A full-service restaurant measures its night by covers. Covers are the guests who sat down, ordered, and turned into revenue, and they are the number every operator watches. But covers are the end of a story, not the beginning. Before a cover there is a person at the door, reading the wait, glancing at the menu, deciding whether to stay. Some of them stay and become a cover. Some of them look at the queue, or wait two minutes without being greeted, and walk back out. Those people never appear anywhere in the restaurant's data. They are demand that arrived and left, and the booking sheet has no idea they existed.
That invisible walk-out is the gap between footfall and covers, and at peak it can be large. A restaurant that looks full on paper, every table turned, a healthy cover count, may also be turning away a steady trickle at the door that it never measures and never staffs for. This is a different question from how fast you turn a table. It starts at the entrance, and it is the one the booking sheet structurally cannot answer.
What is the difference between covers and footfall in a restaurant?
Covers count the guests who sat down and ordered. Footfall counts everyone who walked in, including the people who looked at the wait, read the menu, and left. The gap between the two is the demand a restaurant never served. Measuring both shows how many walk-ins are lost at peak, whether a queue or a slow greet is the cause, and how front-of-house staffing should track the door rather than the booking sheet. It is a different question from table turn, and the door is where it starts.
Note the distinction from quick service. A quick-service operation measures throughput, how many people it can move through a queue and a till in an hour. A full-service restaurant measures conversion of a sit-down decision, where the loss happens at the door before anyone reaches a table. Same word, footfall, very different funnel.
The restaurant-specific pain point: a full booking sheet hides the walk-ins you turned away at the door
The trap is that a busy night feels like a successful night. Tables are full, the kitchen is slammed, covers are up, and the manager goes home satisfied. What that picture hides is the door. On the busiest nights, when the restaurant is at capacity and the wait is longest, walk-in demand is also at its highest, and that is exactly when the most people give up and leave. The restaurant is losing the most customers on the nights it feels most successful, and nothing in its system records it.
Reservation data makes this worse, not better, because it tells a confident and incomplete story. The booking sheet shows the covers that were planned and the covers that arrived. It is silent on the walk-in who saw the "fully booked" sign and went next door, the couple who waited at an unattended host stand and gave up, the group that looked at a twenty-minute queue and decided against it. A restaurant relying on reservations to understand demand is reading half the room. The other half walked in and walked out, and the only way to see them is to count the door.
There is a second-order cost that compounds the first. A walk-in who is turned away or ignored does not just represent one lost table tonight; they form a judgment about whether the restaurant is worth trying again. A couple who stood unacknowledged at the host stand for two minutes and left will often not come back, and they will not tell the restaurant why, because as far as the restaurant's data is concerned they were never there. So the loss the door cannot see is not only tonight's covers but a slice of future demand, which is why operators who eventually measure the gap are often surprised by how large it is and how cleanly it maps onto their slowest-recovering nights.
Measuring the walk-in conversion gap
The measurement is the comparison of two numbers that already half-exist. Covers come from the POS or the reservation system. Footfall, the count of everyone who walked through the entrance, comes from a door counter. Put them side by side over the same hours and the gap is the walk-in demand that did not convert.
That gap is the metric worth managing. A small, steady gap is normal; not everyone who looks in intends to eat. A gap that balloons at peak is a signal: the restaurant is hitting a wall, and the wall is either capacity, the queue, or the greet. The same conversion logic that retail uses on entries applies directly here, with footfall as the top of the funnel and covers as the bottom. Tracked over weeks, the conversion gap also shows whether a new menu, a refit, or a marketing push pulled more people through the door, separate from whether it filled more tables, and it sharpens what "footfall" actually counts so you are not misreading door swings as unique diners.
Reading the gap by daypart and by night of the week is where the diagnosis sharpens. A gap that is flat all week but spikes only on Friday and Saturday evenings points squarely at a capacity or queue limit on the busiest service, which is a different problem from a gap that appears every weekday lunch, where the likelier cause is a host stand left unattended during a predictable rush. The shape of the gap, not just its size, tells the operator which lever to pull, which is why the comparison is worth doing continuously rather than as a one-off audit. A single week's snapshot can mislead; a multi-week pattern read against the staffing rota does not.
Front-of-house staffing to the door, not the reservations
Most front-of-house staffing is built around the reservation book and a fixed idea of service. The problem is that the reservation book does not predict the door. Walk-in pressure, especially around the start of a service or a known busy window, can arrive in a wave that the booking timeline smooths over. If the host stand is unattended for the two minutes that wave hits, the restaurant loses covers it had the capacity to serve.
Footfall data fixes the input. Staffing the door, the greet, and the early-service pinch points to the measured arrival curve, rather than to the reservation count, keeps a body at the host stand exactly when walk-ins surge. This is the same staff-to-traffic discipline any high-traffic venue uses, and it is especially sharp at predictable crunch points like the lunch rush, where a slow greet for fifteen minutes can quietly cost a dozen covers.
