Neutral months for traffic surveys and why they matter

The question that decides most traffic survey arguments is not which months are neutral. It is whether the count you already hold is still usable, and if it is not, how long the re-survey will add to the programme. Neutral survey months traffic conventions exist to answer the first half of that. The second half is negotiated with the highway authority, at scoping, before anyone goes into the field.

Get the scoping conversation right and a three-year-old count with a growth factor will carry a transport assessment. Get it wrong and you are waiting until the following May.

What a neutral month means in practice

A neutral period is one in which observed traffic is a fair picture of the typical condition across the year. The conventions are not written into policy. They come from survey practice, from trip rate database protocols, and from what highway authorities have accepted for decades.

The working definition has three parts. The month must be outside the main holiday periods. The days must be ordinary weekdays, with Tuesday, Wednesday and Thursday preferred. And the week must not contain or immediately follow a bank holiday.

Surveys that meet all three are accepted without argument. Surveys that fail one can often be defended with evidence. Surveys that fail two are usually re-run.

The month-by-month position is set out in traffic count neutral months, and there is little value in repeating it here. What follows is the part that is harder to find: how to use the conventions when the ideal survey window has already passed.

Periods usually excluded and why

August and late December are excluded for obvious reasons, and the exclusions around them are less obvious.

The week before a school holiday often shows reduced traffic as families travel early. The first week back after a long break shows higher flows as everyone returns at once. Half term weeks affect the morning peak far more than the evening peak, because the school run disappears while commuting continues. And in retail locations the six weeks before Christmas produce flows that would be unrepresentative in the other direction.

Local effects matter as much as national ones. A university town empties in late June and refills in late September, on a calendar that does not match school terms. A coastal authority sees summer flows that dwarf the annual average. An agricultural district has a harvest period with heavy vehicle movements that would distort a count in either direction depending on the road.

These local patterns are precisely what the highway authority knows and you may not. Ask.

School terms and the morning peak

The school run is the single largest seasonal variable on most local networks, and it is worth understanding its shape rather than simply avoiding holidays.

Its effect concentrates in a short window, typically between 08:15 and 09:00, and it is directional towards schools rather than towards employment centres. On a road serving a large primary school, the morning peak during term can be 20 minutes earlier and noticeably sharper than the peak during a holiday week.

Two practical consequences. A count taken in a holiday week understates the morning peak on residential and school-serving roads, sometimes substantially, and a highway officer will reject it for exactly that reason. And the peak hour you identify from a holiday count may be the wrong hour, which then propagates through the whole assessment.

Where the development itself generates school trips, the interaction becomes the assessment rather than a survey technicality, and it belongs in the Transport Assessment explicitly.

Agreeing survey dates at scoping

This is the step that saves the most time and is most often skipped.

Write to the highway authority before commissioning anything, setting out the proposed survey locations, the survey dates, the survey types, the duration and the peak periods to be covered. Ask three questions:

1. Does the authority hold recent counts at any of these locations that can be used instead?
2. Are there local calendar effects that would make the proposed dates unrepresentative?
3. Will the authority accept the proposed survey as the basis for the assessment, subject to the results?

A written answer to the third question is the thing worth having. It does not bind the authority on the outcome of the assessment, but it does close off the argument that the baseline is invalid, which is otherwise the easiest objection for a consultee to raise late.

Scoping should also settle the study area, the assessment years, the committed development to be included and the growth methodology, because those decisions interact with the survey. There is no point counting at a junction that everyone later agrees is outside the study area.

Where trip generation itself is contested, the threshold questions belong in the scoping note rather than in the survey specification. Parking demand surveys carry their own timing rules, and a Parking Survey following the recognised overnight methodology has to avoid holiday periods for the same reasons.

Using older counts with growth factors

Most authorities treat traffic count data as usable for around three years, and some accept older data where the network and land use around the count point have not changed.

The method is to apply growth factors from the national trip end forecasting dataset to bring the base year flows forward to the assessment year. It is routine, it is accepted, and it has three failure points worth checking.

The first is local change. A new supermarket, a housing scheme completing, a school expanding or a road closure within the count's catchment invalidates the assumption that growth alone explains the difference. Check what has been permitted and built near the count point since the survey.

The second is the pandemic period. Counts taken during 2020 and 2021 are not representative of anything, and counts taken in 2022 sit in a period when travel patterns were still unsettled. Most authorities will not accept a base year from those years without a validation count, and it is reasonable for them to ask.

The third is the direction of the adjustment. Growth factors generally increase flows, which increases the assessed impact of the development. An applicant relying on old counts is not gaming the system; they are usually accepting a slightly more onerous baseline in exchange for programme certainty.

A short validation count, a single day at one location, comparing current flows against the factored historic figure, resolves most disputes about whether older data still holds, at a fraction of the cost of a full survey programme.

When a re-survey is unavoidable

Four situations where there is no way around new fieldwork.

• No usable count exists at the locations the study area requires.
• The available data predates a material change to the network or to nearby land use.
• The authority has declined to accept the factored baseline in writing.
• The assessment depends on turning movements, queue lengths or journey times that the historic data does not contain.

That last case is common. An authority holding automatic traffic counter data on a link cannot supply the junction turning counts a capacity assessment needs, and classified counts have to be commissioned regardless of how good the link data is.

Where re-survey is unavoidable and the season has passed, options are limited but not absent. A survey in a marginal period, presented with a comparison against historic data from the same location in a neutral period, can be adjusted and accepted. A survey in a clearly unrepresentative period generally cannot.

Building the survey into the programme

Treat survey timing as a fixed point in the planning programme, in the same way as ecological survey seasons.

Work backwards from the target submission date. The assessment needs perhaps six to eight weeks after the data arrives. The data arrives a week or two after the survey. The survey has to fall in a neutral period. And the scoping exchange that agrees the survey needs several weeks before that.

Miss the autumn window and the next reliable one is the spring, which is the same programme risk that ecology brings, described for habitat work in botanical survey season and habitat condition assessment.

Where the scheme falls below the threshold for a full assessment, a Transport Statement still needs a defensible baseline, and the same survey conventions apply to it.

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