The New 2026 NPPF & Design Quality: Chapter 14, Design Review, & What the Framework Does Not Say
- Jonathan Braddick
- 6 days ago
- 15 min read
The National Planning Policy Framework published on 17th of August 2026 is the first full rewrite since 2012. This is a note on what it changes for design, and on what it confirms — and does not confirm — about the use of design review panels.
On Monday 17th of August 2026 the Ministry of Housing, Communities and Local Government published a new National Planning Policy Framework (NPPF) 2026. It is not a revision or a tidy-up of a footnote or two; it is a rewrite, and the first since 2012.
The document has grown from 82 pages to 130. The paragraph numbers have gone, replaced by policy codes which read rather like development plan policies — PM, DM, S, HO, L, GB, DP, TR and HE among some twenty prefixes — and the plan-making and decision-making policies now sit in separate, clearly labelled boxes. Somewhere in England there is a design and access statement carrying forty cross-references to numbered paragraphs, and its author has had a difficult week.
For anyone whose working life involves arguing for design quality — an architect, a planner, a local authority urban design officer, or an applicant who would rather not spend eighteen months in a character-and-appearance argument — this is a better document to argue from than the one it replaces, and the reasons are worth setting out carefully.
The change which matters most to anyone who commissions, provides or relies on design review is one altered verb in policy DP4(2)(a), and I will come to it. Those looking for Paragraph 84(e) will find it alive, well and renumbered, in the housekeeping at the foot of this note.
What Chapter 14 of the 2026 NPPF now says ...
The design chapter of the new NPPF is now Chapter 14, "Achieving well-designed places", policies DP1 to DP4. Old paragraphs 131 to 141 have gone as numbers, though most of their substance survives in tighter form. Some familiar names have gone with them: the National Design Guide is not referenced anywhere in the Framework, and neither is Building for a Healthy Life, although the latter survives in the draft guidance as a place assessment tool. The National Model Design Code survives only in generic form, as "Model Design Codes" in DP2(2) and in the glossary. In their place the Framework points repeatedly to a "Design and Placemaking planning practice guidance", to which I will return, because there is a difficulty with it — not least that it has not yet been published.
DP1, a plan-making policy, asks development plans to set clear design expectations: a vision informed by an evaluation of the area's existing characteristics, identification of where design guides, codes and masterplans are needed, locally-specific design policies where they add something, and — this is the new part — "setting out in what circumstances design review and other design processes will be required."
DP2 sets the ground rules for local design guides, codes and masterplans, which are to understand character, be informed by community engagement, be proportionate rather than exhaustive, and be monitored and reviewed.
DP3 is the substantive decision-making policy. Proposals should respond to their context so that they integrate with and enhance their surroundings, expressly without precluding innovation or change, and particularly where an increased scale or density of development is justified. Seven named principles follow: Liveability, Climate, Nature, Movement, Built Form, Public Space and Identity.
DP4 covers the design process and the expectations it places on local planning authorities. It is worth saying plainly, since the point is easily lost, that all of this is policy expressed as "should". The Framework is national policy of critical importance as a material consideration; it is not itself a statutory duty.
Three changes to the 2026 NPPF which give design real weight ...
The design refusal test has been wired into the new presumption. Poor design has been a stated ground for refusal for some years; paragraph 139 of the 2024 Framework opened with the words "Development that is not well designed should be refused". That is not the change. The change is where the test now sits in the machinery.
DP3(3) reads: "Development proposals should be refused if, without clear justification, they conflict with paragraph 1 of this policy or relevant aspects of the principles in paragraph 2, or with any explicit design standards set out in the development plan (including those in locally-specific policies, guides, codes or masterplans)."
Eleven policies in the whole NPPF are drafted as "should be refused" policies, and the company design now keeps is instructive: peat extraction, ancient woodland, protected landscapes, flood risk and harm to designated heritage assets among them. This matters because of the way the new presumption operates. Policy S4 provides that development within settlements should be approved unless the benefits are substantially outweighed by adverse effects, which is a deliberately high bar for refusal.
But S4(2)(c) lists, as a circumstance likely to clear that bar, a failure to comply with "one of the national decision-making policies which state that development proposals should be refused in specific circumstances." DP3(3) is one of those policies, so in a Framework built around a strong presumption in favour of development, design conflict now sits on a short and named list of matters capable of displacing it. One caveat is worth noting: the 2024 test also caught a failure to reflect "government guidance on design", whereas DP3(3) is anchored to the development plan and to the DP3 principles, which makes local design policy correspondingly more important than it was.
