Why a site visit still decides the design and access statement

A site visit for a design and access statement supplies the content nothing else can: what the street is actually made of, where the levels fall, which windows look at the site and from what height, and how people currently move across and around it. A desk-based draft can assemble the policy framework, the planning history and the constraints, and it can do that well. What it cannot do is see. Every design and access statement that reads as generic is generic in the same place, which is the context analysis, and case officers spot it instantly because they know the street.

So the question is not whether to visit. It is what to look for, and how to record it so that the observations end up in the document rather than in your memory.

What a desk based draft cannot see

Aerial imagery and street-level photography are useful and they are out of date. A satellite view shows roof forms and boundaries. It does not show that the render on the neighbouring pair has been replaced with insulated board, that the boundary wall is a different brick from the houses, or that the pavement narrows to 900mm outside the site.

Four categories of information reliably require a visit.

• Material reality: what the walls, roofs, windows, thresholds and boundaries are actually made of, and their condition
• Level information: the fall across the site, the relationship to the pavement, the position of any retaining structures
• Occupied relationships: which windows overlook, at what height, and what they currently see
• Movement and use: where people walk, where cars park in practice rather than in the marked bays

Each of those turns into a paragraph in the Design and Access Statement, and each one is the sort of specific that persuades an officer the applicant understands the place.

Materials and the grain of the street

Officers assess design against local character, and the evidence for local character is the street. What matters is the specific: brick type and bond, mortar colour and joint profile, window proportion and reveal depth, whether cills are stone or concrete, the roof pitch and covering, the eaves detail, and the rhythm of solid to void across the frontage.

Go with a notebook and record the things you would otherwise guess. Count the storeys on each adjoining property. Note the ridge height relationship. Photograph a typical window including the reveal, because reveal depth is the single detail that most often distinguishes a scheme that reads correctly from one that does not.

Look for the exceptions as well. A street of 1930s semis with one 1970s infill tells you what the authority has accepted before and what the design guide was reacting to. That observation is worth a sentence in the statement, and it is not available from a plan.

Levels, drainage and boundary conditions

Levels decide more than people expect. A rear extension on a site falling 800mm away from the house needs steps, a retaining wall or a raised deck, and each of those changes the overlooking picture and the neighbour's objection.

On site, record the fall across the site in both directions, the threshold level relative to the pavement, the position of manholes and gullies, and any evidence of water: staining on walls, moss lines, silt deposits at low points. Photograph the boundary on both sides at the same time, because boundary ownership and construction are a frequent source of dispute and a dated photograph settles it.

Drainage evidence found on site feeds into the drainage work rather than the design statement, but it should be found on the same visit. A gully serving the rear of the site is worth knowing about before the design fixes a patio level over it.

Neighbouring windows and overlooking

This is the observation with the greatest effect on the design, and the one most often guessed.

Stand where the proposed development will be and look. Note every window that faces the site, its floor level, its approximate size, and whether it is obscured glazed or has a permanent screen. Note whether the room behind it appears to be habitable. Where a window is at first floor and faces a proposed terrace, that is a design problem which is far cheaper to solve now than at committee.

Then reverse the exercise. Stand in the position of the neighbour's window as best you can from the pavement, and assess what the proposal would present to it. The Overshadowing Study and any daylight work will quantify the effect later, but the design response is decided by what you saw.

Do this politely and from land you are entitled to be on. Knocking on a door and asking is often more productive than it sounds, and a neighbour consulted before the application is a neighbour less likely to object to it.

Access, footways and crossing points

The access half of a design and access statement is the half most often thin, and it is the half that is easiest to evidence on a visit.

Record the footway width at the narrowest point, the presence and condition of dropped kerbs, the position of the nearest formal crossing, the bus stop distance and whether it has a shelter, the parking controls and their hours, and the gradients on any approach. Note whether the existing access has adequate visibility in both directions and what obstructs it: a hedge, a parked car, a boundary wall.

Photograph the approach from both directions at eye level, and photograph the site entrance from across the road. Those two images do more for an access narrative than three paragraphs of description.

Where the site is likely to need highway authority input, the observations also tell you whether the Transport Statement or a fuller assessment will be needed, and that affects the programme rather than just the document.

Photographing the site methodically

An unstructured set of forty phone photographs is nearly useless three weeks later. A structured set of fifteen is worth more.

1. Number a plan with photograph positions and arrows showing the direction of view, and follow it.
2. Take a wide shot and a detail shot at each position.
3. Include something of known size in detail shots, a door or a brick course, so scale can be read later.
4. Photograph each boundary along its whole length, corners included.
5. Photograph the site from the street at the points where the proposal will be visible.
6. Note the date, time and weather, because shadow direction in the photographs will be asked about.
7. Photograph anything you would otherwise have to describe from memory.

Do the numbered plan first. Retrofitting positions onto photographs afterwards takes longer than planning them, and photographs without positions cannot be used as figures in the statement.

Feeding observations back into the draft

The failure mode is a good visit and a document that does not reflect it. Two habits prevent that.

Write the context section on the day, in rough, while the observations are fresh. Not polished text, just the specifics: brick type, ridge relationship, footway width, the window that matters. Then the draft has facts to be built around rather than adjectives to be filled in.

And replace every general statement in the draft with something you saw. "The site sits within an established residential area" becomes "the site is the end plot in a terrace of six 1890s two-storey houses in yellow stock brick with slate roofs and stone cills". The first sentence could describe most of England. The second could only describe one street, and that is the whole point of the document.

PlanScope's drafts insert "[TO BE CONFIRMED]" flags where site-specific facts are missing rather than inventing them, and those flags are the list of things to take to site. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. Drafts can be sent for review through the professional directory once the site observations are in.