The Hire You Can’t Interview – Why a Family Café Lives or Dies by Its ePOS System

I’ve got a habit that annoys my wife. We go somewhere for lunch, the kids climb the chairs, and I watch the till. Not the food. The till. Fifteen years of putting these systems into cafés and pubs does that to a person. And I can usually tell within ten minutes whether a place is going to have a bad Saturday, because the software gives it away long before the owner notices.

Here’s what I mean. Owners will spend a week fretting over a new barista. They’ll do two interviews, a trial shift, a chat with the last employer. Then they’ll pick the system that runs the whole building because the rep was friendly and it was £29 a month. Nobody trial-shifts the software. Nobody can, really, because the only trial that counts happens when the room is full, and by then you’ve already signed.

And full is the whole point. The real test of restaurant technology often comes during a packed service, with tableview.com entering the ePOS system for UK restaurants debate over how ordering, payments and floor activity should be coordinated under pressure. On a quiet Tuesday, my old Nokia could probably run a café. It’s the wet half-term Saturday that matters, the one where the delivery apps won’t stop, a grandad wants to pay for his bit separately, and the printer in the kitchen decides it’s had enough.

The job nobody writes down.

If you actually wrote the job description for the thing sitting by the till, you’d never hire a person for it. It takes orders from the counter, the tables, the phone, and two apps, sometimes in the same minute. It has to get each one to the cook in the right order, with the “no onions” intact. It should know there are three lasagnes left before the customer asks. It has to take a foreign card without a fuss, split a bill four ways while a queue watches, and then, at ten o’clock when you’re knackered, tell you what sold and what to buy tomorrow.

That’s a lot. I’ve met exactly one person who could hold all that in her head through a rush, and she retired in 2017.

Fake a Saturday before you sign

You can’t put an ePOS system for restaurant use through a real rush during a trial. But you can fake one, and I’ve done it in car parks with four phones and a very patient owner. Get five people ordering at once from different devices. Have one change their mind halfway through and add a nut allergy, then check the kitchen screen to see whether the change arrived in writing or vanished. One pays with a Dutch card. One asks to split by item rather than in half.

Then, and this is the bit reps hate, turn the router off.

• Some systems keep taking orders and quietly catch up when the wi-fi comes back.
• Some freeze completely, and you’ll find that out on a bank holiday instead of in a car park.
• A few sort of half-work, which is honestly worse, because you don’t notice the missing orders until the one-star review turns up

The screen in the kitchen

The single change I nag small owners about most is getting rid of paper tickets. I resisted this for years. Kitchen screens felt like chain-restaurant kit, all wrong for a place with twelve tables and a dog bowl by the door. I was wrong, and a café in Leamington proved it.

Their printer jammed at twenty to one on a May bank holiday. Six remakes in an hour, two families walked, and the owner told me later she’d lost the best part of £600 that day on a system she’d chosen because it was cheap. She put in a restaurant POS system with a kitchen display about three weeks later. The month after, remakes were down to four. They’d been running close to thirty. I don’t yet have a complete explanation for why the effect is bigger in tiny kitchens than in big ones. It just is.

What it changes for a family place is boring and useful.

• The kids’ meal flagged to go first goes first, because the cook can see it sitting at the top of the screen.
• Modifications arrive as text instead of as a memory

The money bit, without a spreadsheet

Owners feel the rush as stress. The accountant feels it as leaks that never show up as a line. Say you do 280-odd covers over a weekend. Eight remakes at cost is roughly a month of subscription gone. Four extra minutes per bill across forty tables is nearly three seatings you didn’t get. One missed online order adds up faster than anyone likes to admit. And the sourdough. I once stood next to a bin of unsold sourdough on a Sunday night that cost more than the owner was paying for the entire till that month, and nobody knew it was being over-ordered because nobody knew what had actually sold.

A decent ePOS system puts those numbers on your phone after the kids are asleep. Owners who’ve never seen the data are usually a bit shocked by which dishes they thought were popular and which ones lose money every time they leave the pass. But the reports are only worth anything if the orders were captured properly during the rush in the first place. So it comes back round to the Saturday again.

Small places get more out of this, not less.

There’s a belief that all this is for the big operators; in my experience, it’s backward. A 120-cover restaurant has a manager on the pass and enough bodies to soak up a chaotic hour. The family café has the owner, a sister-in-law, and a sixteen-year-old who’s good but sixteen. When the software does the coordinating, those three can run a service that used to need six, and somebody gets a Tuesday off for the first time since opening.

Parents tend to get this quicker than most, to be fair. You already plan for the six o’clock meltdown at four. Do the same with the café. Ask every rep what happens when the internet drops. Ask how many taps it takes to fix a wrong order while people watch. Ask whether the cook sees changes in writing.

If they can’t answer the router question, say thanks and move on. There’s another bank holiday in a few weeks.