"Significant weight" has become "substantial weight", in both of the places where design is concerned. DP3(3) gives substantial weight to compliance with relevant development plan policies when assessing design quality, and DP3(5) gives substantial weight to "outstanding or innovative designs which promote high levels of sustainability, or which help raise the standard of design more generally in an area, so long as they are consistent with the overall form and layout of their surroundings."
Both read "significant" in December 2024. In fairness this is partly a Framework-wide tidy-up, since "significant weight" has been retired from the document altogether, but the operative phrase for design is now the strongest one the Framework uses. The proviso at the end of DP3(5) repays reading twice before anyone builds a case on it.
Value engineering has quietly run out of road. DP4(2)(c) provides that the local planning authority should "not allow the quality of approved development to be materially diminished between permission and completion, as a result of changes proposed following its initial approval (for example through changes to approved details)." The 2024 wording asked authorities to "seek to ensure" this; the new wording does not ask. Anyone who has watched a carefully argued material palette evaporate at discharge of conditions will recognise this as the least discussed sentence in the chapter, and perhaps the one most likely to be tested first.
What the 2026 NPPF says about design review ...
Two provisions, quoted in full, because precision matters more here than paraphrase.
DP1(1)(d), plan-making. Development plans should set out clear design expectations by:
"Setting out in what circumstances design review and other design processes will be required."
DP4(2)(a), decision-making. The local planning authority should:
"Ensure that they have access to, and encourage the appropriate use of design review and other design tools and processes (especially for strategic sites and other significant projects), and take into account their outcomes, including any design recommendations made by design review panels."
Three observations follow, and then I will leave the point alone.
The phrase "design review" appears three times in the 2026 NPPF. It appeared once in the 2024 version, at the tail of paragraph 138, and then only as "design review panels"; the tool itself was referred to obliquely as "design advice and review arrangements", in a list alongside workshops and assessment frameworks. It is now named in its own right, at the head of the sentence, and it appears in plan-making policy for the first time.
The verb has also changed. In 2024 authorities were to "make appropriate use of" design tools, whereas in 2026 they are to "encourage the appropriate use of" design review, and one does not encourage oneself. The ordinary reading of that word extends beyond passive availability: it points outwards, towards applicants, promoters and design teams, whose appropriate use of design review the authority is now expected to promote rather than merely tolerate. That is an interpretation rather than a settled position, but it is difficult to read the change any other way.
The outcomes must then land somewhere, since authorities are to "take into account their outcomes, including any design recommendations made by design review panels" — the outcomes, that is, rather than the outcomes of any particular panel.
Which brings me to the point worth making once and clearly. The 2026 NPPF expects local planning authorities to have access to design review, to encourage its appropriate use, and to take the outcomes into account. It does not prescribe a provider. It does not say that an authority may use only a panel of its own commissioning, and it does not confine an applicant or a design team to a single panel. Those propositions are simply not in the text.
Two qualifications belong with that, and stating them is a matter of accuracy rather than modesty. Taking an outcome into account is not the same as accepting it: the decision-maker remains responsible for deciding what weight an outcome carries, and that weight will turn on the panel's independence, expertise, process, reasoning and engagement with the relevant policy. And where a scheme has been reviewed before, the full design review history should be disclosed, because duplication and conflicting advice help nobody. The argument here is for provider neutrality, not for shopping until the answer improves.
What the 2026 NPPF does instead is send the reader to guidance, and the question of how design review should be conducted has always been answered elsewhere.
The rule book has not changed, because it did not need to ...
The profession's reference work on how design review should be conducted remains Design Review: Principles and Practice, first published in 2013 by Design Council Cabe with the Landscape Institute, the Royal Institute of British Architects and the Royal Town Planning Institute, and updated in 2019. It has never sat inside the NPPF and has never needed to; its ten principles have outlasted five versions of the Framework for the straightforward reason that nobody has improved on them.
The first is Independent: design review is conducted by people unconnected with the scheme's promoters and decision makers. The remaining nine are Expert, Multidisciplinary, Accountable, Transparent, Proportionate, Timely, Advisory, Objective and Accessible.
