How do I know if I’m over-staffed?

Compare your labour cost to your sales. If labour is more than 30% of your revenue, you are over-staffed. For example, if you serve £5,000 a night and pay £1,800 in wages, you are at 36%. Aim for 25% or lower. Use a report that tracks labour as a percentage of sales—not just total wages.

What if my staff refuse to cross-train?

Start with one role they already do well. If a server is good at taking orders, train them to run food to tables during slow periods. Offer a small bonus (£20 a shift) for extra skills. If they still resist, ask why—it might be a scheduling issue, not a training one.

Can I really add £5 per table without annoying customers?

Yes, if you frame it as a recommendation, not a sale. Instead of “Would you like fries?” say, “Most customers add fries—they go really well with the burger.” It works because you are not pushing, you are guiding. Track which upsells work (e.g., desserts sell better after 7pm) and train staff to suggest them at the right time.

What’s the fastest way to cut labour costs without losing service?

Focus on idle time first . If you eliminate 30 minutes of waiting per server per shift, you save £1,500 a month for three staff. Next, speed up table turnover—even 10 minutes per table adds £420 a night. Use data to match staff to demand, not guesswork.

Do I need a fancy POS system to fix this?

No. You need a system that shows you labour as a % of sales , kitchen bottlenecks , and customer wait times . Berryin’s Starter plan covers these basics for £90 a month—no hidden fees, no commission. If you need custom reports or multi-branch control, upgrade to Growth or Pro. Start by running a labour-cost report for the last month. If your numbers are over 30%, pick one fix—dynamic scheduling, upselling, or faster turnover—and test it for a week. The savings will cover the cost of the tools you need.

Labour Shortages: How to Keep Staffing Costs Down

Labour shortages are costing you £10k a month. Here’s how to stop it.

The problem is not that you cannot find staff. It is that you cannot keep them. Turnover in hospitality is 25% higher than in other sectors, and the cost of replacing one server is £1,500—before you account for the lost sales while the new hire learns the menu, the till system, and how to handle a drunk customer without calling the police. If you lose two staff a month, that is £36,000 a year in direct costs, plus the opportunity cost of slower service, more errors, and customers who walk out because the queue is too long.

The usual fixes—raising wages, offering flexible shifts, or cross-training—only go so far. Wages are up 12% year-on-year, and flexible shifts mean you are paying for time staff are not working. Cross-training turns a server into a runner, a runner into a cleaner, and a cleaner into a host, but it does not fix the core issue: you still need more bodies to cover the same hours, and the market will not supply them at a price that keeps your margins intact.

The answer is not to hire more. It is to work smarter with the staff you have. That means three things:

Stop paying for idle time.
Turn every minute of labour into revenue.
Make the hours you pay match the hours you need.

None of these require you to fire staff or cut service. They require you to see labour as an asset to optimise, not a cost to absorb.

1. Stop paying for idle time

A server spends 40% of their shift waiting. They wait for tables to be seated, for orders to be taken, for food to arrive from the kitchen, for customers to decide what they want, and for the till to process a payment. You pay them for that time, but it generates zero revenue. If you have three servers on a Friday night, and each one is idle for 2.5 hours, that is £1,500 a week in dead money—before you account for the fact that those servers could have been serving customers instead.

The fix is not to cut staff. It is to eliminate the gaps where they are not working. Here’s how:

A. Front-of-house: Speed up the customer journey

Pre-book tables. If 60% of your diners book ahead, you can staff shifts to match demand, not guesswork. A host can seat a table in 90 seconds; without a booking, they spend 10 minutes trying to fill gaps. That is 50 minutes of idle time per table, per shift.
Use a digital menu. Customers who browse a tablet or QR-code menu decide 30% faster than those with a paper menu. Faster decisions mean less time spent hovering, fewer “I’ll just have a look” moments, and servers who can move on to the next table.
Train staff to upsell in 10 seconds. A server who can suggest a starter or dessert while taking the main order adds £5 to the bill without adding a minute to the table time. That £5 covers the cost of their idle time for the next customer.

B. Back-of-house: Stop the kitchen from holding up service

Track bottlenecks in real time. If the fryer is the slowest station, and it is down for 20 minutes every hour, that is 1.3 hours of idle time per server per shift. Fixing the fryer (or cross-training someone to cover it) saves £300 a week in labour costs.
Use a kitchen display system that shows order times. If a server sees a dish is taking 15 minutes instead of 5, they can tell the customer upfront and keep them engaged—no more pacing the floor waiting for food.
Batch similar orders. If you get three orders for the same burger at once, cook them together. It saves 10 minutes of labour per order, and those 10 minutes add up to £1,200 a month in saved wages.

The cost of inaction: If you do nothing, idle time eats 30% of your labour budget. That is £3,000 a month for a 10-seat restaurant, £15,000 for a 30-seat one. The fix costs nothing—just better scheduling and a few small process changes.

