Competence for a parking survey and its analysis

Who carries out a parking survey is a question with no licence attached, which is why the results vary so much in quality. There is no chartered discipline for counting parked cars at two in the morning. What there is, is a published methodology that most London boroughs and a growing number of other authorities require, and a set of recurring errors that cause a highway officer to reject a survey and ask for it again at your cost.

The competence that matters divides in two: the fieldwork, which needs trained enumerators and careful recording, and the analysis, which needs a transport planner who understands what the stress figure does and does not show.

What the methodology asks of the surveyor

The overnight beat survey methodology that authorities refer to asks for a count of parked vehicles and a measurement of available kerbside capacity within a defined catchment around the site, carried out in the small hours when residential demand is at its highest.

The parameters that recur are two survey nights, one midweek and one at the weekend, within a window in the early hours, covering streets within a short walking distance of the site. Boroughs and county highway authorities set their own variations on the night, the window, the radius and the capacity assumption, and those variations are not interchangeable.

So the first competence test is procedural. Ask the surveyor which authority's version they are working to, and ask them to show you the authority's own requirement in writing. A surveyor who cannot name the source of their parameters has not checked.

Defining the survey area without bias

The catchment is where bias enters, and it is the thing an officer checks first.

The area should be drawn as the streets a resident of the development would realistically park in, measured as walking distance along the street network rather than as a radius on a plan. That distinction matters where a railway line, a river or a dual carriageway sits within the nominal radius, since a street two hundred metres away as the crow flies may be a fifteen minute walk.

Three errors recur. Extending the catchment until it picks up quieter streets further out, which dilutes the stress figure. Stopping the catchment short of a street that is plainly within walking distance because it is heavily parked. And drawing the boundary mid-street, so part of a road is counted and part is not.

Record the catchment on a plan with every street named and the boundary explained in a sentence. Where the authority has specified the catchment, say so and attach the correspondence.

Night working, training and recording

The fieldwork is unglamorous and the quality depends on it.

Enumerators need training on three things: classifying vehicles consistently, recording accurately against a street-by-street schedule, and not missing vehicles parked on frontages, in bays obscured by larger vehicles or across private accesses. They also need a night working risk assessment, high visibility clothing, identification, and in most areas a courtesy notification to the local police, because people walking residential streets at three in the morning with a clipboard attract attention.

Recording should be contemporaneous and timestamped. A tablet-based count with GPS logging is now standard and makes the audit trail straightforward. A paper count transcribed the next morning is not wrong, but it is harder to defend if challenged.

Note the conditions. Date, day of the week, start and finish times for each street, weather, and anything unusual such as roadworks, a street party or a nearby event. Those notes answer the officer's first question about whether the night was representative.

Measuring the capacity, which is where errors hide

Parking stress is a fraction, and most disputes are about the denominator rather than the numerator.

Available capacity is the usable kerbside length divided by an assumed length per vehicle, and both parts of that go wrong.

Usable length excludes everything the highway code and the local restrictions take out of use: dropped kerbs and vehicle crossovers, junction protection, bus stops and clearways, loading bays during their hours of operation, double yellow lines, zig-zags outside schools, and taxi ranks. On a terraced street where most houses have a crossover, the usable length can be a fraction of the gross kerb length, and a survey that measures the street end to end overstates capacity badly.

The length per vehicle assumption also needs stating. A figure of five to six metres per car in unmarked kerb space is the normal range, with a longer allowance where a space sits between two obstructions and the vehicle has to be manoeuvred in. Marked bays are counted as bays. Use the figure the authority specifies and state it in the report.

Measure the deductions rather than estimating them. This is the single most effective thing a surveyor can do to make a survey defensible.

Quality control before the report goes out

Four checks, none of them costly.

Independently re-count a sample of streets on the same night and compare. A discrepancy above a small tolerance means the recording protocol needs tightening before the data is used.

Photograph each street, timestamped, so the condition can be verified later.

Sense-check the two nights against each other. A street at ninety per cent on the midweek night and fifty per cent at the weekend needs an explanation, and it is usually a commuter or a visitor effect worth describing rather than smoothing over.

Have someone who was not on the survey read the capacity schedule against the plan. Transposed street names and double-counted frontages are common and embarrassing.

The stress calculation and who should interpret it

The calculation is parked vehicles divided by capacity, expressed as a percentage, reported per street and in total for the catchment.

Interpretation is a transport planner's job, and it involves three judgements. Whether the baseline shows constrained conditions, which authorities commonly treat as the case above around eighty per cent, with severe constraint higher still. What additional on-street demand the development would generate, derived from local car ownership data for the relevant area rather than from a national average. And whether the resulting figure is acceptable against the authority's standards and the policy position on car-free or car-capped development.

Be straight in the report about what the survey does not show. It is a snapshot of overnight residential demand on two nights. It says nothing about daytime demand, nothing about seasonal variation, and nothing about how behaviour changes when a new development arrives. A report that claims more than that invites a challenge on all of it.

The report should carry the surveyor, the analyst, their qualifications and the dates, and our note on the named author and date on a planning report explains how much that front page affects the way an officer reads the rest.

How the highway authority reviews the result

Officers check in a predictable sequence, so check the same things first.

Was the survey carried out on the specified nights, within the specified window, outside school and public holidays. Does the catchment match what was agreed or what the methodology requires. Were the capacity deductions measured and listed. Is the length per vehicle the authority's figure. Does the development demand come from local data. Is the conclusion supported by the per-street figures rather than by the catchment average, which can hide a saturated street.

That last point catches a lot of reports. A catchment averaging seventy per cent with the three streets nearest the site at ninety-five per cent will be read as showing a problem, and a conclusion resting on the average will be rejected.

Where the authority's own officers disagree with the analysis, the resolution is usually at the level of one input. Our note on changing consultant mid-application covers how to keep the survey data usable if the team changes while this is being argued.

When a repeat survey is unavoidable

Four situations mean doing it again.

The survey fell in a school holiday, a public holiday week or an unusual period, including anything that changed local travel patterns. The nights or the window did not match the authority's requirement. The capacity was measured without deductions. Or the survey is more than about a year old and the authority asks for current data.

Budget for the possibility on any application where parking is the main issue, and reduce the risk by agreeing the methodology, the catchment and the survey nights with the highway authority in writing before anyone goes out. That email is worth more than any amount of argument afterwards.

A Parking Survey report can be structured before the fieldwork so the surveyor knows exactly what to record, with the analysis and the Transport Statement built on it once the counts are in. Reviewers with the relevant transport competence can be approached through the professional directory.

This is general guidance, not legal advice. Check the position 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. For specialist technical reports, qualified professional input may be required.