FlowsBook a callSign inCreate free accountSign up
Footfall and mobility

Comparing footfall in two cities without fooling yourself

Two cities run on different clocks, weekends and scales. How to set footfall from Madrid and Riyadh side by side so the comparison actually means something.

Part of the guide Location intelligence automation

By the xMap team · · 5 min read

To compare footfall in two cities, do not set one street's people-hours directly against another's. Measure both the same way, then read each against a reference spot in its own city, such as the main station or the busiest shopping street, and compare the shapes: the share of the day in each part, the busiest hour in local time, weekdays against the local weekend. Absolute levels across cities are the least reliable part of the data; how a place behaves compared with its own city is the most useful.

This is for retail and restaurant brands opening in a second country, consultants writing a market-entry note, and landlords marketing one location to tenants from another city. It explains why two cities are harder to compare than two streets, a method that holds up, a made-up worked example in Madrid and Riyadh, the traps, and a flow that produces the comparison.

Why two cities are harder than two streets

Mobility Footfall measures people-hours of presence around a place: someone who stays three hours counts three times. It is scaled from a panel of devices to the whole population, and each day arrives about 10 days after it happened. Within one city those facts affect every site alike, so a comparison is fair. Across cities, four things can differ.

  • The clock. Each place's hours are given in its own local time, and the field footfall_timezone says which. An evening peak at 21:00 in one city and 18:00 in another is a real difference in habits, not an error.
  • The weekend. The Days setting follows the place: Saturday and Sunday in most countries, Friday and Saturday in much of the Gulf. Compare weekend with weekend, not Saturday with Saturday.
  • The scale. How well the device panel represents each country can differ, so a street with twice the people-hours of another in a different country is not reliably twice as busy.
  • The history. Footfall starts on 17 October 2026. Windows longer than the days available are filled with what exists, and the step says so; use the same window in both cities.

A method that holds up

  1. Pick the candidate site in each city and one reference spot in each: a place every local would agree is busy, such as the main shopping street or the central station.
  2. Measure all four with the same radius, the same time window and the same Days setting.
  3. Divide each site's footfall per day by its own city's reference. A site at 0.4 of its reference is a fairly busy spot by local standards wherever it is.
  4. Compare the shape of the day: the busiest hour, the share of footfall in the morning, midday, evening and late night, and the busiest and quietest days.
  5. Only then look at the raw levels, and treat a gap between countries as a hint, not a fact.

A worked example in Madrid and Riyadh

The figures in this example are made up, to show the method. A coffee brand with shops in Madrid is choosing a first site in Riyadh. It compares its best Madrid shop with the candidate, each against a reference spot in its own city, over the same 14 days, all days, within 250 m.

Madrid shopMadrid referenceRiyadh candidateRiyadh reference
Footfall per day (people-hours)18,00060,00011,50052,000
Share of its city's reference0.301.000.221.00
Busiest hour, local time13:0018:0021:0021:00
Share in the evening and late night38%45%61%58%
Busiest daySaturdaySaturdayThursdayFriday

Made-up figures to show how a two-city comparison reads.

On raw people-hours the Riyadh candidate looks much weaker. Against its own city it is a little less busy than the Madrid shop is in Madrid, 0.22 of the reference against 0.30, which is a gap worth knowing but not a reason to walk away. The bigger lesson is the clock: the Madrid shop lives on lunch; the Riyadh spot lives after dark, with three-fifths of its footfall in the evening and late night and the week peaking on Thursday. The brand keeps the site on the list and plans opening hours, staffing and the menu around the evening, not around a Madrid lunch rush.

Mistakes to avoid

  • Reading people-hours as visitors. A spot where people linger, such as a mall with cinemas, scores high on people-hours without more visitors.
  • Comparing different windows. A week in one city and a month in another mixes holidays and paydays. Keep the windows identical.
  • Mixing weekends. Use the Days setting and let it follow the place.
  • Expecting sales. Footfall measures presence near a place, not who comes in or buys.
  • Forgetting where visitors come from. Trips into a place, with the districts they start in, are measured in Sao Paulo only so far. Do not plan a two-city comparison on them.

Producing the comparison with an AI employee

One flow measures every site and reference across both cities and writes the comparison.

  1. My data: a table of the sites and the reference spots, with a column saying which city and which role each has.
  2. Mobility Footfall: within 250 m, the last 14 days, all days.
  3. xMap LLM, on the whole list: divide each site by its own city's reference, compare the shape of the day and the week, and write six lines using only these figures.
  4. Make a report for the expansion committee, or Email it.

What lands in the inbox is a table and a short note: each site's level against its city, the busiest hour in local time, the part of the day that carries it, and the busiest day. Run the flow again a month later and compare the notes, now with more days behind each figure. Templates to start from: the five best spots for a pop-up shop, where to open next and market entry study, refreshed monthly.

Questions

Can I compare footfall between countries at all?

Yes, as long as you compare each site with its own city first. Raw people-hours across countries are a hint, not a measure.

Why is the busiest hour so different between cities?

Hours are given in local time and habits differ: lunch-led cities peak at midday, hot cities often peak late in the evening. That difference is usually the most useful finding.

How many days of footfall should I use?

The same number in both cities, and as many as exist. Footfall starts on 17 October 2026 and arrives about 10 days late, so recent windows are short.

Does footfall cover every city?

It covers most of the world from 17 October 2026, though not yet Australia. A place with no data yet says so, so run your sites once before planning around them.

Templates that do this

Read next

More on footfall and mobility · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.