Daylight, Sunlight and Amenity

Privacy distances between facing windows in dense schemes

6 min readBy PlanScope Editorial Team

Separation standards are local guidance, not national policy. On a dense site the argument is about harm to living conditions rather than about the dimension.

The privacy distance between facing windows that most English councils apply is 21 m habitable room to habitable room, with roughly half that to a blank wall or a boundary. Those numbers appear in supplementary planning documents and residential design guides, not in national policy, and they have no statutory force at all. On a dense urban site they are frequently unachievable, which is why the argument is about harm rather than about the number.

Understanding where the figure came from helps, because it explains why it stops being useful above a certain density.

Where the standards came from

The 21 m convention descends from interwar public health guidance about the spacing of two-storey terraced housing on suburban layouts, expressed originally in feet. It was designed around back-to-back houses of a particular height, on flat ground, with windows facing each other squarely.

Almost none of those assumptions hold in a four-storey courtyard block or a mansion block infill. The figure survives because it is easy to apply and easy to defend at committee.

National policy has moved away from rigid numeric standards towards a design-led approach. The National Design Guide and the National Model Design Code both frame privacy as a quality to be achieved through layout, orientation and detailed design. In London, the London Plan takes the same line, and a scheme that meets the design objective by other means is not automatically unacceptable because a dimension falls short.

That does not mean the standard can be ignored. It means the burden is on you to show the outcome is acceptable.

The figures officers actually use

Local guidance varies, and the pattern is consistent enough to summarise.

Relationship Commonly applied distance
Habitable room window to habitable room window, directly facing 21 m
Habitable room window to blank wall or gable 12 m to 14 m
Habitable room window to a shared boundary 10.5 m
Upper floor window to a private garden or amenity space Judged on overlooking, often around 10 m
Oblique or offset relationships Reduced, sometimes by a third, where the angle is more than 30 degrees

Check the actual adopted document for the authority rather than relying on a general figure. Some boroughs apply no numeric standard at all. Others apply the standard only above two storeys, or exempt town centre locations, and a few operate different figures for flatted schemes and for houses.

Habitable room is the term that does the work. Kitchens are treated inconsistently, bathrooms and circulation are not habitable, and a window serving a room that already has a second aspect carries less weight.

Applying standards in high density layouts

In a perimeter block, the courtyard dimension is set by the block, not by the windows. Reaching 21 m across a courtyard on a constrained site often means losing the block form entirely, and the resulting layout is usually worse for daylight, overlooking of the street and defensible space.

So the design question becomes which relationships to prioritise. Living rooms and bedrooms warrant the separation. Kitchens, studies and secondary bedroom windows can often be handled differently.

Ground floor relationships matter more than upper floors in one respect and less in another. Direct views into a ground floor living room from a footpath are intrusive at short range, and are usually solved by level change, defensible planting or a raised finished floor level rather than by distance.

Above third floor level, downward angles mean that facing windows overlook less than the plan dimension suggests. Sectional drawings show this in a way that plans never will.

Design alternatives to raw distance

The techniques that work are ordinary and they need to be in the drawings rather than in the text.

  • Offsetting windows so that facing openings do not align, which removes the direct view without changing the separation.
  • Angled or splayed reveals and oriel windows that direct views along the facade rather than across the gap.
  • Winter gardens and inset balconies, which push the glazed line back and put a threshold between the two rooms.
  • Obscure glazing and fixed panes to secondary windows, secured by condition, provided the room retains an openable clear window elsewhere.
  • Level changes, boundary treatment and structural planting where the relationship is at ground level.
  • Privacy screens to balcony flanks, which deal with the more common complaint about overlooking from outdoor space rather than from rooms.

Obscure glazing is the one to use sparingly. Officers dislike it applied to primary windows, and a scheme relying on it across several units reads as a layout problem rather than a design solution.

How officers weigh a shortfall

The test in the officer report will be whether the relationship causes unacceptable harm to living conditions, judged against the adopted policy. A numeric shortfall is evidence towards that, and it is not conclusive.

Four things move the judgement. What the rooms are used for. Whether the relationship is direct or oblique. What the existing context already looks like, because a shortfall in an area of tight Victorian terraces reads differently from the same shortfall in a 1970s cul-de-sac. And whether the harm falls on existing residents or on future occupiers of your own scheme, since officers generally weigh harm to existing neighbours more heavily.

Be honest about the shortfall in the submission. An application that presents 16 m as though it complied invites a refusal that a candid application would have avoided.

