FlowsBook a callSign inCreate free accountSign up
Footfall and mobility

Footfall during festivals: measuring what a city celebration does to its streets

A festival fills some streets and empties others. How to set up a before, during and after comparison, read the evening shift, and report the effect honestly.

Part of the guide Location intelligence automation

By the xMap team · · 4 min read

Footfall during festivals is best measured as a comparison: the same streets in the festival weeks against the weeks just before and after, with a street far from the action as a check. Measured presence usually climbs around the festival sites and along the routes people walk to them, shifts later into the evening, and sometimes drops in ordinary shopping streets that people skip that week.

This guide is for festival organisers reporting to funders, city tourism and events offices, business associations deciding whether a festival helps their members, and shops and restaurants near the route. It sets out how to plan the measurement, how to read it, an invented example, what to avoid saying, and an AI employee that produces the report.

Plan the measurement before the festival

Because Mobility Footfall reports an average day over a recent window, and each day arrives about ten days later, the comparison is built from a series of weekly runs. Start them at least two weeks before the festival so you have a normal baseline, and keep them going for two weeks after.

  • Festival sites: the main stage, the market or lantern trail, each at 200 m so they read their own block of the map.
  • Routes: the streets between the station and the sites.
  • Neighbouring trade: the restaurant street and the main shopping street nearby.
  • A control street in another part of the city, untouched by the festival.
  • Weekends versus weekdays: many festivals peak on two weekends; set Days to weekends only on a second block to see them clearly.

Put the points in a table on the My data page. The data starts on 17 October 2026, so festivals before that date cannot be measured; Day of the Dead events in Mexico City at the start of November 2026 are among the first big celebrations inside the record.

Reading the effect

  1. Lift at each point: festival-week average day divided by the baseline weeks' average.
  2. Lift at the control: if the control rose too, part of the change belongs to the season or the weather, not the festival. Take it off.
  3. The evening shift: compare the evening and late-night shares of the day. Most festivals push presence later; that matters to restaurants and to night transport.
  4. Displacement: a shopping street that fell while the festival sites rose may be losing its regular customers for the week.

An invented example

All figures here are made up. A city in Spain runs a two-week autumn street festival. The weekly series covers eight points. Over the festival fortnight, the two sites run at roughly three times their baseline, the walking route at about double, the restaurant street up by around half with its late-night share clearly higher, the main shopping street slightly down on weekend afternoons, and the control street flat. The organiser's report to the city says the festival lifted the district, moved the busy hours into the evening and drew some weekend afternoon trade away from the shopping street, and suggests a festival trail through that street next year.

In Sao Paulo, the same flow can add Where visitors come from for the festival month: which districts the trips into each site set off from and how people travelled. Trip data is available only in Sao Paulo so far.

What to avoid saying

  • "The festival attracted N visitors." People-hours are presence, not people. Say how much busier the area was.
  • "Spending rose." Footfall is not sales. Pair it with traders' own figures if you want to talk about money.
  • "Up on last year." Not until a second edition falls inside the data.
  • Figures on the closing night. They arrive about ten days later; plan the report for the third week after.
  • Anything about individuals. All figures are aggregates; never use them to follow or single out anyone.

An AI employee that writes the festival report

  1. Schedule: every Monday from two weeks before to three weeks after the festival.
  2. My data: sites, routes, neighbouring streets and the control.
  3. Mobility Footfall: 200 m, last 7 days, with Footfall, Busy hours and Busiest days; a second block on weekends only.
  4. Traffic Flow: live road speeds and the jam factor around the sites at the time of each run, for the transport team.
  5. xMap LLM: write a paragraph per point, using footfall_trend for the change on the week before and the baseline weeks' figures, pasted in as a reference note once the first two runs are in.
  6. Make a report: a PDF for funders and the city, with the caveats on its first page.

Each Monday the organiser gets an updated PDF; the last one, three weeks after closing, is the festival's effect report. For the rest of festival planning, see Heat plan for an outdoor festival, Food vendors to invite to my festival and Residents near the venue, for a noise notice, the events and venues page, and the location intelligence automation guide.

Questions

Can footfall count festival visitors?

No. It measures people-hours of presence around each point. Report how much busier the area became, and use ticketing or counters for visitor numbers.

How far ahead should I start measuring?

At least two weeks before the festival, so the baseline is normal weeks, and keep going two weeks after it ends.

Why use a control street?

To take out what the whole city did that week. If the control rose too, part of the lift came from the season, weather or a holiday.

When will the figures for the last day be ready?

About ten days after it, since each day arrives late. Plan the final report for the third week after closing.

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.