Independence is not a decoration on that list; it is the first item, and everything after it rests on it. The guidance is also candid about where the pressure on it falls. It records that "Sometimes panels are managed and funded by a local authority or a group of local authorities, and sometimes they are run independently", and that where they are run directly by an authority it is "preferable for the panel to be managed corporately, detached from the council's development management function." Panels sponsored by or closely linked to an authority, it says, "must deal with the crucial but sensitive issue of maintaining independence from it." It even notes that panels may be funded from "a proportion of the charge for pre-application advice".
And where that pressure cannot be resolved, the guidance says what should happen. Where panel members are associated with a planning authority, have a financial interest in a scheme, or have declared support for it at councillor level, "it may be advisable to refer the project to an independent panel that operates with a wider remit outside the immediate locality."
That is not the Panel's view of the matter. It is the profession's, settled in 2013 and left standing in 2019, and it disposes of the question rather neatly. A panel created and controlled by the decision-maker under a continuing contract, and funded out of the pre-application advice budget, may be entirely competent design review; what it is not is design review independent of the decision-maker. A Panel controlled by the promoter has the mirror problem. Both can be useful. Neither is the thing the first of the ten principles describes, and it is that thing to which DP4(2)(a) now invites a decision-maker to have regard.
The Government's draft guidance takes the same starting point, opening its treatment of design review with the words "Design review is an independent, early evaluation of development proposals by a multidisciplinary panel of experts." An effective panel, it continues, "operates under clear terms of reference to ensure transparency, objectivity and public benefit." For further reading it signposts Urban Design Learning's National Design Review Code of Conduct (2025).
A difficulty worth noting ...
Chapter 14 anchors itself to the "Design and Placemaking planning practice guidance" four times over. DP1(2) says that guidance on applying the plan-making policy "is set out in" it; DP3(4) says its principles "should be used to inform how this policy is applied in the absence of locally-produced design policies, guides, codes or masterplans"; DP4(1) points to its Part 2; and the seven principles in DP3(2) are expressly described as the seven features it sets out.
At the time of writing, that guidance has not been published. Its own consultation ran from 21 January to 10 March 2026, the GOV.UK page still carries the words "We are analysing your feedback", and no response to it has been issued. What exists publicly is the January 2026 consultation draft, and the entry in the Planning Practice Guidance collection remains "Design: process and tools", dated 1 October 2019. A Framework which cites its own guidance four times in three pages is, for the moment, footnoting a book which has not yet been printed.
The timetable is at least known. The government's response to the NPPF consultation, published alongside the Framework itself, states that it "will publish the final Design and Placemaking Planning Practice Guidance in the autumn", and that it also intends to publish "a series of Model Design Codes" for common development types such as large housebuilder-led sites and small urban sites. The draft is a good and detailed document, and there is no reason to expect the final version to depart from it substantially.
In the meantime, two things carry the argument in the gap between a Framework which gives design real teeth and the guidance meant to explain how to use them: the development plan, and the best practice the profession already recognises. Which is a reasonable argument for ensuring that a design review process visibly meets that standard, rather than approximately resembling it.
Where the practical opportunities sit ...
For applicants and design teams, three of them are immediate.
DM1(1)(b)(ii) provides that proposals for major development should be accompanied by a concise planning statement setting out "the outcome of pre-application engagement and the extent to which the proposal has changed in response to this engagement." An independent design review report, together with a documented account of what changed as a result, can form a substantial part of exactly that evidence, and it is difficult to think of a cleaner way of assembling it.
DM3(1)(c) provides that the authority should take account of planning matters raised during pre-application engagement, "including any positive responses to this engagement." Design review outcomes generated through pre-application engagement fall squarely within what a decision-maker is directed to take into account.
DP3(5) offers substantial weight for outstanding or innovative design, and substantial weight calls for substantiation. An independent, multidisciplinary, on-the-record assessment is how "outstanding" stops being an adjective in a covering letter and starts being persuasive evidence. The same holds for HO11(1)(e), the old Paragraph 84(e), which is unchanged.
For local planning authorities, three more.
DP1(1)(d) means that the next local plan needs a paragraph setting out when design review is required. A well-drafted trigger — by scale, by sensitivity, by strategic site status — turns design review from an occasional favour into a predictable and budgetable part of the process, and tells applicants what is expected of them before they arrive.
