Black Friday and Christmas: plan the peak on footfall data
To plan Black Friday and Christmas on footfall data, measure arrivals per hour through the peak weeks, with dwell, capture rate and zone flows, against a baseline of normal weeks.
On this page
Most retailers know the shape of the year: a long build, a surge in November, a December sprint and a quiet January. Many do not know the shape of their own peak, because the main instrument reading it is the till, and the till only sees the people who bought. So the peak weeks get planned on last year’s revenue, a staffing template and a guess that the queue will form at the same hour as before.
How has the Black Friday peak changed shape, and why does revenue data hide it?
Industry reports suggest the concentrated Black Friday spike has been spreading into a longer Black Week, with purchases distributed across many days, although Black Friday itself remains one of the most intensive days of the year in visits. Swedish trade bodies have pointed to e-commerce growing across the week while the single Friday softens, and shopping centre operators have seen visitor flows on the day itself decline even in years when the period as a whole held up (HUI Research reported centre footfall down around three percent on Black Friday in 2018, with somewhat larger drops in the big cities). For a store, that means staffing for the Friday alone no longer fits a demand curve that now stretches across more than a week.
Revenue can look fine while the pattern underneath it shifts, which is why revenue data hides the change. Visitor data shows that pattern: arrivals per hour across the whole week, dwell as the floor fills, and the capture rate when the street outside is at its busiest.
What should you measure in the peak weeks?
In the peak weeks, measure three things: arrival curves per hour, capture rate, and zone flows with dwell.
- Arrival curves per hour: planning staff by the hour instead of by the day affects whether customers are served or give up when volume peaks (see staff planning with footfall data).
- Capture rate: November streets carry some of the year’s densest passing traffic, so a window that draws in even a slightly larger share of passers-by in the peak weeks is worth more than it would be in most other months.
- Zone flows and dwell: campaign tables and seasonal displays can pull traffic deeper into the store or cause congestion at the door. The event impact measurement that reads campaign lift shows which happened.
Why does the baseline become more useful every year?
Much of the value of measuring the peak comes from the comparison, so each year of data makes the baseline more useful. A continuous series, measured with the same method, replaces “it felt busier than last year” with a number per store and per hour, year on year. The January review can then start from figures: which hours were understaffed, which days deserve the campaign budget, and which stores converted the traffic they were given and which did not.
With measurement running well before the next peak season, you have a clean pre-peak baseline. Wi-Fi based counting on existing access points makes that timeline realistic. The retail metrics glossary defines the measures this guide relies on.
Frequently asked questions
Why measure footfall during Black Friday if the tills already show sales?
Because tills record only the visits that ended in a sale. Footfall adds how many came, when they arrived, how long they stayed and where they went. A weak day can then be traced to weak traffic or weak conversion, and you can fix the right problem before Christmas week repeats it.
Is Black Friday still one day of store traffic?
Black Friday remains one of the most intensive days of the year in visits, but industry reporting shows buying spreading across the whole Black Week, so the peak is no longer only one day. Hour-by-hour visitor data shows how your own customers spread across that week, and staffing and campaign timing should follow their pattern.
What should we compare the peak against?
Against your own baseline. A continuous visitor series shows the peak as a lift over normal weeks, store by store and hour by hour, and because the metric and the method stay the same, the year-on-year comparison stays like for like.
Can we get measurement running before the next peak season?
Usually. Wi-Fi based counting typically uses the access points a store already has, so deployment before the next peak season is realistic, and each extra week of data before the peak strengthens the baseline.