Optimising site capacity under London Plan policy D3
Density is now an output of the design process in London, not an input. Here is how to build a design-led capacity study that an officer can follow and a borough will accept.
The London Plan 2021 does not tell you how many homes a site can hold. Policy D3 asks you to work it out, by testing the site's context and constraints through design, and then to justify the number you arrive at. That is the design-led approach to optimising site capacity, and it replaced the density matrix that earlier London Plans used to set dwellings and habitable rooms per hectare against public transport accessibility.
For an architect this changes what the submission has to contain. A density figure is now an output of the design process, not an input to it, and a capacity claimed without the testing behind it is the argument most likely to fail at officer level.
What replaced the density matrix
The 2016 London Plan carried a density matrix: find the setting, find the public transport accessibility level, read off a range. It was used as a cap as often as a guide, and schemes that sat inside the range were treated as acceptable while schemes above it needed justification.
Policy D3 removed the matrix. In its place the plan asks that all development make the most efficient use of land, with the form and layout responding to the site's context, its connectivity and the capacity of its infrastructure. The word that matters is "optimise", which is deliberately not "maximise". A site optimised under D3 is one where the quantum of development is the most the site can carry while still producing acceptable homes, acceptable neighbouring amenity and a form that fits its setting.
Boroughs still set standards that constrain the answer, and the plan is explicit that optimisation does not override design quality, so the two work together rather than in tension.
How a design-led capacity study is built
The study is a piece of reasoning, presented in sequence, and the sequence is what an officer follows.
- Establish the site and its context: the prevailing building form, storey heights, plot depths and rhythm of the surrounding blocks, plus any designations touching the site.
- Record connectivity: the public transport accessibility level, walking distances to stations and town centre boundaries, cycle infrastructure and the local plan's allocation if the site has one.
- Identify the constraints that fix the envelope: daylight to neighbours, overlooking distances, tree constraints, rights of way, sunlight to amenity space, protected views and any local views policy.
- Test three or four options at different quanta, drawn properly, each assessed against the same criteria.
- Select the option that performs best and state why the others were discarded.
- Express the result as a number, with the density stated but presented as a consequence of the design rather than as the justification for it.
The temptation to skip to step five is strong, particularly when the client has already fixed the unit count. Resist it. The discarded options are the evidence that optimisation happened.
Site context, connectivity and capacity
Three inputs carry most of the weight, and each has a document that supports it.
Context is established by survey and analysis, not assertion. A statement that the area is characterised by four-storey mansion blocks needs a street elevation showing them. Where the proposal exceeds the prevailing height, the case has to be made on the site's own attributes, such as a corner location, a boundary with a wider road or a transition to a taller adjoining block.
Connectivity is objective and easy to evidence. The public transport accessibility level is published, and walking distances can be measured. A site at PTAL 6a in a town centre carries a different capacity from an identical site at PTAL 2 in a suburb, and the plan expects that difference to show in the scheme.
Infrastructure capacity is the input most often skipped. Primary school places, health provision, water and sewerage capacity, play space provision and open space deficiency all bear on how much development a site can absorb. Where the scheme generates a child yield, the play space assessment needs to demonstrate that the on-site provision follows the Mayor's standard rather than a convenient reading of it.
Testing options rather than asserting a figure
An option test is only useful if the options differ in substance and are assessed on the same terms. Three schemes at 42, 44 and 46 units is not a test. Three schemes at 30, 45 and 60 units, each with its own massing, daylight consequences and amenity outcomes, is.
For each option, record the same measures: unit count and mix, gross internal areas against the Nationally Described Space Standard, dual aspect percentage, single aspect north-facing units, private and communal amenity space, daylight and sunlight results for the proposed units and for neighbours, and the effect on the street elevation. Where an option fails, say which measure it failed on.
A daylight and sunlight assessment run at option stage rather than after the scheme is fixed will change the design, which is the point. Late testing produces a scheme that has to be defended rather than one that was optimised.
Where boroughs still apply density ranges
Some borough local plans retain density guidance, site allocations with indicative capacities, or tall building policies that set height thresholds and identify where tall buildings are and are not appropriate. Those sit alongside D3 rather than being displaced by it, and the development plan has to be read as a whole under section 38(6) of the Planning and Compulsory Purchase Act 2004.
The practical consequence for a scheme above an indicative allocation capacity: the allocation figure is indicative, so exceeding it is not a breach, but the design-led case has to be stronger and the borough will expect the infrastructure and amenity consequences to be answered directly. Where a tall buildings policy applies and the site is outside an identified location, the height argument is a much harder one and rarely rescued by capacity reasoning alone.
Evidencing the process in a design statement
The capacity study belongs in the design and access statement, as a labelled section with the option drawings included rather than as an appendix nobody opens. What an officer needs to see, in order: the context analysis, the constraints plan, the options, the assessment of each against stated criteria, the selected option and the resulting density expressed in both homes and habitable rooms per hectare.
Consistency across documents matters more than length. If the capacity study says 58 units and the energy strategy in the energy statement models 54, the reviewer stops reading the argument and starts checking the arithmetic. Same for the accommodation schedule, the affordable housing statement and the transport documents.
Common reasons a capacity argument fails
The recurring ones make a fair pre-submission checklist.
Options that differ only cosmetically. A density figure quoted from a comparable scheme rather than derived from this site. Context analysis that describes the area in adjectives instead of measurements. Daylight testing run once, at the end, on the scheme the client wanted. Infrastructure and child yield left to the section 106 negotiation. Single aspect north-facing units presented without explanation. And a capacity claim that depends on the neighbouring site being redeveloped at the same time, with no mechanism to secure it.
The scheme that passes is usually the one where the drawings show the thinking, and where the number at the end is the smallest claim in the document rather than the largest.
Note: this article covers the London Plan position. Outside Greater London the design-led principle applies through national policy and local plan design policies instead, and the borough-specific standards described here do not.
PlanScope drafts design and access statements, energy statements and the supporting documents around a capacity study, structured to UK planning policy and flagged where site evidence is missing. Browse the 66 report types and start a draft. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.
Frequently asked questions
Does the London Plan still have a density matrix?
No. The 2016 plan's matrix of dwellings and habitable rooms per hectare against public transport accessibility was removed in the London Plan 2021. Policy D3 asks instead for a design-led approach in which capacity is established by testing the site's context, connectivity and infrastructure through drawn options, with the density figure reported as the outcome.
What does optimising site capacity actually mean?
It means the most development a site can carry while still delivering acceptable homes, acceptable amenity for neighbours and a form that responds to its setting. It is not the same as maximising units. A scheme that reaches a high density by producing single aspect north-facing flats with poor daylight has not optimised the site under policy D3.
How many options should a capacity study test?
Three or four clearly different quanta, each drawn and assessed against the same criteria, is the normal expectation. Options separated by two or three units are not a test. Record unit mix, space standards, dual aspect percentage, amenity space, daylight results and the street elevation for each, and state which measure caused each discarded option to fail.
Can a scheme exceed the capacity in a site allocation?
Yes, because allocation capacities are indicative rather than a ceiling. The design-led case has to be correspondingly stronger, and the borough will expect the infrastructure, amenity and townscape consequences to be answered directly. Where a tall buildings policy identifies where height is appropriate and the site sits outside it, the argument is much harder.
Where should the capacity study sit in the application?
Inside the design and access statement as a labelled section with the option drawings included, not buried in an appendix. Keep the unit count consistent with the accommodation schedule, the energy strategy and the transport documents, because inconsistent numbers move the officer from assessing the argument to checking the arithmetic.
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.