Choosing a depot location by drive time to customers

Choose a depot by drawing, around each candidate site, the area a van can reach within the same short drive, then counting what falls inside each area: the businesses you serve, the people who order from you, and the depots of rivals. The candidate whose drive-time area holds the most of your demand, with the fewest rivals sharing it, is usually the best place to start serious talks with landlords.

This is for operations directors, network planners and founders of delivery, wholesale, field-service and rental businesses opening a first depot in a city, or a second one to relieve the first. It covers why drive time beats distance for this decision, what to count inside each area, a worked example in Lyon, the traps, and an AI employee that runs the comparison.

Why drive time, not distance

A depot's job is to put vans at customers' doors quickly. Two sites 5 km from the city centre can be very different: one sits by a ring road and reaches half the city in a quarter of an hour, the other sits behind a river with two bridges and reaches far less. A circle on a map treats them as equal. A drive-time area follows the roads, so it shows the difference.

The Travel area block draws that area around each candidate: within a travel time by car, on foot or by bike, or within a distance along the roads. For now it reaches up to 15 minutes or 10 km by road. When the routing service cannot draw an area, the block falls back to a circle at a typical straight-line speed and says plainly that the area is an estimate, in the step and in every email.

What to count inside each area

• Your customers. A table of your accounts or recent delivery addresses, uploaded to My data, shows how much of today's work each candidate would reach.
• Future customers. Nearby places set to search the area counts the kinds of business you sell to: restaurants for a food wholesaler, garages for a parts distributor, clinics for a medical courier.
• Residents. For home delivery, Demographics gives the people living inside each area, in countries the population grid covers (France, Canada, India and Australia among them, not the United Kingdom or the United States).
• Rivals. Other depots, cash-and-carry warehouses or courier hubs inside the same area share the same customers.
• The roads at the door. Traffic Flow at each candidate, measured at the hour vans would leave, shows whether the first minutes of every route start in a jam.

A worked example: a food wholesaler in Lyon

The figures here are made up. A food wholesaler supplying restaurants and cafes wants its first depot in Lyon and has three candidate units: one on the eastern ring road, one in an older industrial area by the river, one north of the city near the motorway.

For each candidate, the flow draws a 15-minute drive area, counts restaurants and cafes inside it, counts residents inside it, and checks road speeds at the gate at 06:00. The eastern unit reaches the most restaurants and cafes and the most residents. The river unit reaches fewer, because the bridges cut its area in two. The northern unit reaches the fewest restaurants but has free-flowing roads at 06:00. The wholesaler shortlists the eastern unit, keeps the northern one as a second choice for a later delivery-only depot, and drops the river site.

Traps when comparing depots

• Drawing at an unrealistic hour. A drive-time area describes typical travel, not the 07:30 crawl. Pair it with Traffic Flow at the hours vans leave.
• Counting every business. Count only the kinds you serve. A wholesaler that counts banks and offices will pick the wrong site.
• Trusting an estimate as real. If the step says an area was estimated, the comparison is rougher. Run it again later or treat the result with care.
• Forgetting rivals. A site that reaches many restaurants also reaches the rival who already supplies them.
• Comparing more than a few sites at once. Three to five candidates keep the comparison readable; more turns it into a search, which is a different job.

The AI employee that compares the candidates

1. My data: the candidate sites, each with a name and position.
2. Travel area: 15 minutes by car around each incoming place.
3. Nearby places: the kinds of business you serve, inside the areas; each comes back named with the area it fell in.
4. Demographics: residents in each area, where the country is covered.
5. xMap LLM, the whole list: a short comparison of the candidates, counts first.
6. Make a report: a PDF for the property meeting.

What lands is a short report: for each candidate, what it reaches, how the roads look at the gate, and a plain recommendation with its reasons. For two or three exact spots, Compare sites in the app measures them side by side without building a flow. Templates nearby: Who my warehouse reaches in 15 minutes, for same-day delivery, Warehouses within 15 minutes' drive of the port and Where to put micro-fulfilment hubs.