How to find businesses near a train station, within a real walk
Draw a 5, 10 or 15 minute walk around one station or a whole line, then list the shops, cafes and offices inside it. Why a walk beats a circle, with a worked example.
Part of the guide Location intelligence automation
To find the businesses near a train station, start from the station's position, draw the area a person can walk in 5, 10 or 15 minutes along real streets, and search a places database for the kinds of business you want inside that area. Do the same for every station on a line and you get one list per station, each place tagged with the station it is near.
Station surroundings matter to many teams: distributors of newspapers, snacks and drinks; landlords marketing a unit near a new line; coworking and gym operators; commuter-focused cafes; and transport planners checking what passengers can reach. This guide explains why a walk is better than a radius, how to cover a whole line, a made-up example and the mistakes that throw the list off.
A walk, not a circle
A 500 m circle around a station looks tidy but treats a railway cutting, a river or a fenced depot as if people could walk straight across. On the far side of the tracks, a cafe 300 m away as the crow flies can be a 12-minute walk via the nearest bridge. In xMap Flows, the Travel area block draws the area reachable on foot within a set time along the road network, and the blocks after it search only inside that shape.
Pick the time by purpose:
- 5 minutes: the station concourse and the first block around each exit. Right for kiosks, newsstands and snack shops that live on passing commuters.
- 10 minutes: what people will walk to at lunch or after work. Right for cafes, gyms, pharmacies and dry cleaners.
- 15 minutes: the edge of what most people treat as near the station. Right for offices, clinics and landlords describing a location.
Finding the stations first
You can type each station in yourself, or let the flow find them. Nearby places has station types such as Train station and Subway station; search them across the city and you get each station with its position. Travel area then takes up to 25 incoming places a run, so a long line or a whole metro network may need to be split into runs of 25 stations.
A worked example
This example is made up. A snack and drinks distributor in Seoul wants independent convenience stores and newsstands near the stations of one subway line, to offer a station display stand.
- Nearby places lists the line's 22 stations, using the Subway station type within the area the line runs through, and the rep removes the few that belong to other lines.
- Travel area draws a 5-minute walk around each of the 22.
- A second Nearby places searches inside those walks for convenience stores and newsstands, independents only.
- The run returns 140 shops, each with a field saying which station's walk it was found in.
- Sort and limit by station name, so the list reads in the order of the line.
- Reps work the line from one end to the other, a few stations a day.
Because the walk follows the streets, two stations close together share some shops; Remove duplicates keeps each shop once.
What else to measure around a station
- How busy the area is. Mobility Footfall gives people-hours of presence by hour and day, so the commuter peaks show. It measures presence, not passengers, and arrives about 10 days late; there is none for Australia.
- Who lives near it. Demographics counts residents inside the walk in countries with population data, such as France, Canada, India and Thailand, but not the United Kingdom or the United States.
- What is missing. Counting a type inside each walk, such as pharmacies, shows the stations with none.
Mistakes to avoid
- One radius for every station. A city-centre interchange and a suburban halt need different distances.
- Ignoring the tracks. A straight circle puts businesses on the wrong side of the railway in your list.
- Mixing lines. A station served by three lines is one place; do not count its shops three times.
- Forgetting closed places. Leave Only places that are open switched on, or old units turn up in the list.
- Reading footfall as ridership. People-hours around a station include everyone nearby, not just passengers.
Running it as an AI employee
For a distributor or a landlord with a network to watch, an AI employee can redo the list each month and say what changed. The location intelligence automation guide covers the data behind it.
- Schedule, monthly.
- My data with your stations, or Nearby places to find them.
- Travel area: walking, 5, 10 or 15 minutes, around each incoming place.
- Nearby places inside the walks, with your types.
- Remove duplicates, then Only new.
- Show on a map and Email.
The monthly email lists new businesses station by station, with a map of the walks. Ready-made starts include Newsstands near stations, for magazine distribution, Busy stations with no convenience store and, for planners, Who and what is within a 15-minute walk of a planned station.
Questions
How far is near a station?
It depends on the business. Five minutes on foot suits kiosks and snack shops, ten minutes suits cafes and gyms, and fifteen minutes is about the edge of what people call near the station.
Why use a walking area instead of a radius?
A radius ignores tracks, rivers and fences. A walking area follows the streets, so a place across the railway with no bridge nearby is not counted as close.
Can I do every station on a line at once?
Yes. Travel area takes up to 25 incoming places a run, so a longer line or network can be run in parts.
Can I see how busy each station area is?
Mobility Footfall gives people-hours of presence by hour and day around each place. It is not a passenger count and it arrives about 10 days late.
Templates that do this
Read next
More on business and places data · Every article · Templates