Roger Fields spent ten years as an accountant before opening his own restaurants in New York, and Restaurant Success by the Numbers is the book he wishes someone had handed him first. Its argument is uncomfortable and useful: most restaurants that close were never viable on paper, and the owner could have known that before signing the lease. For anyone opening a café, bistro or bar in Europe, it is the clearest walk-through we know of the arithmetic between a dream and a business plan.
The big idea
A restaurant is a set of numbers that either fit together or do not: the concept decides the average spend, the location decides the rent and the footfall, the menu decides the food cost, and break-even tells you whether the whole thing can pay you a wage. Check the fit before you build.
Who should read it
Read it if you are planning a first restaurant, taking over an existing one, or running a busy place that somehow never has money in the bank. It is written for independents, not chains, and assumes no finance background. Skip it if you want inspiration about food, service or design; the reason to buy this book is the numbers.
Key takeaways
- The famous 90% failure rate is a myth; restaurants close for a short, predictable list of reasons, and most are avoidable.
- Concept, target market and location have to agree with each other; falling in love with two of them before checking the third is the classic first-timer mistake.
- Forecast sales bottom-up (seats, turns, spend, days), then cut the result; overstate costs and understate sales.
- Break-even is fixed costs divided by what is left of each euro after variable costs, and it lets you test a lease, a price or a hire in one minute.
- The money you need is investment plus a contingency plus months of working capital; the reserve is the last thing to cut, not the first.
The 90% failure myth, and what actually closes restaurants
The number everyone repeats, that nine in ten restaurants fail in their first year, has no serious research behind it. Fields digs up the academic studies and finds first-year closure rates closer to a quarter or a third, with survival improving sharply once a place has made it through year three. The myth does two harmful things at once: it scares off the careful, and it hands the reckless an excuse, since if it is a lottery anyway, why plan?
What closes independents, in his experience as an accountant with restaurant clients, is a short and repetitive list. Rent the sales could never carry. Too little cash to survive the slow first months. Too much borrowed for the fit-out, so the repayments eat the profit. A concept that does not match the neighbourhood it lands in. And nobody watching food, drink and labour costs week by week. Franchises fail far less often not because their food is better, but because somebody researched the site and the concept before the money was spent.
His framing is sober: start smaller than your dream, set goals you can measure, accept that the capital you can raise defines the size of place you can open, and surround yourself with people who know what you do not. The rest of the book is that research, done in numbers, without a franchise fee.
Concept, market and location must agree
Fields opened his first restaurant, a Mexican place near Times Square, with a fun concept, a fully equipped kitchen and low rent, and it struggled from day one. The neighbours turned out to be subsidised-housing tenants and theatre-goers with conservative tastes and modest budgets. Only when he rebuilt the concept around who actually walked past the door did the place take off. His lesson: you may fall in love with a concept or with a location, but never with both before you have checked that one fits the other.
The checking is unglamorous. Who lives, works and passes within walking distance, what they earn, what the busy places nearby charge and who sits in them, how many restaurant seats the area already has per resident. In a Belgian or Dutch town that means counting terraces on a Saturday, reading the menus of the three busiest competitors, and asking the local business association who opened and closed lately. Radius depends on concept: a specialist, higher-priced restaurant can pull guests from twenty kilometres away, while a lunch spot draws from a few streets.
Underneath all of it sits the unique selling proposition: one thing your target guest can see and value that the place across the street does not offer. A homogeneous concept, another pizzeria in a street with four, competes on price and cannot raise its prices; a differentiated one can, and guests will travel and pay for it.
Forecast sales conservatively: seats, turns, spend, days
The feasibility study starts with a sales forecast built bottom-up rather than wished for. For each service, estimate the number of seats, how many times each seat is filled (the turns), and what a guest spends on food and on drink, separately, because dinner guests order wine and dessert and lunch guests do not. Multiply by the days you open, then by 4.33 for a month and 52 for a year; twelve months of four weeks quietly loses four weeks of wages.
A 50-seat bistro in Ghent open five evenings, turning its seats 1.5 times at €38 a head, forecasts about €14,250 a week; add three lunch services at one turn and €22 and you are near €17,500. Then cut it, because sales take months to build, August empties the town, and a Tuesday is not a Saturday. Fields is blunt: underestimate sales and overstate costs, never the other way round.
