The catchment definition section of a retail impact assessment

A retail impact assessment stands or falls on its catchment definition, because every turnover, market share and trade diversion figure that follows is arithmetic performed inside the zones you draw. Draw them badly and the impact percentages are precise answers to the wrong question. The catchment section of a Retail Impact Assessment has three jobs: set the study area and its zones, justify the method used to draw them, and name every centre whose trade the application site could plausibly draw away.

The National Planning Policy Framework (August 2026 edition) applies a default floorspace threshold of 2,500 m2 gross for retail and leisure proposals outside a centre that are not in accordance with an up-to-date plan, unless the local plan sets a lower figure. Many adopted plans do set a lower one, sometimes 500 m2 in smaller market towns. Check the adopted plan before you scope anything.

Drawing the primary and secondary zones

A catchment is normally built from postcode sectors rather than smooth shapes on a map. That is a practical choice: population and expenditure data are published at that geography, and a household survey can be stratified by the same zones, so the survey results, the expenditure pot and the diversion matrix all line up without interpolation.

The primary zone is the area from which the proposal would draw the bulk of its trade. The secondary zone covers households who would use the store occasionally, usually on a linked trip or when passing. Some assessments add a tertiary zone for bulky goods, where people travel much further for a single purchase than they would for a weekly food shop.

Say plainly why each zone boundary sits where it does. A river with one crossing, a dual carriageway, a district boundary that changes the bus network: these are the reasons a zone stops, and an officer reading the report wants to see them stated rather than inferred from the shading.

Drive time, distance and the zone boundary problem

Four methods turn up in practice, and they are not interchangeable.

For a food store, a drive-time band of around 10 minutes for the primary zone and 20 for the secondary is a common starting point, then adjusted to sit on whole postcode sectors. For comparison and bulky goods, the bands widen considerably. The honest approach is to draw the isochrones, overlay the sector geography, and record every place you moved a boundary and why.

Urban schemes need more care. Where a large share of shopping trips are made on foot or by bus, drive time flatters the catchment in one direction and understates it in another. Say so in the text rather than leaving the reader to spot it.

Choosing which centres to assess

The list of centres to be assessed is where most arguments start. The Framework's impact test looks at the effect on existing, committed and planned public and private investment in a centre, and at the vitality and viability of town centres in the catchment. That means you assess centres, not just the nearest competing store.

Work through four groups:

1. Defined centres in the adopted plan that fall inside the catchment, at every tier in the hierarchy.
2. Defined centres just outside the catchment boundary where the local plan or a neighbouring authority has identified a regeneration scheme or a site allocation.
3. Out-of-centre stores that trade in the same category, because they affect the diversion arithmetic even though they are not centres for the purpose of the test.
4. Centres in adjoining authorities where the boundary cuts through the catchment. A cross-boundary centre left out of the assessment is the fastest route to an objection from the neighbouring council.

Planned investment deserves its own paragraph. If the local plan allocates a site for retail-led regeneration, or a scheme has permission but has not been built, the assessment has to consider whether trade diversion would undermine it. Pulling the relevant consents and allocations together is the same discipline as the planning history section of a planning statement, and the same sources answer both.

Household survey data and how old is too old

Most contested assessments rest on a zone-stratified household survey that asks where people do their main food shop, where they go for top-up, and where they buy each non-food category. The survey converts into market shares by zone, and those shares drive the diversion.

Two things get challenged. The first is sample size per zone, because a zone with a thin sample produces a market share with a wide margin of error that then gets applied to a large expenditure pot. The second is age. Councils commonly push back on surveys more than about three years old, and shopping patterns since 2020 have moved enough that older data invites a fight. There is no national rule on survey age, so ask the case officer at pre-application stage what they will accept, and record the answer.

If you are relying on a survey commissioned by the council for its own retail study, check the fieldwork date rather than the publication date of the study. They are often two years apart.

Turnover densities and where the numbers come from

Company sales densities, applied to net sales area, generate the turnover of existing stores and the projected turnover of the proposal. Use published company accounts and recognised published market data, state which source and which year, and adjust where a store is known to trade above or below its company average.

Then set out the expenditure side: population by zone, expenditure per head by category, growth assumptions, and the deduction for special forms of trading. Online shopping is the deduction that gets argued over hardest, because a generous allowance shrinks the available pot and makes the proposal's impact look worse, while a mean one does the opposite. Show the figure, show its source, and show the sensitivity if it moves by a few percentage points.

An illustrative example. Take a proposed 3,000 m2 gross food store, around 2,000 m2 net. Apply the operator's published sales density, deduct the online share, and you have a design-year turnover. Split that turnover across the zones using the survey-derived market shares. The resulting diversion from each centre is the number the whole assessment turns on, and every step in that chain should be traceable in a table. Treat those figures as illustration of the method, not as a benchmark.

Presenting trade diversion

Give the diversion twice: as absolute expenditure drawn from each centre, and as a percentage of that centre's existing turnover. The percentage is what a member reads, and the absolute figure is what an inspector checks.

Set out a base year and a design year, and explain the gap. A design year too far ahead lets forecast expenditure growth absorb the diversion and produce a comfortable result. Test the cumulative position as well as the solus one, with committed schemes in the catchment included. Then apply judgement: the Framework asks whether the adverse impact is significantly adverse, which is a planning conclusion about the health of the centre, not a threshold percentage.

Keep the retail sequential test in its own section and its own logic. It asks about site availability and suitability rather than trade, and it is a different exercise again from the flood sequential test, which readers sometimes conflate. If that comparison is useful for your own team, the sequential test section inside a flood risk assessment sets out how the flood version works.

Where the assessment is most often attacked

Catchments get challenged on a short list of recurring points. A zone boundary that excludes a centre which would otherwise show material diversion. Survey fieldwork older than the council's retail study. A turnover density taken from a year when the operator traded unusually well. An online deduction with no cited source. A design year chosen to suit the answer. And, often decisive, an assessment that treats planned investment as background rather than as one of the things the test protects.

None of these are fatal if the report anticipates them. Write the method section so that a reader who disagrees with one assumption can see exactly which table changes, and offer the sensitivity test rather than waiting to be asked for it.

Retail evidence sits alongside a wider pack for a major application, and the supporting documents interact: browse the 66 report types to see which of them your scheme is likely to need before you commission anything.

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