Restaurants rarely close because of one bad decision. They close because of accumulation. The Restaurants Association of Ireland's April 2024 report on closures and RTÉ Prime Time's reporting both describe the same pattern: busy rooms, thin margins, and owners who found out too late where the money went.
At Resk (formerly Elyx) we sort the damage into two piles. Macro costs, meaning rent, energy, wages and VAT, which no software fixes and we won't pretend ours does. And leaks, which are yours to close. There are six that come up again and again. Here's what each one looks like in a real week, how to spot it in your own numbers, and what fixed looks like.
1. Missed demand
The scene: 7:40pm on a Friday. Full room, two staff down, and the phone rings. It rings out. The guest doesn't leave a voicemail, because nobody under sixty leaves voicemails. They book somewhere else from the same phone before your next table is cleared. You never find out it happened, which is the worst part. Missed demand doesn't appear on any report. It's invisible revenue.
How to spot it: pull your phone provider's call log for the last month and count unanswered calls during service hours. Most owners have never done this. The number is usually a shock, and even a blunt assumption about how many of those were booking attempts turns it into money. Add the enquiries that came through Instagram or the website contact form and got answered a day later.
What fixed looks like: every call answered, including at 7:40 on a Friday, with the intent captured and the booking made under your rules. A booking call landed during the rush, and it became a booking instead of a story you never heard. Your morning brief tells you how many calls came in after close and what each guest wanted.
2. Unpaid revenue
The scene: Thursday, before the weekend push. Two deposits for Saturday's big tables were requested but never paid. A payment link from last week's private hire failed and nobody saw the failure notice. Money is waiting in payment links, and it will keep waiting, because chasing it is nobody's actual job. Then Saturday's eight-top no-shows and the unpaid deposit means you carry the whole cost of the empty table.
How to spot it: list every payment link and deposit request from the last 60 days and mark which ones were actually paid. Then check your no-show count for the same period against which of those tables had a paid deposit attached. The gap between requested and collected is the leak, and it's usually concentrated on exactly the bookings that hurt most when they vanish.
What fixed looks like: unpaid deposits and failed links get chased automatically, and the manager gets a clean list before service instead of a nasty surprise after it. Deposits protect covers before guests arrive. No-shows still happen. They just stop being free for the guest and fatal for the night.
3. Weak margins
The scene: Monday morning, an invoice lands. A supplier price change is eating margin on the salmon, which touches four dishes. Nobody connects the invoice to the menu because the invoice lives in a folder and the menu lives in your head. Three months later the accountant asks why food cost crept up two points, and by then you've sold a thousand plates at the wrong price.
How to spot it: pick your five best-selling dishes and cost them today, at current invoice prices, not the prices from when the menu launched. If you can't do that inside an hour because the recipe costs aren't written down anywhere, that's the finding. The dishes that sell most are where margin drift does its quiet damage.
What fixed looks like: supplier invoices link to the menu items they affect, so a price change surfaces the week it happens with the decision already framed: reprice, renegotiate or re-spec. You decide in July, not October. The menu stops silently paying for the supplier's problem.
4. Stock pressure
The scene: Saturday night, you 86 the special at 8pm because prep ran short, while a fridge of Tuesday's over-order edges towards the bin. Both happened in the same week. Both were predictable from numbers you already had. Waste and stockouts feel like opposite problems, but they're the same leak: ordering by gut feel while the sales data sits unread in the POS.
How to spot it: track two things for a fortnight. What you threw out, honestly weighed, and what you ran out of during service. Put a euro figure on each. Then compare your last four supplier orders against the actual sales in the same window. If the orders look identical while the sales varied, you're ordering from habit.
What fixed looks like: order suggestions built from what actually sold, adjusted for the bookings already in the diary for next week. Stock counts that take minutes and feed the same picture. Less in the bin, fewer 86'd dishes on your busiest night, and the cash that was sleeping in the walk-in back in the account.
5. Labour mismatch
The scene: a wet Tuesday with three staff watching an empty room, then the following Saturday running two short because a party of twelve booked after the roster was posted and nobody adjusted. Wages are among the heaviest lines on your P&L, and they got heavier again in January. Yet in most restaurants the roster is still built from memory of what this week looked like last year.
How to spot it: for each day of last month, put labour hours next to covers served. The ratio will swing wildly, and the expensive days aren't the busy ones. They're the quiet days staffed for a rush that never came, and the busy days where overtime and comps papered over the gap. Owner burnout, which the closure research keeps flagging, lives here too: the gap in every roster is usually filled by the owner.
What fixed looks like: the roster is built against actual bookings and realistic walk-in patterns, and when a big party lands after posting, someone is told before it becomes Saturday's crisis. You can't control the wage floor. You can control whether you're paying it to the right number of people on the right nights.
6. Cashflow gaps
The scene: month end. The bookkeeper sends the figures and there's less in the account than the busy month suggested there should be. You spend the evening doing archaeology on your own business. The closure coverage keeps describing versions of this pattern: restaurants trading well by the till and failing by the bank account, with owners who saw the real picture too late to change course.
How to spot it: honestly answer one question. Do you know, today, this week's take against this week's outgoings, or will you find out in five weeks when the accounts are done? If the answer is five weeks, every decision you're making right now is based on old information.
What fixed looks like: one daily operating brief that shows money in, money owed, money leaking, and the moves already prepared. Cashflow stops being a monthly autopsy and becomes a daily glance. When you do need funding for stock or equipment, the operating evidence to support it is already assembled rather than reconstructed in a panic.
Add it up
None of these six kills a restaurant in a week. All six run at once, and they compound while your attention is on the pass. The fix starts with a number, because you can't prioritise what you haven't measured.
That's what the free leak audit is for. Five quick answers. One clear number: what the leaks you control are costing you. Run it at resk.ai/leak-audit, or book a demo and start a 30-day trial. The macro costs aren't going anywhere. The leaks can.