Staffing to Footfall: How to Match Rotas to When Customers Actually Arrive
Staffing to footfall means building your rotas around when customers actually arrive, not around habit or the shift pattern you inherited. Most stores schedule the same cover every Tuesday because that is how it has always been done. The customers, meanwhile, arrive on their own curve.
When the rota and the curve do not match, you pay for it twice. You pay for staff standing around in a quiet morning, and you pay again in lost service and lost sales during the afternoon rush.
This post gives you a practical, honest method to turn hourly visitor counts into a staffing plan. You will see how to read the real arrival curve, translate it into cover, flex on the day, and check your work each season. The numbers here are illustrative, so treat them as a worked example rather than a promise.
Key takeaways
- Rotas built on habit overstaff quiet hours and understaff rushes. Hourly footfall gives you the real curve to plan against.
- Pull counts by hour and day of week, with staff excluded, so the curve reflects customers and not colleagues walking the floor.
- Translate the curve into cover with a target customers-to-staff ratio, then round the result into realistic shift blocks.
- Use live occupancy and queue wait time to flex staffing on the day, on top of the planned rota.
- Re-check the curve each season and treat sale events and holidays as planned exceptions.
Why rotas drift away from real demand
Most rotas start with tradition. Someone set the shifts years ago, the team got used to them, and they carried forward through every reprint of the schedule. Gut feel fills the gaps. The result rarely matches when people actually walk through the door.
There are two failure modes, and they both cost money.
The first is overstaffing the quiet hours. You have four people on the floor at 10am on a Wednesday when two could cover comfortably. Those extra hours are pure labour cost with little sales to show for them.
The second is understaffing the rush. The after-work peak hits, queues build, fitting rooms back up, and customers leave without buying. That lost sale does not show on the wage bill, which is exactly why it is easy to ignore.
Footfall data replaces the guesswork with the real curve. When you can see how many customers arrive in each hour of each day, you can plan cover against demand instead of against memory. The counts come from your existing entrance cameras, they are anonymous, and the video stays on-site. No badges, no app, no facial recognition. You are measuring how many bodies crossed the line, by hour.
Get the real arrival curve from hourly counts
Start by pulling footfall by hour and by day of week. A single weekly total tells you almost nothing useful for scheduling. The pattern lives in the breakdown: Monday at 11am behaves nothing like Saturday at 3pm.
Before you trust the curve, exclude your own staff. If colleagues walking in and out of the entrance get counted as visitors, your morning numbers inflate exactly when the team arrives, and your curve bends in the wrong place. Staff exclusion keeps the count to customers, so the shape you plan against is real demand and not your own people clocking in.
Now look for the repeatable shape. Most stores have one. A typical day might run like this:
- An opening lull while the early shift sets up and few customers arrive.
- A midday build as the lunch crowd and errand-runners come through.
- An after-work peak as people stop in on their way home.
- A weekend that shifts everything earlier and flatter, with a longer plateau rather than two sharp spikes.
Your shape will be your own. The point is to find it.
Finally, average several weeks together. One rained-out Saturday or one local event can throw a single day off badly. Four to six weeks of the same weekday, averaged hour by hour, smooths out the noise. If you already produce a weekly footfall report, you have most of this data; you are just slicing it by hour instead of by day.
An illustrative method to translate the curve into cover
Once you have the curve, you need a rule that turns arrivals into people on the floor. The simplest workable rule is a target ratio of customers to staff that matches the service level you want.
Pick the ratio that fits your format. A high-touch fashion floor where staff fit and advise might aim for one colleague per 15 customers an hour. A self-service convenience format might run comfortably at one per 40. There is no universal number. Choose the one that keeps service where you want it.
Then work hour by hour. Divide the expected arrivals in each hour by your ratio to get the cover that hour needs.
Here is a clearly illustrative example for one weekday, using a target of one colleague per 20 customers per hour. These figures are made up to show the method, not a result you should expect.
| Hour | Expected arrivals | Raw cover (÷20) | Rounded cover |
|---|---|---|---|
| 09:00–10:00 | 18 | 0.9 | 1 |
| 10:00–11:00 | 32 | 1.6 | 2 |
| 11:00–12:00 | 55 | 2.8 | 3 |
| 12:00–13:00 | 70 | 3.5 | 4 |
| 13:00–14:00 | 48 | 2.4 | 2 |
| 16:00–17:00 | 60 | 3.0 | 3 |
| 17:00–18:00 | 80 | 4.0 | 4 |
| 18:00–19:00 | 38 | 1.9 | 2 |
The raw number is rarely a whole shift, so round to realistic blocks. People do not arrive in one-hour shifts. Group the hours into shift shapes your rota software and your team can live with, and round up where service matters most.
