Retail planning for municipalities

A radius does not show who is served

Whether a neighbourhood has shops nearby is often decided with a circle drawn around the nearest store. If a home is inside the circle, it counts as served. A railway line without a crossing, a wide main road or a river does not appear in that calculation. For someone who shops on foot, those barriers decide how long the walk is.

For everyday shopping, walking is the only option for many people: older residents, for example, and households without a car. An ageing population, rising mobility costs and the concentration of food retail make it more urgent to know which neighbourhoods are losing their local store, and which never had one.

In Germany, many municipalities steer retail development with a retail and centres strategy (Einzelhandels- und Zentrenkonzept). It sets out where new retail should go, defines the central shopping areas the municipality wants to protect, and feeds into land-use planning. For the strategy to hold, it needs clear evidence: who can reach which store, where a store is missing, and what a new or closed store would change. A radius cannot answer these questions reliably.

How we analyse access to retail

The method is an accessibility analysis with shops as the destinations. It starts with a definition of what counts as served: which shops count, by which mode, and within what time. That definition drives the result. A 10-minute walk gives a different map from a 15-minute walk, and both can be justified, so the threshold is always part of the result. Our glossary explains what makes a service gap.

Next, we calculate the catchment area of every store along the street network, on foot, by bicycle, by car or by public transport. Combined with population data, it shows how many people can reach a store in the chosen time and how many cannot. A supply gap is then a specific area with a known number of residents.

A heatmap gives the full picture. It shows the travel time to the nearest stores for every grid cell, or gives larger stores more weight than small ones. This makes thin coverage visible even inside areas that a radius would count as served.

Where several stores compete for the same residents, the Huff model estimates the share of demand each one attracts. It uses the attractiveness of each store, such as its sales floor area, and the distance to it.

Finally, we run scenarios. A planned store is added or an existing one removed, and the result is compared with today. You see how many more people a new site would serve, and who loses access when a store closes.

Retail analysis in GOAT

In GOAT, the tools are under Accessibility Indicators. Catchment Area calculates isochrones for Walk, Bicycle/Pedelec, Car and Public Transport, with time steps you choose yourself, such as 5, 10 and 15 minutes. Heatmap - Closest Average shows the average travel time to the nearest stores. Heatmap - Gravity weights stores by Destination Potential, for example by sales floor area. Huff Model calculates the share of demand each competing store attracts. The GOAT documentation explains each setting.

Add store locations, population data and administrative boundaries to your project from the Catalog, or upload your own data, such as your own survey of stores. You can save the steps as a Workflow. When a store opens or closes, run the analysis again with the same method.

For the City of Bonn, we are building a retail planning application on GOAT, embedded in the city’s Urban Data Platform. It is designed to run accessibility analyses, a local supply review scheme and forecasts automatically. In Freiburg, the city planning office and the local transport operator VAG have used GOAT on a subscription since 2022, including to review the city’s retail strategy.

You can run the analysis yourself in GOAT, or ask us to do it for your municipality. The projects are listed directly below.