The desk is the most expensive thing at the venue
A two-court padel venue turns over somewhere between £8,000 and £15,000 a month. A person on the desk every evening costs about £2,200 a month at the 2026 living wage with employer costs; through the whole day it is nearer £6,600. Nothing else on the site, not the lights, not the booking platform, not the insurance, comes close. Most of what that person does is open the gate for people who have already paid, and close it after people who have already left.
So the case for an unstaffed venue writes itself, and the padel industry knows it. What it has been vaguer about is the one thing that makes it work: when a player stands at a locked gate with no staff in sight, how does the venue know they belong there?
The obvious answer, and the doubt
The obvious answer is the booking. Every booking has a time, a court and a person attached. If the gate knows the bookings, it can let the right person in at the right time and nobody else. That is the whole idea, and any decent system can do the letting-in part.
The doubt is whether the booking really does identify a person. Booking platforms are built to sell court time, not to run doors. Some bookings are made by staff on a customer's behalf. Some accounts are created in a hurry on a phone. If a meaningful share of bookings cannot be tied to the person who turns up, those players arrive at a locked gate with a valid booking and no way through, and the unstaffed venue has failed at the exact moment it was supposed to save money.
Nobody we asked could put a number on that share. So we did.
What we looked at
We took a month of live bookings from two venues on the same booking platform, both with two courts, in different parts of the country and with different clienteles. Around 650 active bookings between them, once cancellations were removed. For each booking we asked one question: does this booking carry enough about the person who made it for a gate to recognise them when they arrive? Nothing personal was read or kept; the count ran on the venues' own equipment.
Here is what came back.
| Bookings | Booker identifiable | Not identifiable | |
|---|---|---|---|
| Venue A | roughly 300 | all of them | 0 |
| Venue B | roughly 350 | 342 | 9 |
| Together | about 650 | 98.6% | 1.4% |
Nearly every booking could open a gate. About one in seventy could not. That sounds like a footnote until you picture the seventieth player, in the rain, at 9pm, with three friends and no staff.
Who the seventieth player is
The nine bookings were not random. They came in two shapes, and the shapes matter more than the count.
The coach. Six of the nine were private classes, entered by staff at the desk in the manager app. Those bookings exist to block the court, not to sell it, so they carry a placeholder rather than a customer. The person who turns up is the coach, sometimes with a bag of balls and a group behind them. A venue that runs classes will see this every single week, and the more coaching it does the more of these it will have.
The quick sign-up. Three were ordinary bookings from the mobile app by a single player whose account was created with the bare minimum. They are real customers who paid real money, and there is not enough on the booking for a door to recognise them.
What did not appear is just as useful. Not one booking made on the web or the iOS app fell into either group. Not one group booking did. The problem is narrow and it is predictable.
The guests are not the problem
Something else in the data looked alarming and wasn't. Roughly one participant in nine was a guest added to a booking by name only. If the gate needed to recognise every player individually, that would be a fifth of the bookings with a gap.
It doesn't. The booker is the person who arrives first and lets the group in, and the booker was identifiable on every booking except the nine above. Group bookings are a solved problem as long as the door opens for the person who paid.
What this means for the design of an unstaffed venue
The numbers change what "unstaffed" has to mean. It cannot mean a gate that opens for bookings and nothing else, because that leaves a coach and the odd customer locked out every week. It has to mean three things working together.
The booking opens the gate, and does the heavy lifting. For 98.6% of arrivals, the person who booked proves it at the gate and it opens for the window around their booking. Fifteen minutes before, ten minutes after, is about right: early enough that nobody waits in the car park, tight enough that whatever let them in is useless an hour later. Entry should belong to that booking and that window and to nothing else.
Staff and coaches need their own way in. The cleanest answer to the coaching problem is a separate way in for the team that the venue controls from its own dashboard, changes routinely, and can withdraw in one tap if it leaks. A venue that can do that in the morning does not lie awake wondering who has it.
A person is still reachable, just not on site. The quick sign-ups are one in two hundred. A phone number on the gate and a manager who can let someone in remotely covers them. What the venue should never do is design for this case as if it were common. It isn't.
The night at the gate
Put those together and an evening at an unstaffed venue looks like this. At 6:45 a group arrives for a 7pm booking. The booker checks in from their phone at the gate and it opens. The lights on their court come on at 6:59. At 8:02 the lights go off. A coach arrives at 8:10 for an 8:15 class and lets herself in the way the team does. At 9pm someone who signed up in a hurry rings the number on the gate, and the manager, at home, lets them in from the dashboard.
Everything that happened is in a log the venue can read in the morning: every open, when, on which booking, and every failed attempt.
The objections we hear
What if the internet goes down? Ask the vendor exactly this, and expect a straight answer. A gate that needs the cloud to decide whether to open will, at some point, leave people in the car park. The venue's own equipment should be able to make that decision on its own with the day's bookings it already has.
What if someone lets a friend in? They have let in someone who is now on the court with them, for the length of their booking. That is what they paid for.
What about cancellations and no-shows? A cancelled booking should stop opening the gate the moment the cancellation reaches the venue. A no-show costs nothing; the lights never came on.
What about people who follow someone through? A gate is a gate. This is no worse than a staffed venue where the receptionist is on the phone, and the log at least tells you when it happened.
What to do before you take the desk away
- Run this count on your own bookings for a month. It takes a day of data and an afternoon, and it tells you exactly how many people your gate will fail.
- Decide how coaches get in before anything else. If you run classes, this is not optional.
- Put a phone number on the gate for the one in two hundred, and a log behind the gate for the morning after.
- Make sure the gate still opens for bookings with the broadband down.
- Only then cancel the evening shift.
An unstaffed venue is not a venue with no staff. It is a venue where the bookings do the work staff used to do, and the staff are somewhere more useful.