Leave headroom for the work that footfall does not capture. Breaks, deliveries, stock moves, and back-office tasks all pull someone off the floor. If a colleague is on a break during the 17:00 peak, they are not covering it. Build that into the plan rather than discovering it on the day.
Layering occupancy and queue on top of footfall
Footfall sets the plan. It cannot see the day you are actually having. For that, you layer live signals on top.
Real-time occupancy tells you how many people are in the store right now, not how many arrived this morning. When occupancy climbs past your planned level, that is your cue that the day is running hotter than the curve predicted and you may need to pull someone onto the floor.
Queue and wait-time data is the sharper trigger. A growing wait at the checkout is the clearest sign that you are under-covered at the front end, and it is the one customers feel most. Set a wait-time threshold, and when the queue crosses it, that is the signal to move a colleague to the till before the queue turns into walkouts. Our guide on measuring queue wait time and abandonment covers how to set sensible thresholds for your store.
To go deeper on combining these signals on the day, the practical guide to real-time occupancy and queue management pulls it together.
One last operational note: give in-the-moment flexing a single owner. The plan is a shared document, but the decision to move someone to the front end at 5:15pm needs one duty manager making the call. Shared responsibility for live flexing usually means nobody flexes.
Estimating the labour-cost payoff
The honest framing of the benefit is this: staffing to footfall does not magically cut your wage bill. It moves hours from where they are wasted to where they are needed, and trims the genuine waste.
Trimming overstaffed quiet hours frees budget you can spend on the real peaks. If you take one colleague off a slow Wednesday morning where the curve clearly does not need them, that hour can fund better cover during the Saturday afternoon rush, where it actually earns its keep.
To estimate your own payoff, use your numbers, not ours. For example, a store running one extra colleague for three quiet hours a day, at an hourly wage of, say, $16, is spending roughly $48 a day, or around $1,000 a month, on cover the curve does not support. Some of that is genuine saving; some is better redeployed to peaks. Run the same arithmetic on your own rota and wages.
Frame it honestly. The benefit is fewer wasted hours and better cover where it counts, not a fixed percentage saving. This is a planning aid built on real demand, not a promise of a specific number, and you should not present it as one.
Keeping the schedule honest over time
A curve is a snapshot of a season, not a permanent truth. Patterns shift, so re-check the curve every season. Summer trading, darker winter evenings, and back-to-school weeks all bend the shape. A rota tuned to spring footfall will be quietly wrong by autumn.
Compare planned cover against actual footfall to see where you missed. Lay the two side by side, hour by hour, after a few weeks of running the new plan. Where you planned three and the floor needed four, you have found an understaffed hour. Where you planned four and the curve was flat, you have found waste. This comparison is the whole feedback loop.
Feed sale events and holidays in as exceptions rather than letting them distort the baseline. A clearance weekend or a public holiday is not your normal Tuesday, and folding those days into the average corrupts the curve you plan ordinary weeks against. Handle them as named exceptions with their own cover. Our guide to occupancy management through peak season covers planning for the busy stretches in more depth, and if you run a counter-service format, the staffing-to-covers playbook for restaurants and QSRs applies the same idea to seated and quick-service demand. Stores chasing the underlying pattern first will want our piece on finding your real peak hours with footfall data.
Different formats need different ratios and rhythms. A high-touch fashion and apparel floor staffs very differently from a grab-and-go store, so tune the method to your own service model rather than copying someone else's.
Frequently asked questions
How do I use footfall data to plan staff schedules?
Pull your visitor counts by hour and day of week to find the real arrival pattern, then set a target ratio of customers to staff and work out the cover needed in each hour. Round the result into realistic shift blocks that your team and rota software can actually run.
Should I exclude staff from the counts I schedule against?
Yes. If employees are counted as visitors, your curve is distorted, usually right when the team arrives and leaves. CountPort can exclude staff so the footfall you plan against reflects customers only.
Does staffing to footfall really cut labour cost?
It can reduce wasted hours by trimming cover in quiet periods and moving it to real peaks. The exact saving depends on your wages and current schedule, so treat any figure as your own estimate rather than a guaranteed result.
How is footfall counted for scheduling?
CountPort counts visitors anonymously from your existing entrance cameras and records the totals by hour. All video is processed on-site, detection is body-only, and there is no facial recognition. Accuracy depends on camera placement, mounting height, lighting, and how crowded the entrance gets, so a good setup matters.
See it on your own traffic curve
The fastest way to judge whether your rotas match real demand is to look at your own hourly footfall next to your current rota. If they do not line up, you have found money on the table.
Request a demo and we will show you how CountPort turns your existing cameras into the hourly curve you can build a smarter schedule against.
