What a draft can say about five year housing land supply
A draft can set out the five year housing land supply test and structure a site-by-site critique. It cannot know an authority's current position, and a plausible wrong figure costs more than a blank one.
An AI-assisted draft can set out the five year housing land supply test correctly, quote the statutory framing, structure the argument and tell you what evidence the case needs. It cannot tell you what the supply position is. Any draft that states a figure for an authority's supply, or asserts that the authority cannot demonstrate five years, is producing a number it has no way of verifying, and that is the single most dangerous sentence a planning statement can contain.
The distinction matters because supply arguments are won on site-by-site evidence. A statement that gets the framework right and the arithmetic sourced is useful. A statement that gets the framework right and the arithmetic invented loses a developer their credibility with the case officer for the whole application.
Where the supply position comes from
The supply position is not a national dataset. It is produced by each local planning authority, from its own housing requirement and its own schedule of sites, and it moves.
The requirement side comes from the adopted plan or, where the plan is out of date, from the standard method. The supply side is a list of specific deliverable sites with assumptions about lead-in times, build rates and the number of completions falling within the five year window. A buffer is then applied on top of the requirement, with a larger buffer where past delivery has been poor.
What this means in practice is that the position has an author, a date and a document. It appears in an authority's annual monitoring report, in a standalone five year housing land supply statement, or in the housing topic paper supporting a plan examination. The approach to finding and reading those documents is set out in checking the housing land supply position for an authority, and it is the first task on any application where supply might matter.
Why a draft cannot know the number
AI drafting has no access to that authority's current monitoring report, and even where a figure appeared in its training data it would be a snapshot of unknown date. Three things make the risk worse than a stale figure normally would.
Supply positions change several times a year. A new monitoring report, a large permission, a withdrawn site or an appeal decision can move the figure.
The figure is contested by design. Authorities and appellants routinely produce positions a year or more apart on the same evidence base, because they disagree about individual sites rather than about the method.
And the consequence of being wrong is asymmetric. Overstating a shortfall invites the authority to demonstrate otherwise and undermines everything else in the statement. Understating it gives away the strongest argument a developer may have.
PlanScope's drafting handles this by inserting a "[TO BE CONFIRMED]" flag where a site-specific or authority-specific fact is missing, rather than filling the gap. That is deliberate. A flag you have to clear is a nuisance. A plausible wrong number is a liability, and the wider pattern is set out in what drafts get right about structure and wrong about evidence.
What a draft can legitimately do
Quite a lot, as long as it stays on the structural side of the line.
| A draft can | A draft cannot |
|---|---|
| State the five year supply test and the buffer mechanism | State an authority's current supply figure |
| Set out the glossary definition of a deliverable site | Judge whether a named site meets it |
| Explain the consequence of a shortfall for the planning balance | Assert that a shortfall exists |
| Frame the site-by-site analysis and list the evidence needed | Supply the evidence |
| Draft the argument around a figure you have sourced | Source the figure |
| Reproduce the Housing Delivery Test consequences | Report this year's result |
Used that way, the draft saves the repetitive part of the work. The policy framing, the test wording, the structure of a site-by-site critique and the connection between a shortfall and the weight given to policies are all stable material that does not need rewriting for every application.
Deliverability and the evidence behind it
The national glossary definition is where most supply disputes actually live. A deliverable site has to be available now, in a suitable location, and achievable with a realistic prospect that housing will be delivered within five years.
Sites with detailed planning permission are generally treated as deliverable unless there is clear evidence that completions will not begin within five years. Sites without detailed permission, including allocations, outline permissions and permission in principle, need clear evidence that completions will begin within the period. That asymmetry is where a developer's analysis earns its money.
The evidence that moves an inspector is specific and dull.
- Planning history for each site, with dates and current status.
- Whether a section 106 agreement is signed, and whether any conditions requiring discharge before works start remain outstanding.
- Lead-in times evidenced from comparable local schemes rather than from a generic table.
- Build rates evidenced from recent local completions, outlet by outlet.
- Ownership and whether the owner has stated an intention to develop.
- Physical constraints, including remediation, ground conditions, access land assembly and utilities capacity.
- Whether a site has appeared in successive supply schedules without delivering.
None of that can be drafted. But all of it can be structured in advance, which is the useful division of labour. The definitional groundwork is covered in deliverable and developable sites in housing supply.
The delivery test and its consequences
The Housing Delivery Test is a separate, published annual measurement of completions against requirement, and its consequences are mechanical rather than argued: an action plan at one threshold, a larger buffer at the next, and the presumption in favour of sustainable development at the lowest.
Because the result is published centrally and dated, it is one of the few figures in this field that can be stated with confidence once you have looked it up. Look it up. Do not let a draft recall it, and quote the publication date alongside the percentage, as the treatment in the Housing Delivery Test explains.
The same caution applies to the National Planning Policy Framework itself. The Framework was renumbered in December 2024 and re-issued on 17 August 2026 with policy-coded chapters in place of paragraph numbers, so a paragraph number recalled from an older edition will be wrong. State the edition you are citing and verify the reference on gov.uk at the time of writing.
Presenting a contested supply case
Illustrative example, not a real case: a developer promoting 85 dwellings on an unallocated edge-of-settlement site believes the authority's claimed 5.3 year supply is overstated. The useful sequence is to take the authority's published schedule, select the 12 largest sites, test each against the deliverability definition with the evidence listed above, quantify the completions that should be removed, restate the supply with the buffer applied, and only then draft the planning balance argument. The draft document does the fifth and sixth steps well and cannot do the first four at all.
Write it so that every number has a citation to a published document and every judgement about a site has a stated basis. Where you rely on an appeal decision, describe it as indicative of how inspectors have approached an issue rather than as binding precedent, because it is not.
And have a planner read it. A contested supply case is the point at which an AI-assisted draft most needs a qualified professional over it, which is what the professional review route exists for.
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, not legal advice; check the position with your local planning authority or a qualified professional before relying on it.
Frequently asked questions
Can an AI draft state an authority's housing land supply figure?
No, and it should not try. The position is produced by each authority from its own requirement and site schedule, changes several times a year, and is contested site by site. A drafting tool has no access to the current monitoring report, so any figure it produces is unverifiable. Source it from the authority's published document instead.
Where is the five year supply position published?
In the authority's annual monitoring report, in a standalone five year housing land supply statement, or in a housing topic paper supporting a plan examination. Each carries an author, a base date and a methodology. Quote the document and the base date alongside the figure, because a position six months old may already have been superseded.
What makes a site deliverable?
The national glossary definition requires the site to be available now, in a suitable location and achievable with a realistic prospect that housing will be delivered within five years. Sites with detailed permission are generally treated as deliverable absent clear contrary evidence. Sites with only outline permission, an allocation or permission in principle need clear evidence that completions will begin.
What does the Housing Delivery Test change?
Its consequences are mechanical rather than argued, running from an action plan requirement at one threshold, to a larger buffer on the five year requirement at the next, to the presumption in favour of sustainable development at the lowest. Because the measurement is published centrally and dated, it can be stated confidently once you have looked up the current publication.
Should a statement cite NPPF paragraph numbers?
Only from the edition you are actually using, stated explicitly. The Framework was renumbered in December 2024 and re-issued on 17 August 2026 with policy-coded chapters replacing paragraph numbers, so a number recalled from an earlier edition will be wrong. Verify the reference on gov.uk at the time of writing rather than copying it from older output.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.