How footfall changes on public holidays, and how to measure it yourself

Public holidays move people around: districts that depend on offices often go quiet, while shopping streets, parks, family neighbourhoods and places of worship can get busier. How much depends on the holiday, the city and the street, so the only reliable answer for your sites is to measure them: keep a weekly record of footfall around each one, then compare the week that held the holiday with the ordinary weeks either side. There is no holiday button in the data, and with figures starting on 17 October 2026, the record you start now is what you will compare next year against.

This article is for store, restaurant and venue managers planning stock and staff for holiday periods, marketing teams timing campaigns, and analysts asked why one week looked strange. It covers why a holiday is hard to see in a single figure, a simple weekly method, an example around a festival in Delhi, the traps, and a flow that keeps the log for you.

Why one figure hides the holiday

Mobility Footfall averages a window of days: the last 7, 14, 30 or 90, ending about ten days ago because each day arrives late (the reason is in why footfall data arrives days later). A single holiday inside a 30-day window is one day in thirty; its effect is diluted almost to nothing. Inside a 7-day window it is one day in seven, which you can see.

The block also offers all days, weekdays only or weekends only. A weekday holiday behaves more like a weekend in many places, so it pulls a weekdays-only average down and leaves the weekend figure alone. That is a useful tell.

A weekly method that works

1. Every week, measure each site for the last 7 days, weekdays only, and separately for the weekend if your business trades then.
2. Keep each week's footfall_daily, footfall_from, footfall_to, busiest day and day split as a new row in a sheet.
3. When a holiday falls inside a week's dates, mark that row.
4. Compare the marked week with the average of the two weeks before and the two after, not with the single week next to it.
5. Look at the day split as well as the total: a holiday often moves people from morning to afternoon even when the daily total barely changes.

The block's own footfall_trend, the last 7 days against the 7 before, will also flag a holiday week, but only against one neighbouring week, which may itself have been unusual. Your own log is steadier.

An example around a festival in Delhi

A made-up case. A sweets and gifting chain in Delhi keeps the weekly log for four outlets: one in a market street, one in a residential colony, one beside a mall and one in an office area. When a major festival falls inside a week, its marked row reads like this against the average of the surrounding four weeks (invented for illustration):

Average daily footfall in people-hours, weekdays only. Invented figures for illustration; not real figures for Delhi.

The chain learns that its market and residential outlets need more stock and staff in festival week, the mall outlet a little more, and the office outlet less, perhaps with staff lent to the market street. Next year, with this year's row in the sheet, it can plan a month ahead rather than react.

The same method fits any holiday in any city: a long weekend in Toronto, a national day in Cape Town, the first days of November in Mexico City.

Traps to avoid

• Blaming the holiday for everything. Weather, school terms, a big match or roadworks in the same week can move footfall as much. Note them in the sheet too.
• Comparing with the week right before. Shopping before a holiday can lift that week, making the holiday look quieter than it was.
• Using 30-day windows. They smooth the holiday away.
• Forgetting which days are the weekend. The block counts weekends as the place's own: Friday and Saturday in most of the Gulf, Saturday and Sunday elsewhere.
• Expecting history before October 2026. The footfall record starts on 17 October 2026, so last year's holidays are not in it.
• Reading people-hours as shoppers. A family neighbourhood gets busier on a holiday partly because people stay home; they are present, not necessarily buying.

Let an AI employee keep the log

1. Start with Schedule, every Monday morning.
2. Bring in your sites with My data.
3. Add Mobility Footfall: the last 7 days, weekdays only, with Footfall, Busy hours, Busiest days and Trend, 200 m.
4. Add a Google Sheets step to append one row per site, with the dates covered (see Google Sheets, Docs and Drive).
5. Add xMap LLM asked to note any site whose footfall moved more than a set share on the week before, and which part of the day moved.
6. Send it by Email. In the inbox each Monday: a short note of the sites that moved, the dates the figures cover, and the sheet growing week by week.

Add a second Mobility Footfall block set to weekends only if your weekend trade matters; the two side by side show whether a holiday behaved like an extra weekend at your sites.