The catchment question: where do your visitors actually come from?

Most sites hold a confident, detailed theory about their visitors, built from years of operating in a place. It is also frequently wrong, and the errors land on decisions that matter a great deal: where to open next, how to allocate a marketing budget, which tenant to put in the unit coming free in spring. Measuring where visitors come from is what catchment analysis is for, and the intuitive model breaks down once you do it.

The short version

• A catchment area is the zone a store, shopping centre or venue draws its visitors from, usually divided into primary, secondary and tertiary zones.
• Car-park plates, loyalty card postcodes and staff recall are heavily filtered signals, so the intuitive model overweights the zones that make themselves visible.
• Cellular network analytics gives visitor counts by origin district per day and season, anonymised and aggregated, with no individual identified.
• For a new site, how many people live nearby is only a starting point; the question is how many people from where will come. The same origin counts serve tenant mix, marketing spend and public services.

What is a retail catchment area?

A catchment area is the geographic zone from which a location draws its visitors. Retail geographers conventionally divide it into three zones: a primary area covering the nearest population that generates the large majority of visits, a secondary zone contributing a further share, and a tertiary zone of occasional and destination-driven visitors.

Catchment analysis at district scale works without surveys or loyalty card inference and shows where visitors originate before they appear at your location. Cellular network analytics makes this possible at the scale real planning decisions need.

The resulting map is often surprising. A shopping centre with a strong regional reputation may draw most of its visitors from two or three adjacent districts, with the district three kilometres away barely visible. Or the reverse: a location that the operator considers local turns out to pull heavily from a distant residential area with no obvious alternative. Either pattern can work for a business, and both affect every decision that depends on who your customer base is.

Why does the intuitive model fail?

The intuitive model fails because operators build catchment assumptions from what they can observe: licence plates in the car park, accents at the information desk, postcodes on delivery orders. These are real signals, but heavily filtered ones. The car-park sample misses everyone who arrived on foot. Loyalty card postcodes only cover cardholders, who are not a random slice of visitors. Staff recall anchors on memorable conversations and unusual cases, while the median visitor who arrives, shops, and leaves without incident barely registers.

The resulting model overweights the catchment zones that make themselves visible and underweights the ones that don’t. Marketing spend, signage placement, and transport negotiation all flow from that model, so the error compounds over years without anyone noticing.

How does catchment data change a site selection decision?

For a new site, “how many people live nearby” is only a starting point; the question catchment data answers is “how many people from where will come.” Site selection analysis is where catchment errors are most expensive. A site in a city centre with good public transport connections may draw from a much wider area than a retail park on the urban fringe despite having a smaller residential population within walking distance.

Population figures can also mislead in the other direction. A market town that already has a well-established shopping destination may appear to have attractive population numbers until the catchment data shows that the existing destination has already captured the effective audience, and the new site would be competing for visitors who have already settled into a habit. Getting this wrong is expensive, so the analysis belongs before the lease is signed.

How do malls use origin data for tenant mix?

Mall operators and property owners use catchment analysis in two directions at once. The first is inward: understanding which tenant categories over- or under-index for the actual visitor base the mall draws. A centre that attracts a disproportionate share of its visitors from younger urban districts may be underserved on categories that appeal to that audience, regardless of what the conventional mall formula says.

The second is outward: demonstrating the catchment to prospective tenants. An anchor tenant negotiating a lease wants the headline footfall number, and also where those visitors come from and whether that population overlaps with the brand’s own customer base. Origin data gives the landlord something concrete to show that headline traffic figures cannot.

How does origin data steer marketing spend?

District-level origin data changes how marketing budgets are allocated. If the catchment shows that visits from a district ten kilometres away have been declining over three years, perhaps because a competitor opened in that area, that is a different problem from a decline that affects the whole catchment uniformly. The first calls for targeted communications and possible partnership with local transport; the second suggests a broader brand or product issue.

For cities and municipalities, the same logic applies to public services and leisure venues. A library that draws almost entirely from the surrounding district is serving a narrow catchment that may reflect transport barriers more than demand in the wider area. A cultural venue trying to demonstrate its regional significance to funders can use origin data to show exactly which districts it reaches and which remain underserved. How people move once they are in the district is a separate question, and the footfall pulse of a neighbourhood shows whether a high street really serves the people around it.

What does catchment data look like in practice?

The data deliverables from a cellular network analytics deployment give you visitor counts broken down by origin district per day and season, so you can see which districts are growing in contribution and which are shrinking. The routes visitors choose once they are on site come from Wi-Fi or sensor counting, not from network data. With that, you can compare your catchment profile to similar locations and model what a new transport link or a competitor opening might do to the distribution.

All of this is anonymous and aggregated. Cellular network analytics works at the scale of districts and time periods; it does not work at the level of individual journeys or identifiable people. The dataset contains no home addresses, only the district a group of visitors originated from, expressed as a count. That level is enough for every decision described above, and catchment analysis has always made sense at that level.

The catchment you thought you had

Where visitors come from is the most basic question a site can ask about itself. It comes before conversion, dwell time and revenue per visit, because all of those questions depend on first knowing who is arriving.

Origin analytics, done at district scale, is built to correct the assumption most sites start from, without identifying anyone. That assumption is usually close enough to be plausible, but rarely close enough to stake a site selection decision on.

Bumbee Labs is the company behind the only footfall method in Europe approved by a data protection authority.