The expectation that design guides and codes sit within a plan in order to carry weight is not new. What has changed is that no new supplementary planning documents may be brought forward, and PM4(1) restricts their replacement, the supplementary plan, to two purposes only: locally-specific design standards, and responding quickly to unanticipated change between plan-making cycles. Existing supplementary planning documents do not evaporate on 17th of August; authorities were advised in earlier consultations, as the government's consultation response records, to review them and decide whether their content should remain as guidance or be absorbed into new-style local plans. The practical effect is the same in the end, which is that development plan status matters more than it did, and design codes drifting about as informal documents matter less.
L3 introduces a minimum of 35 dwellings per hectare within a reasonable walking distance of a well-connected station, rising to 45 where the service frequency is at least double the qualifying minimum, and it expects development within settlements to contribute to an increase in local density. DP3(1) simultaneously asks proposals to respond to their context. To its credit the Framework reconciles the two expressly, and it leans towards density in both directions: DP3(1) says that context should not preclude increased scale or density justified under L2 and L3, and L3(2)(a) says that character should be taken into account "but should not preclude development which makes the most of an area's potential". The direction of travel is therefore settled, and the argument will be about how far — an argument which is considerably cheaper to have at design review than at appeal.
For everyone, the design chapter now runs to four policies and roughly three pages, which is rather less than most of us spend on the covering letter, and it repays reading properly.
A free standard, offered to anyone who wants it ...
If design review is to carry the weight DP4(2)(a) now invites, the governance behind it needs to be visible, testable and consistently applied. That is a fair expectation for an authority to hold, and an equally fair one for an applicant to hold.
It is also, as it happens, a gap the government has just declined to fill. Question 149 of the NPPF consultation asked about the approach to design review. The response records that consultees raised "the quality of design review panels and the weight to be given to design review in decision-making" and "sought greater clarity on how and when design review should be used in practice". The government took DP4 forward as consulted on, with one minor amendment about strategic sites, on the basis that the policy "provides an appropriate framework for the use of design review and related processes". The sector asked for a standard on panel quality, and national policy has left the sector to supply it.
Earlier this year the Panel published an open-source Code of Conduct for design review panels. It is not the only reference point in the field, and that is no bad thing, since a sector with more than one published standard is a sector taking governance seriously. Ours is free to use, free to adopt and deliberately model-neutral: it is not tied to any single delivery model and presupposes no particular commercial arrangement. It covers seven areas — panel composition, the commissioning relationship, panel management, feedback standards, policy alignment, conflict management and transparent governance — and it incorporates the Nolan Principles of Public Life.
Its central proposition is that independence is structural rather than verbal. That means a panel manager who is not an officer or agent of the commissioning body or of the authority; conflicts declared, managed and recorded; panellist identities not disclosed in advance; the written feedback document, reflecting the collective view of the Panel, as the output rather than the opinion of the loudest voice in the room; and contextual information shared openly in the session rather than passed privately beforehand.
It is offered to independent panels, to panels run by or on behalf of local planning authorities, and to applicants, design teams, inspectors and the public. It can be adopted, adapted, or simply used as a checklist against whatever arrangement is already in place. There is no charge and no catch. A sector which agrees what good governance looks like is a sector whose advice carries weight, and that serves everybody, whoever happens to run the panel.
In short ...
The new Framework is longer, better organised and considerably more directive than the one it replaces. Design has moved from a matter of aspiration to a ground for refusal wired directly into the presumption, and good design has moved from significant weight to substantial weight. Design review is named in its own right in national policy and appears in plan-making policy for the first time, with authorities expected not merely to use it but to encourage its use.
What the Framework asks for is access, encouragement and proper regard for the outcomes. What good practice has asked for since 2013 is that the process be independent, expert, multidisciplinary, proportionate, timely and transparent. Those two things have always fitted together rather neatly, and they still do.
Finally, some housekeeping for completeness ...
Paragraph 84(e) is now policy HO11(1)(e). The wording is identical, word for word: the design must be "truly outstanding, reflecting the highest standards in architecture", help "to raise standards of design more generally in rural areas", and "significantly enhance its immediate setting". Same test, same bar, new number. Appeal and committee decisions interpreting that limb remain relevant, subject as always to their own facts and to the changed Framework around them, and the body of reasoning on what "truly outstanding" means has not been disturbed. The surrounding policy has been redrafted — the chapeau now reads "should only be supported where", and limbs (b) and (c) have changed — but limb (e) has not moved a comma.
The Design Review Panel arranges independent, multidisciplinary design review across England. Our Code of Conduct is published openly and is free to use at designreviewpanel.co.uk/code-of-conduct.
Sources and further reading
Comments