FlowsBook a callSign inCreate free accountSign up
Population and demographics

Daytime vs nighttime population: mapping where a city fills up and empties out

Residents show where people sleep; footfall shows where they spend the day. How to rank parts of a city on both and find the places that swell by day or by night.

Part of the guide Location intelligence automation

By the xMap team · · 4 min read

Daytime vs nighttime population is the gap between where people sleep and where they spend their waking hours. The night side is close to the resident count: a population grid tells you how many people live in each small area. The day side is measured differently, as footfall, the people-hours of presence recorded around a point. Because the two are in different units, the honest way to compare them is by rank: put every part of the city in order on residents, then in order on daytime footfall, and look at the places whose two ranks are far apart.

City services use this to place public toilets, waste bins and street cleaning; police and security plan patrols; retailers decide whether a branch should open early or late; emergency planners ask how many people would be in an area at 2 pm on a Tuesday. This article sets out a grid method that works across a whole city rather than one site at a time.

Why ranks, not a ratio

It is tempting to divide footfall by residents and call it a daytime multiplier. Do not. Footfall counts hours, not heads: one person who stays eight hours near a point counts eight times. Residents count heads, with no hours at all. A ratio of the two has no meaning you can explain to anyone.

Ranks avoid the problem. If a block is 5th of 50 for residents and 48th for midday footfall, it is a place people live and leave during the day. If it is 47th for residents and 3rd for midday footfall, it fills up while its few residents are out. The size of the gap between the ranks is the swing.

Laying a grid over the city

  1. Make a table of points about 1 km apart across the part of the city you care about, each with a short name such as "C4". Up to 50 points fit in one run of footfall; a larger city is done in batches.
  2. Read the table with My data.
  3. Demographics, each incoming place, within 500 m: residents around each point.
  4. Mobility Footfall, each incoming place, within 200 m (the point's own cell), on weekdays, with Footfall and Busy hours: the average day's footfall, the busiest hour and how the day splits between morning, midday, evening and late night.
  5. Data analyst, asked to rank the points on residents and on midday footfall and give each point the difference between its two ranks.
  6. Show on a map, coloured by that difference, and a table by email.

Use the same distances for every point. Mixing a 500 m count here with a 1 km count there makes the ranks meaningless.

Reading the swing

PatternResidents rankMidday footfall rankWhat it usually is
Fills by dayLowHighOffices, markets, campuses, hospitals
Empties by dayHighLowResidential suburbs, commuter districts
Busy bothHighHighMixed centres with homes above shops
Quiet bothLowLowIndustrial land, parks, the edge of the city
Fills lateLowLow at midday, high late nightNightlife and transport hubs

How to read the gap between the two ranks. The last row needs the late night share from Busy hours, not midday.

A made-up example

Say a city services team in Mumbai lays 50 points over the island city and its nearest suburbs. The made-up result: the points around the business district rank in the bottom ten for residents and the top ten for midday footfall, so public toilets and lunchtime street cleaning go there; a band of residential points to the north ranks high for residents and low at midday, with an evening peak, so waste collection there moves to early morning before people leave; two points near a station rank low on both at midday and high late at night, which goes to the night patrol plan.

Limits to keep in view

  • Footfall covers most of the world from 17 October 2026, not yet Australia. Each day arrives about 10 days later, so the latest figures are always about ten days old, and the history is short.
  • Footfall is scaled from a panel of devices to the whole population. It is a measure of how busy a spot is, good for comparing spots, not a head count.
  • The population grid covers about 100 countries, among them India, Thailand, Egypt, France, Canada and Australia, but not the United Kingdom or the United States. Both pieces of data exist together only where both cover the city.
  • Weekdays and weekends swing differently. Run the footfall step twice, once for each, before planning anything that runs seven days a week.

Once the grid is set up, a monthly Schedule keeps the map current as the footfall history grows. For other ways to combine residents, places and movement, see location intelligence automation.

Questions

Can I add footfall and residents to get a total daytime population?

No. Footfall is people-hours of presence and residents are people, so the numbers cannot be added or subtracted. Compare them by rank instead.

How far apart should the grid points be?

About 1 km works for a city-wide view. Closer points give more detail but use more of the 50 places a footfall run takes.

Does it work in Sydney or Melbourne?

Not fully yet. Residents are covered in Australia, but footfall data does not cover Australia so far.

How recent are the daytime figures?

Each day's footfall arrives about 10 days later, and the data starts on 17 October 2026, so a window of the last 7 days covers the week ending about ten days ago.

Templates that do this

Read next

More on population and demographics · 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.