Footfall heatmap, city centre edition: mapping presence block by block
Cover a downtown with a grid of points, read each block's average day and busiest hour, colour them in bands and see where the crowd really gathers, and when.
Part of the guide Location intelligence automation
To build a footfall heatmap, city centre streets are covered with an even grid of points, each point is measured on its own small block of the map, and the blocks are coloured by how busy their average day is. In xMap Flows that is a table of grid points, the Mobility Footfall block set to each point's own block, a Calculated field that sorts the results into bands, and a map coloured by band.
It suits planners, retail and property analysts, outdoor advertising sellers and anyone who has heard "the busy end of the street" and wants to see it. Below: how to lay the grid, how to read hot and cold blocks, a made-up example, the usual errors, and the flow.
Why hexagons, and how big they are
xMap's mobility data is counted on H3 hexagons, a grid of cells that tile the earth. The cell Mobility Footfall reads when its circle is 200 m or less is about 0.1 square kilometres, roughly 350 m from side to side. So a point every 350 m or so, set to 200 m, reads one cell each with no overlap and no gaps. Wider circles add up every cell whose middle is inside, which smooths the picture and blurs the hot spots you are trying to find.
A run measures up to 50 places. A grid at that spacing over a centre about 2 km by 2 km needs roughly 36 points, which fits in one run. A larger downtown can be split into two or three tables, each run on its own.
Laying the grid
- Draw the boundary you care about: the old town, the business district, the area inside a ring road.
- Make a CSV of points every 350 m inside it, with columns name, lat and lng. Name each point by its row and column, such as B4, so the map legend is easy to talk about.
- Upload it on the My data page. It becomes a table the My data block can read.
- Add a few named points on top of the grid for places you already know: the main station, the market, the cathedral square. They help you check the picture against what you see on the ground.
Reading hot and cold blocks
Footfall is people-hours of presence: a block with offices full of workers from nine to six can look as hot as a shopping street with a fast turnover of visitors. The total alone hides that, so read each block's busiest hour and the split of its day next to it:
| What you see | What it usually means |
|---|---|
| Hot all day, peak at midday | Offices and lunch trade |
| Hot, with most of the day in the evening | Restaurants, bars, cinemas |
| Hot in the morning and early evening only | A station or a bus interchange |
| Warm, busiest at the weekend | Shopping and leisure streets |
| Cold next to a hot block | A barrier: a river, a rail line, a fenced site, a one-way road nobody walks along |
A rough guide to reading a block, to be checked on the ground.
The cold blocks are often the most useful. A cold block wedged between two hot ones is a gap in the walking routes, and the reason is usually physical. A cold block on the edge is simply where the centre stops.
A made-up example
Nothing here is a real reading. A team maps the centre of Lisbon with 34 grid points and five named ones. The hottest cells sit along the main square and the streets running down to the river, with their busiest hour in late afternoon. Two cells near the main station peak at 08:00 and 18:00 and are quiet between. One cell between the shopping streets and the hill district is cold, and a site visit finds a long stretch of blank walls and a steep stair. The team repeats the exercise in Mumbai's Fort area: the business streets peak at midday and empty in the evening, while a market cell is hottest after dark.
For a shop that lives on browsing, the warm weekend cells beat the hot commuter cells. For a breakfast kiosk, the station cells are the answer.
Mistakes to avoid
- Mixing circle sizes. A grid where some points use 200 m and others 400 m compares different areas. Keep one setting.
- Leaving holes. A missing point looks like a cold block on the map. Check every row came back with footfall_days filled in.
- Reading a quiet day as a quiet block. Use the last 30 days, not 7, so one rainy week does not paint a street blue.
- Treating no data as zero. A cell that says no footfall data here yet has no figures, which is different from no people.
- Forgetting the delay. Each day arrives about ten days late, and the data starts on 17 October 2026, so a long window may hold fewer days than its name; footfall_from and footfall_to say which.
The flow that draws it every month
- Schedule: the first Monday of each month.
- My data: the grid table.
- Mobility Footfall: 200 m, last 30 days, returning Footfall and Busy hours.
- Calculated field: a line such as category = if(footfall_daily > 40000, "Busiest", if(footfall_daily > 15000, "Busy", "Quiet")), with thresholds you choose after the first run.
- Show on a map: colour points by Category, updating the same map each month so its link stays the same.
- Email: the map link and the ten busiest cells with their peak hour.
The inbox gets a map with three colours, a legend you can click to hide a band, and a short table. The map can be shared by a public link with a landlord or a council officer. For ideas that build on it, see Score every neighbourhood for a new branch and Where to put the next billboard, and the location intelligence automation guide.
Questions
How many points do I need for a city centre?
One every 350 m or so. A centre about 2 km across needs around 36 points, which fits in one run of up to 50 places.
Can the map show a smooth heat layer instead of coloured points?
The saved map can switch to a heat view, but the clearest reading of footfall is points coloured by band, because each point is one block with its own figure.
Why is a block next to a busy street cold?
Usually something on the ground: a river, a rail line, a fenced site or a road people do not walk along. Check it in person before drawing conclusions.
Does this work outside big cities?
Yes, wherever footfall is covered, which is most of the world but not yet Australia. Smaller centres need fewer points and a longer window.