Restaurant Management: A Practical Guide for Owners and Managers

Restaurant management is the job of keeping four things true at once: the food is good, the guests come back, the staff stay, and the money works. Most guides treat those as separate subjects. In a real restaurant they are one subject, because every decision you make about one of them shows up in the other three by Friday.

This guide covers the whole job for someone who already has the doors open: what a manager actually does all day, the handful of numbers that tell you whether the week went well, how to control food and labor, how to divide responsibility once you have more than a few people, and which systems earn their subscription. If you have not opened yet, start with the true cost to open a restaurant instead; the budget you set before signing a lease shapes nearly everything below.

The Bottom Line

• Restaurant and foodservice sales are forecast to pass $1.55 trillion in 2026, but more than 9 in 10 operators call food, labor, insurance, energy and swipe fees significant challenges, and 42% reported their restaurant was not profitable last year (National Restaurant Association, 2026).
• Prime cost (food plus labor) is the number that decides the month. Most independents aim for 28-35% food cost, 25-35% labor, and a combined figure under 65% of sales.
• Accommodation and food services posted a 3.5% monthly quits rate in August 2026, the highest of any US industry (U.S. Bureau of Labor Statistics, 2026). Scheduling and training are cost control, not HR paperwork.
• A weekly review beats a monthly one. By the time a monthly P&L arrives, the month you could have fixed is over.

What restaurant management actually covers

Restaurant management is the coordination of six areas that each run on a different clock: money, people, food, guests, compliance, and the channels orders come through.

The last row is the one most operators manage by accident. Orders now arrive through several doors with wildly different economics, and your own website is the only door where you keep the full margin and the customer’s contact details. If you are choosing or replacing that platform, we compare restaurant website builders in a separate post. This one stays on the operating side of the business.

Nothing on that list is optional, but they are not equally urgent. Money and people are where a restaurant dies. Food and guests are where it gets a reputation. Compliance is where it gets shut down.

What does a restaurant manager do day to day?

A restaurant manager spends the day moving between three jobs: setting the shift up, holding the standard during it, and recording what happened so the next shift starts better.

A workable shape for a single day:

• Before service. Check the schedule against the reservation book and yesterday’s sales. Walk the line and the walk-in. Confirm prep is where it should be and that nothing is about to expire. Read the shift notes from the night before.
• Pre-shift huddle. Five minutes, standing up. Covers expected, the 86 list, specials, one service point to work on, one thing that went well yesterday.
• During service. Expedite when it’s busy, watch tickets times, handle the two or three guest problems that need a manager, and keep an eye on who is carrying the shift and who is drowning.
• After service. Cash and comps reconciled, closing checklist signed off, shift notes written for tomorrow: what sold out, what sat, what broke, who did well.
• Once a week, off the floor. Two hours with the numbers, the schedule, and the order guide. This is the part that gets skipped and the part that pays.

The career side of this role (pay bands, titles, how people get there) is a different subject. This post is about doing the job, not getting it.

The five numbers to check every week

Five figures tell you almost everything about a restaurant’s health: total sales, food cost percentage, labor cost percentage, prime cost, and cash in the bank.

Targets vary by format. A pizza shop runs food cost in the low twenties and labor higher; a steakhouse does the reverse. What matters is your own trend, not the benchmark.

Prime cost is the one to post on the office wall. It bundles the two costs you actually control into a single figure, and once it passes roughly 65% of sales there is usually not enough left to cover rent, utilities and insurance and still show a profit. Industry reporting for 2026 put average restaurant labor at about 28.9% of sales (Boostpoint, 2026), which leaves very little room on the food side for a full-service menu.

Check these weekly, not monthly. A monthly P&L is a history lesson. A weekly number is a decision you can still make.

How to control food cost and inventory

Food cost is controlled in four places: what you buy, what you store, what you portion, and what you throw away. Software helps with the arithmetic, but the savings themselves come from one of those four.

Count what moves. You do not need a full inventory every week. Count the twenty or so items that make up most of your food spend (proteins, cheese, oil, produce you buy in volume, alcohol if you pour it) and count them on the same day at the same time.

Compare theoretical to actual. Theoretical food cost is what your recipes and sales say you should have spent. Actual is what your invoices and counts say you did. The gap is waste, over-portioning, comps, or theft, and you cannot find it without costed recipes.

Run FIFO properly. First in, first out only works if things are labelled with dates and the labels are believed. One unlabelled container in the walk-in is how a $300 case gets dumped.

Price the menu off real costs. Recost your top sellers at least quarterly. When a key ingredient moves 20%, you either change the price, change the portion, or change the dish. Doing none of the three is a decision too, just a bad one.

Treat waste as data. A clipboard by the bin with three columns (item, rough quantity, why) for two weeks will tell you more than a month of guessing. Prep over-production and spoilage usually dominate, and both are schedule problems disguised as kitchen problems.

Hiring, scheduling and the turnover math

Turnover is the most expensive line in a restaurant that never appears on the P&L. Accommodation and food services had the highest quits rate of any US industry in August 2026 at 3.5% for the month, roughly twice the all-industry average (U.S. Bureau of Labor Statistics, 2026). Every departure costs you a hiring cycle, a training cycle, and a few weeks of slower, more error-prone service.

What moves the number, in rough order of effect:

1. Publish the schedule further out. Two weeks beats one. People quit over unpredictability more than over pay.
2. Schedule to demand, not to habit. Pull the last six to eight weeks of sales by hour. Most restaurants are overstaffed for ninety minutes and underwater for forty.
3. Train past week one. A new server who still doesn’t know the menu in month two will leave in month three. Build in a thirty-day and ninety-day check-in.
4. Cross-train on purpose. Someone who can run host, expo and a section is a scheduling asset and harder to lose, because the job keeps being interesting.
5. Promote from inside and say so. It is the cheapest retention tool available and the one most often left unused.