2. Turn every minute of labour into revenue

Labour is not just about serving food. It is about creating opportunities for customers to spend more. The average table spends £35. If you can add £5 to that—through upselling, modifiers, or faster turnover—you cover the cost of the labour that generated it.

A. Upsell without being pushy

The “recommendation” trick. Instead of “Would you like fries with that?”, say “Our customers love the garlic fries with this burger—they pair really well.” It adds £3 to the bill 40% of the time, and customers feel like they got a personal tip, not a hard sell.
Bundle deals. “Burger + fries + drink for £12” sounds like a discount, but it is a £3 upsell per customer. If you serve 50 customers a night, that is £150 extra revenue with no extra labour.
Train staff to listen for cues. If a customer says “I’m starving,” they are more likely to add a starter. If they say “I’m watching my calories,” they will skip dessert—but you can suggest a lighter option that still adds £4 to the bill.

B. Faster turnover = more tables = more revenue

Set a target for table time. If the average table takes 50 minutes, and you have 20 seats, you can serve 24 customers an hour. If you cut that to 40 minutes, you serve 30 customers—12 more tables, or £420 extra revenue at £35 a head.
Use a digital bill to speed up payments. Customers who pay by card take 30 seconds longer than those who pay cash. If you offer a 5% discount for cash, you cut that time in half and avoid the 1.5% card fee (which is £50 a month at 50 transactions).
Clear tables while the customer is still eating. A server who takes the empty plates away as soon as the main course arrives can seat the next customer 10 minutes earlier. That is 24 extra seats a night, or £840 extra revenue.

The math: If you add £5 per table and serve two more tables an hour, that is £15 extra revenue per customer. At 50 customers a night, that is £750 a week—enough to cover the wages of a part-time server.

3. Make paid hours match needed hours

The biggest waste is paying staff for hours they are not working. If you have a 10-seat restaurant and staff for 12 seats, you are over-serving 20% of the time. If you under-staff, you lose customers. The solution is dynamic scheduling—matching labour to demand in real time.

A. Use data, not guesswork

Track your “busy hours” by day. If Tuesdays are slow and Fridays are packed, staff accordingly. If you have three servers on a Friday night but only need two for the first hour, send one home early and bring them back for the rush.
Monitor no-shows. If 15% of your staff fail to turn up, you are paying for labour you never use. Require advance notice for shifts, and charge staff for no-shows (£50 per missed shift covers the cost of a temp agency call).
Use a labour-cost report. If you know your food cost is 30% and your labour cost should be 25%, but it is running at 35%, you have a problem. A system that tracks labour as a percentage of sales lets you adjust before it hits the bottom line.

B. Cross-train for flexibility

Turn servers into runners. If you have a slow period, redeploy staff to take orders to tables instead of waiting for them. It keeps them busy and reduces the chance of a customer walking out because their food is cold.
Train kitchen staff to help with prep during quiet periods. A chef who can chop veggies for the next day’s soups in a slow hour is earning their wage while not holding up service.
Use part-timers for peak times. If you need an extra pair of hands for lunch but not dinner, hire a student for four hours a day. It costs £60 a day instead of £120 for a full-time server.

The cost of bad scheduling: If you over-staff by 10%, that is £2,000 a month in wasted wages. If you under-staff, you lose £500 a night in sales when customers leave because the queue is too long.

What breaks when you try to cut labour costs

The risks are not what you think. You will not lose customers if you improve service, not reduce it. The dangers are:

Over-automating. If you replace staff with self-ordering kiosks but the system crashes, you lose sales. If you cut too many servers, queues get longer, and customers leave.
Burning out your team. If you cross-train everyone to do everything, no one is good at anything. Staff quit when they feel like cogs in a machine.
Ignoring the data. If you cut labour based on last month’s sales but demand has dropped 20% due to a local event, you will be under-staffed when it matters.

The fix: Test changes in small doses. Try dynamic scheduling for one week, track the results, then adjust. If table turnover speeds up but complaints go up, you are moving too fast.

The tools that actually help

You do not need expensive software to fix this. You need:

A system that tracks labour as a % of sales. If you know your target is 25%, you can adjust before you hit 30%. Berryin’s reporting tools let you set alerts for when labour costs spike.
A way to see kitchen bottlenecks in real time. If the grill is the problem, fix it before it slows down service. Berryin’s kitchen display shows order times so you can spot delays before they hit the customer.
A menu that encourages faster service. If your most popular dish takes 20 minutes to cook, customers will wait—but if you offer a “quick burger” that is ready in 5, they will order it instead of leaving.

What you do not need:

A “staffing algorithm” that promises to cut costs by 15%. Algorithms do not know your kitchen, your staff, or your customers.
More screens. If your staff are spending time staring at tablets instead of serving, you have the wrong system.
A consultant who charges £500 a day to tell you what you already know.