through the comments in the document's own order rather than in the order they arrive. Significance first, then setting, then harm, then mitigation, then the balance. Fixing the harm section while the significance section still cannot support it produces a second round of comments.

Three habits shorten the cycle. Send the reviewer the drawings and the photographs at the same time as the text, because half of all comments are really questions about what is proposed. Record the site visit date and what was inspected, so the reviewer does not have to ask. And where you disagree with a comment, answer it in the document rather than in the covering email, since the document is what the officer reads.

A heritage statement is a report type where councils expect a suitably qualified author and professional review is strongly advised. PlanScope produces an AI-assisted draft structured to policy, citing sources and flagging missing facts, which a heritage specialist then takes ownership of; the professional directory is the route to arranging that review. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. This is general guidance rather than legal advice.