Labor is also where management structure starts to matter. One person cannot hold the standard across sixty shifts a month.

How restaurant management responsibilities get divided

Most independents land on three levels: an owner or general manager who owns the numbers and the hiring, one or two shift managers who own service, and senior hourly leads (a head chef or sous, a bar lead, a shift lead) who own a station or an area.

A division that works:

• Owner / GM. Prime cost, pricing, vendor relationships, hiring, the schedule’s final shape, compliance. Signs the checks and the inspection paperwork.
• Kitchen lead. Order guide, prep lists, food cost at the station level, kitchen training, food safety logs.
• Front-of-house lead. Floor plan and sections, service standards, guest recovery, host and server training.
• Shift manager. Everything inside one shift: the huddle, the saves, the close, the notes.

The rule that matters more than the chart: one owner per number. If food cost belongs to everyone, it belongs to no one, and it will drift for a quarter before anybody mentions it. Writing down who owns which number is usually the single most valuable hour an owner spends.

The restaurant management tech stack that pays for itself

Four systems cover most of what a restaurant needs, and each one should be removing a specific recurring task:

• POS. The source of truth for sales by item, by hour, by channel. If you cannot pull that from yours, it is costing you money.
• Scheduling and time tracking. Should forecast labor cost against projected sales before you publish, not after you run payroll.
• Inventory and recipe costing. Mainly useful for the theoretical-vs-actual gap. Worth it if you have more than about twenty-five menu items.
• Your own website and ordering channel. The only sales channel you own outright.

Resist buying the all-in-one before you have the basics working. An integrated suite that nobody updates is worse than a spreadsheet somebody actually fills in. Make each tool earn its subscription by naming the task it removes and checking in thirty days whether that task is gone.

Managing delivery and takeout as its own channel

Off-premise orders need to be managed as a separate business line with its own margin, not folded into dine-in numbers where their economics disappear.

The core problem is commission. Major marketplace apps charge roughly 15-30% per order, and Uber Eats raised its US Marketplace commission to 20% in March 2026, up from 15% (ChowNow, 2026). Against an industry net margin usually quoted in the 3-5% range, a 20-30% commission can turn a busy delivery night into a losing one, before packaging, promotions and refunds even enter the sum.

What to actually do about it:

• Report delivery separately. Sales, food cost and net contribution per channel. A dish that works at the table can lose money in a box.
• Push repeat orders to your own channel. Marketplaces are good at finding you new customers and bad at letting you keep them. Printed inserts, a QR code on the bag, and a reorder email are the usual tools.
• Price off-premise deliberately. A separate delivery menu with its own prices is standard practice, not sharp practice.
• Pick the items that travel. Fried food, anything that steams in a box, and delicate plating all arrive worse than they left. Trim the delivery menu to what survives twenty minutes in a bag.
• Decide who drives. Marketplace couriers, your own staff, or a delivery provider you book directly each have a different cost per order and a different amount of control over how the food arrives. Work out the per-order cost of each for your own volume before committing to one.

A weekly restaurant management routine

Most of this guide collapses into about two hours a week, done on the same day every week.

1. Pull last week’s sales by day, by channel, and against the same week last year.
2. Count the top twenty food items and calculate food cost percentage.
3. Run labor cost percentage from the payroll report, then add food to get prime cost.
4. Look at the cash position and the next two weeks of bills.
5. Build next week’s schedule against projected sales by hour, and publish it.
6. Review the order guide before you place orders, with the count sheet in front of you.
7. Pick one thing to fix. A dish, a station, a service step, a vendor price. Then name who owns it.
8. Read your last ten reviews. Not to reply to all of them, but to notice the same complaint appearing three times.

Eight steps, one morning. A restaurant that does this every week outperforms one that does a brilliant version of it every quarter.

Frequently asked questions

What is the most important skill in restaurant management?

Reading numbers quickly enough to act on them. Good operators are not necessarily the best cooks or the warmest hosts, but they notice that food cost moved two points this week and know which three things could have caused it.

What is a good prime cost for a restaurant?

Under 65% of sales is the common target for independents, made up of roughly 28-35% food and 25-35% labor. The right split depends on your format: higher food cost and lower labor for a counter-service concept, the reverse for full service.

How often should a restaurant take inventory?

Count your highest-spend items weekly and run a full inventory monthly. Weekly partial counts catch problems while you can still fix them; the monthly full count is what your accountant needs.

Do small restaurants need restaurant management software?

A POS that reports sales by item is close to essential. Beyond that, add scheduling software when you pass about fifteen employees, and inventory software when your menu passes about twenty-five items. Below those thresholds, a disciplined spreadsheet competes well.

How do you manage a restaurant with high staff turnover?

Shorten the time it takes a new hire to become useful. Written station guides, a named trainer per role, and a thirty-day check-in cut both the cost of each departure and the rate of departures, because most people leave in the first ninety days.

Where to start

If you read this and want one change rather than ten, make it the weekly two hours. Prime cost, cash, and next week’s schedule, same morning every week, written down. Almost every other improvement in this guide depends on having a regular moment where someone looks at the numbers and decides something.

From there, pick the area with the worst trend rather than the one you enjoy most. Food cost drifting? Start with counts and costed recipes. Losing people? Start with the schedule. Delivery busy but unprofitable? Start by reporting it separately. Restaurant management rewards doing one thing properly far more than it rewards doing everything a little.