Illustrative example. A courtyard block of flats has facing living room windows 17 m apart across a landscaped court, against a local standard of 21 m. Every affected window is served by an inset balcony 1.5 m deep, the openings are offset by half a bay so that no two windows align, and the court is planted with multi-stem trees along the centre line. The officer report records a departure from the numeric standard and concludes that the living conditions of future occupiers are acceptable. Dimensions here are illustrative.

The point of that example is the ordinariness of the measures. None of them is expensive, and all of them have to be on the drawings at submission rather than offered later in response to an objection.

Evidence that supports a reduced distance

Give the case officer the drawings that answer the question rather than making them measure.

Provide a plan with dimensioned separations between every affected pair of windows, labelled with the room use at each end. Provide sections through the critical relationships showing sightlines from a standing eye height. Provide a schedule listing each relationship, the distance, the standard, and the design response where it falls short.

Where the shortfall is material, support it with quantified daylight work, because privacy and daylight objections usually arrive together and a Daylight and Sunlight Assessment that shows compliant results removes half the argument. Where the concern is outdoor amenity space rather than windows, an Overshadowing Study covers ground the daylight report does not.

Then explain the reasoning in the Design and Access Statement, naming the standard, stating where the scheme departs from it, and setting out the design measures adopted. Officers are far more willing to accept a departure that has been thought about than one they discover with a scale rule.

For a scheme still being scoped, the report index sets out which of the amenity documents a dense residential proposal typically attracts.

This is general guidance, not legal advice. Separation standards are set locally and vary widely, so check the adopted policy with your local planning authority or a qualified professional before relying on it.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.

Frequently asked questions

Is the 21 metre rule a legal requirement?

No. It appears in supplementary planning documents and residential design guides adopted locally, and it has no statutory force. National policy favours a design-led approach, and the National Design Guide treats privacy as something to achieve through layout and detailing. A shortfall is evidence of possible harm rather than an automatic reason for refusal.

Which rooms count as habitable?

Living rooms, dining rooms and bedrooms are habitable in every guide. Bathrooms, hallways, landings and utility rooms are not. Kitchens are treated inconsistently between authorities, so check the local definition. A window serving a room that already has a second aspect generally carries less weight in the assessment than a sole-aspect window.

Do separation distances reduce for oblique relationships?

Many local guides reduce the required distance where windows are offset or angled rather than directly facing, often by around a third beyond about 30 degrees. Some guides say nothing about it, in which case a dimensioned plan and a section showing the actual sightline is the way to make the argument.

Can I rely on obscure glazing to solve overlooking?

For secondary windows, usually yes, secured by condition and provided the room retains an openable clear window elsewhere. For primary living room and bedroom windows, officers dislike it, and a scheme depending on obscure glazing across several units reads as a layout problem rather than a design response. Use it sparingly.

What drawings should I submit on a constrained site?

A plan with every affected window pair dimensioned and labelled by room use, sections through the critical relationships showing sightlines at standing eye height, and a schedule listing each relationship against the standard with the design response. Officers respond far better to a shortfall that has been measured and explained than to one they find themselves.

Sources and further reading

privacyamenityresidential designdensityoverlooking

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.

More in Daylight, Sunlight and Amenity

View all →
Daylight, Sunlight and Amenity7 min read

Daylight modelling inputs that get challenged

The argument in daylight review is almost never about the test. It is about reflectance, geometry, room use assumptions and the baseline that was run.

Read article →
Daylight, Sunlight and Amenity7 min read

Sense of enclosure and how the objection is assessed

Enclosure, overbearing impact and loss of outlook are amenity judgements rather than daylight calculations. What evidence carries weight and which design changes work.

Read article →
Daylight, Sunlight and Amenity7 min read

Obtrusive light at residential windows and the limits used

Obtrusive light limits vary by environmental zone and by curfew period, and the zone assignment is a judgement the whole assessment depends on. The metrics, the figures, and how to make verification work.

Read article →
Daylight, Sunlight and Amenity7 min read

Privacy and overlooking: where the distance rules came from

The 21 metre back-to-back figure is a local design convention with imperial origins, not a national standard. How authorities apply it, and the three arguments that win a reduced distance.

Read article →
Daylight, Sunlight and Amenity7 min read

Overshadowing of amenity areas and the two hour test

At least half a garden should receive two hours of sunlight on 21 March, and where that already fails the loss is judged against a 0.8 factor. What the plots should show and when the test misleads.

Read article →
Daylight, Sunlight and Amenity7 min read

Assessing daylight inside the homes you are building

Internal daylight to new flats is assessed against target illuminance over part of the reference plane for part of the year, not a single average daylight factor. How the methods differ and what fails.

Read article →