What footfall data measures: people-hours of presence, explained

Footfall data measures presence: the total number of hours people spend in the area around a place, added up over a day. It is counted in people-hours, so one person who stays three hours counts three times, while ten people who pass by for six minutes each add up to about as much as one person staying an hour. It tells you how busy a spot is and when, not how many different people visited it, and each day's figures arrive about ten days after the day itself.

This explainer is for shop owners, site selectors, landlords, marketers and analysts who have been given a footfall figure and want to use it without overclaiming. It covers what the unit means, the figures you get, how to compare two places, the traps, and how to fetch the figures for your own sites with an AI employee.

People-hours, in plain words

Imagine counting, every hour, how many people are inside a small patch of the map, then adding those hourly counts across the day. That total is the day's footfall. The measure is built from a panel of devices and scaled up to the whole population, and it is only ever shown as totals for an area: never a person, a device or a single trip.

Two consequences follow. First, footfall rewards time: a park where people linger scores higher than a pavement people hurry along, even if more people use the pavement. Second, residents count too. People asleep at home are present, so a quiet residential street can show high night-time footfall. That is why the hour-by-hour split matters as much as the daily total.

The figures you get

The Mobility Footfall block adds these to each place you give it, for a circle of 50 m to 500 m around it:

• footfall_daily: the average day's people-hours, with footfall_days, the number of days that had data.
• footfall_peak_hour and footfall_peak: the busiest hour of the day in the place's own local time, and its average footfall.
• footfall_by_part: how the day splits between morning, midday, evening and late night.
• busiest_day and quietest_day: the days of the week. Weekends follow the place, so they are Friday and Saturday in most of the Gulf and Saturday and Sunday elsewhere.
• footfall_trend: the last 7 days against the 7 before, once there are 14 days or more.
• visitors_from and how those trips travel: where visitors come from. So far this is measured in Sao Paulo only.

Why the figures are about ten days behind

Each day's footfall arrives about ten days later. So a window of "the last 7 days" covers the seven days that ended roughly ten days ago, and the block shows the dates it used (footfall_from and footfall_to). When the newest days are later than usual, it uses the newest days there are and says so in footfall_note. The data starts on 17 October 2026, so the longer windows of 30 and 90 days fill in as time passes.

For most decisions a ten-day delay does not matter: busy hours and busy days are habits that change slowly. It does matter for anything that needs today's crowd, such as a one-off event, where footfall tells you about the pattern, not the day itself.

Comparing two places: an example

A made-up example. A cafe owner in Toronto, a city where xMap's places database lists 622 cafes within 10 km of the centre (November 2026), is choosing between two corners for a second shop. She asks for 30 days of footfall within 200 m of each. Suppose the answers are these (invented for illustration):

Invented figures to show how the fields read; not real figures for Toronto.

Corner 1 is busier overall, but its peak is late in the evening and a third of its footfall is late at night: likely residents at home and night-time trade, not morning coffee buyers. Corner 2 is quieter in total but peaks at 08:00 on a weekday, with most of its footfall in the morning and midday. For a cafe that makes its money before lunch, Corner 2 is the better fit, though a bigger total sat on the other side.

Traps to avoid

• Reading people-hours as people. A total of 48,000 does not mean 48,000 visitors.
• Comparing circles of different sizes. A 500 m circle gathers far more people-hours than a 100 m one. Use the same size for every place.
• Ignoring residents. Check footfall_by_part: a late-night share that dominates usually means homes, not shoppers.
• Expecting everywhere. Footfall covers most of the world from 17 October 2026, but not yet Australia. A place with no data says "No footfall data here yet", which is not the same as zero people.
• Treating the trend as news. A rise on the week before can be a holiday or the weather. Look at several weeks before acting.
• Following individuals. The figures are aggregates by design, and the Terms forbid using them to single anyone out.

Getting footfall for your sites

1. Bring in your places: My data with your stores, or Nearby places to find candidate sites in Paris, Mumbai or Sao Paulo.
2. Add Mobility Footfall. Choose the circle (200 m keeps each place to its own block of the map), the window, all days or weekdays or weekends, and which figures you want.
3. Add Sort and limit on footfall_daily to put the busiest first, or xMap LLM to suggest the best hours for staff or promotions at each.
4. Send it by Email or into a sheet. In the inbox: each place, its daily footfall, its peak hour and busiest day, and the dates the figures cover.

At most 50 places are measured a run, and places close together share one search. To start from a ready flow, add Mobility Footfall to a template such as Where to open my next cafe, Busy stations with no convenience store or Sampling day: the busiest spots and the weather. For who lives around a site, rather than who is there, see what street-level demographic data shows.