The mismatch is usually structural rather than a one-off lapse. A reservation book that shows tables booked from seven o'clock encourages a front-of-house plan built around seating those bookings, which pulls the available bodies onto the floor at exactly the moment the walk-in wave hits the door. The measured arrival curve almost always shows that wave starting earlier and running heavier than the booking timeline implies, because walk-ins do not book and so leave no trace until they appear. Once the manager can see the real arrival shape, the fix is often as simple as protecting the host stand for a defined window at the start of service, which costs little and recovers covers that were being lost to nothing more than an unattended door.
Counting at the entrance without a camera over the dining room
The obvious objection to counting in a restaurant is that nobody wants to dine under a camera. Guests are eating, talking, relaxing; a lens over the dining room would be intrusive and would change the feel of the place. The counting has to happen at the door and nowhere else, and it has to record numbers, not diners.
Ariadne measures this with Hybrid Fusion, its patented camera-free method. Time-of-Flight depth sensing counts every visitor at the entrances, capturing geometry rather than images, while patented phone signal sensing follows movement through the interior, detecting the signals a phone emits even in airplane mode, and tracks that movement to about one-metre precision. The sensor streams both feeds to Ariadne, where Hybrid Fusion combines them into one trajectory per visit and computes counts, dwell, and paths. The streams carry no identifier: no MAC address, no device ID, no biometric data, and no camera is involved. Identifiers are stored only when a visitor explicitly opts in, which keeps the method GDPR-friendly and outside biometric territory.
For a restaurant that means a small unit at the entrance counting who comes in and goes out, with no footage of the room and no record of any diner. The dining experience is untouched, and the camera-free method sidesteps the privacy conversation entirely, which matters in a setting built on guests feeling comfortable.
From counts to action: greet speed, queue handling, daypart staffing
The point of measuring the gap is to close it. Three levers move it most. Greet speed is the first and cheapest: if the conversion gap spikes when the host stand goes unattended, the fix is simply covering the door during the arrival wave, which the footfall curve now pinpoints. Queue handling is the second: if walk-ins leave at a visible queue length, the data tells you where the tolerance breaks, which informs whether to take a number, offer the bar, or quote an honest wait that keeps people rather than losing them.
Daypart staffing is the third and the most structural. The arrival curve, not the reservation count, should set how many people are on the floor and the door through the day. A daypart with high footfall and low conversion is under-served and over-losing; a daypart with low footfall is over-staffed. Aligning the rota to the measured door, the same way the people-counting-to-action playbook describes, turns the conversion gap from an invisible loss into a managed number.
How a restaurant manager actually uses it
The user is the general manager or the front-of-house lead, and the practical rhythm is a morning review and a weekly pattern check. In the morning, the manager looks at the previous night's footfall against covers by hour. A clean night shows footfall and covers tracking together. A bad night shows them diverging at a specific window, and that window is the conversation: was the host stand covered at 19:30 when the walk-in wave hit, or was everyone seating a large booking. That single read turns "it felt hectic last night" into a specific, fixable staffing note for the next equivalent service.
Over a week, the same manager uses the pattern to reshape the rota rather than react to last night. If the gap reliably opens at the start of Friday and Saturday service, the fix is a dedicated host on the door for that window, costed against the covers the gap shows are being lost. The discipline that makes this work is comparing like with like: a wet Tuesday and a dry Saturday have different footfall for reasons that have nothing to do with service, so the manager reads the conversion gap, the share of arrivals that became covers, rather than the raw door count, because the gap is the part front-of-house can actually move.
FAQ
Is footfall the same as covers in a restaurant?
No. Covers are the guests who sat down and ordered. Footfall is everyone who walked in, including those who left without sitting. The gap between them is the walk-in demand the restaurant did not serve.
How do I measure the walk-ins I lose?
Count everyone who enters with a door counter and compare it to covers from your POS or reservation system over the same hours. The difference, especially how it grows at peak, is the walk-in conversion gap.
Does this work for a restaurant that is mostly reservations?
Yes, and it is often most useful there, because a reservation-led restaurant has the least visibility into walk-ins. The booking sheet shows planned covers; it cannot show the people who arrived hoping for a table and left.
Will I need a camera in the dining room?
No. Counting happens at the entrance using depth and signal sensing, not cameras, so there is no footage of the dining room and no record of individual guests. The dining experience is unaffected.
How do I tell whether the queue or the greet is losing me walk-ins?
Read the conversion gap by window and against your floor state. If the gap spikes when the restaurant is genuinely at capacity, the limit is seats; if it spikes when there were tables free but the host stand was unattended, the limit is the greet. The two have different fixes, and the timing of the gap is what separates them.
Can a single counter handle a restaurant with a separate bar entrance?
Counting needs a sensor at each entrance people actually use, so a separate bar door gets its own unit and the counts combine. That also lets you see the bar trade and the dining trade as distinct streams, which matters if the bar converts walk-ins the dining room turned away.
---
