How to measure footfall after renovation of a shopping centre
A refit is finished and the owners want to know if it worked. How to read people-hours around the centre before, during and after the works, without fooling yourself.
Part of the guide Location intelligence automation
To measure footfall after renovation of a shopping centre, record the people-hours of presence around it in three phases (before the works, while hoardings are up, and after the doors reopen), on the same radius and the same kind of days, and judge the result against a similar centre or street that was not refitted. Read the reopening weeks apart from the settled weeks that follow, because a relaunch pulls a crowd that rarely lasts.
This is for asset managers, landlords and centre marketing teams who have spent money on a new entrance, a food hall or a facelift and now face the question from investors: did more people come? It covers what the figure can prove, how to set up the three phases, a made-up worked example, the traps, and an AI employee in xMap Flows that keeps the record going without anyone remembering to run it.
What a refit can and cannot show in the figures
Footfall in xMap Flows is people-hours of presence in the area around a place, scaled from a panel of devices to the whole population. A shopper who stays two hours counts twice. That makes it a good measure of how busy and how sticky a destination is, which is exactly what most refits aim at: a better food court is meant to keep people longer, a new car park entrance to bring them in from a different side.
It does not count sales, tenants' takings or unique shoppers. A refit that raises people-hours by a tenth while the till receipts stay flat tells you people linger but do not buy, which is useful in its own right. Keep your own door counters and tenant turnover beside the footfall line; the two answer different questions.
- Total presence. The average day's footfall around the centre, the main headline for the board.
- The shape of the day. How the day splits between morning, midday, evening and late night. A new restaurant terrace should push the evening share up.
- Busiest and quietest days. A refit aimed at weekday office workers should lift Tuesday, not Saturday.
- Week on week. The last 7 days against the 7 before, to see whether the reopening bump is fading or holding.
Setting up the three phases
The data has two limits you plan around. Each day arrives about 10 days after it happens, so you only see the first reopened week in the third week. And the footfall history only begins in mid-October 2026, so a refit that finished before then has no before phase at all, and you cannot compare with the same season last year. If your works are still ahead or under way, start recording now.
- Pick the circle and keep it. For a mid-size centre, 300 to 500 metres around its middle covers the building and its car parks. Never change it between phases.
- Before. At least two full weeks while the centre trades normally. Note anything unusual: a public holiday, a big match, a street festival.
- During. Keep measuring while the works run. Hoardings, a closed entrance or a moved bus stop usually push presence down, and that dip is part of the story the investors want.
- Reopening. Mark the first two weeks after the launch as their own phase. Events and offers draw people who will not come back weekly.
- Settled. Weeks three onward are the fair after phase. Compare them with the before phase, weekdays with weekdays and weekends with weekends.
- Control. Measure a similar centre or high street in the same city, untouched by works, exactly the same way all through. Your result is the change at your centre minus the change at the control.
A worked example (made up)
The centre, the figures and the dates here are invented to show the arithmetic. A neighbourhood centre in Lyon closes one wing for six weeks to build a food hall. The owner records footfall within 400 metres of the centre and of a comparable retail street across the river.
| Phase | Centre, people-hours a day | Control street | Evening share at the centre |
|---|---|---|---|
| Before (2 weeks) | 18,000 | 12,000 | 22% |
| During works (6 weeks) | 15,300 | 12,100 | 19% |
| Reopening (2 weeks) | 23,400 | 12,300 | 30% |
| Settled (4 weeks) | 20,700 | 12,200 | 28% |
Invented figures for illustration only.
The settled phase is 15% above the before phase at the centre and under 2% above at the control street, so roughly 13 points of the rise can be put down to the refit. The evening share moving from 22% to 28% says the food hall did what it was built for. The reopening figure of 23,400 would have overstated the effect by more than half; it belongs in the launch report, not the business case.
Mistakes that make a refit look better or worse than it was
- Comparing the launch weekend with an ordinary week. Always wait for the settled phase.
- Changing the radius. A wider circle after the works picks up a busy road or a station and invents growth.
- No control area. School holidays, weather and pay day move the whole city. Without a control you cannot tell the refit from the calendar.
- Reading people-hours as shoppers. Longer stays raise the figure with no extra people. Say so in the report.
- Forgetting the delay. A figure read on the day the works end still describes the works.
Letting an AI employee keep the record
The hardest part of a before-and-after study is discipline: the same measurement, every week, for months. An AI employee in xMap Flows does that for you. Part of the wider work of location intelligence automation, it runs on a timetable and writes the phases down as it goes.
- Schedule: every Monday morning.
- My data: a small table with the centre and the control street, each with a position and a phase column you update when the works start and end.
- Mobility Footfall: footfall, busy hours, busiest days and trend, 400 metres, the last 7 days.
- xMap LLM, on the whole list: write three sentences comparing the centre with the control this week, naming the phase.
- Email to the asset manager, or Make a report once a month for the investors.
What lands in the inbox each Monday is the week's people-hours for both places, the evening share, the busiest day and a short note such as "Centre up 4% on last week while the control is flat; settled phase, week 5". After a few months you have the whole curve, with the dip during the works and the level it settled at, ready for the board. Every run's figures are also saved, so you can download the full series from the run history.
Questions
How long after reopening should I wait before judging a renovation?
Leave the first two weeks out as a launch phase, then judge on at least four settled weeks. Add about 10 days for the data to arrive, so the earliest fair read is around six to seven weeks after the doors open.
Can I compare with the same month last year?
Not yet. The footfall history in xMap Flows starts in mid-October 2026, so the fair comparison is with a before phase you recorded and with an untouched control area measured at the same time.
Does footfall tell me how many people visited the centre?
No. It is people-hours of presence in the area around it, so a visitor who stays longer counts more. Use it to compare phases and places, and your door counters for visitor numbers.
What if the works also changed the roads around the centre?
Note the date in your phase column and consider a second, smaller circle on the side that changed. A new junction or a moved bus stop can shift presence from one entrance to another without changing the total.