Those per-service averages keep working after opening. A falling average spend at dinner is your earliest warning of a service or portion problem, weeks before the accountant sees it in the quarterly figures.
Fixed, variable, and the break-even you can play with
Costs split in two. Fixed costs run whether the door opens or not: rent, insurance, loan repayments, the manager's salary, the bins. Variable costs rise and fall with sales: ingredients, drink, most kitchen and floor hours, card fees, laundry. Food, drink and labour together, what the trade calls prime cost, take roughly 60 to 65 cents of every euro in most restaurants. That is the bad news; the good news is that those are precisely the costs you can manage, while rent is signed for years.
Break-even follows: divide your fixed costs by the share of each euro left after variable costs. If fixed costs are €9,000 a month and variable costs eat 62% of sales, you need €9,000 divided by 0.38, about €23,700 a month, before the first euro of profit, or roughly €950 a day over 25 opening days. Put that figure next to your forecast and you know at once whether the plan has any margin in it.
The real value is the what-if. The corner unit at €700 more rent needs €1,850 more sales every month, forever, to leave the same profit. Trimming variable cost from 62% to 59% lowers break-even by about €1,700 a month. You can test a lease, a price change or a second cook in a minute instead of rebuilding the whole plan.
How much money you really need
How much does it cost to open a restaurant? Fields calls it the wrong first question, because the answer depends on the ones before it: what type, where, how big, fitted out how. Once the concept is fixed, the capital budget lists everything paid before the first guest sits down: deposit and first rent, design and building work, permits, licences and legal fees, insurance, equipment, tableware, opening stock, the first payroll and the launch marketing.
Then come two lines first-timers skip. A contingency of around a fifth of the total, because renovation always finds something behind the wall and every week of delay means rent and salaries without a sale. And working capital: cash to cover at least three months of fixed costs while trade builds. His own first budget came up €30,000 short. A €250,000 fit-out that opens with an empty account is the most common way a good concept dies in its first winter.
The budget is also what you show investors, so it has to survive contact with reality; going back for more money is the fastest way to lose their confidence. If the pro forma shows €90,000 net profit on €280,000 invested, that is a credible pitch. The same plan with the reserve quietly deleted is not a plan.
The lease and the money: read the fine print
Rent is the one fixed cost you cannot renegotiate afterwards, so it deserves the most suspicion before signing. Fields's benchmark for table service is total annual rent of no more than about 10 to 12% of forecast sales; on €600,000 a year that caps rent near €5,000 to €6,000 a month, whatever the agent says the street is worth. He then wants the owner to understand every clause: when rent actually starts (never before the space is usable), fixed versus index-linked increases, what a net lease passes through in taxes and common charges, whether you may sell or sublet, and whether you are personally guaranteeing the whole term.
He is equally direct about financing. Your own money first, because nobody backs an owner who risks nothing; then family and investors who expect a real return for real risk; and only then the bank, which for a European start-up usually means a loan against your personal guarantee or a public guarantee scheme. Landlord contributions and rent-free months are money too, and often the cheapest. Equipment leasing gets you a range when the bank says no, but read the total repayable first: leases rarely allow an early exit.
And one rule he repeats until it sticks: do not sign the lease until the financing is confirmed. A signed lease with no money behind it is a personal debt with a restaurant's logo on it.
Opening day and beyond: watch the numbers weekly
Full tables are not profit. If the till is busy and the bank account is not, the diagnosis is nearly always the same four leaks: spoilage, waste, theft and over-staffing, and the common cause is that nobody is counting. Five cents lost on every sales euro sounds like nothing until you multiply it by €500,000 a year for ten years.
The tools are simple and weekly. Par levels for every item, expected weekly use plus a small safety stock divided by the number of deliveries, keep stock low, fresh and countable. Cost of sales is opening stock plus purchases minus closing stock, set against that week's sales, and any gap of two points or more from target gets investigated the same day. Standard recipes and a portion scale, because a few grams too many on one dish is worth well over €1,500 a year. Every drink rung in before it is poured, and someone other than the bartender counting the bar.
Each year, raise prices by a couple of percent quietly rather than by ten percent loudly; a €14.50 dish at €14.90 is noticed by nobody. Owners who look at food, drink and labour cost every week, Fields says, are consistently more profitable than those who look monthly, because lost profit cannot be recovered, only stopped.
Put it into practice
- Draft a one-page feasibility study: seats, turns, spend per service, fixed and variable costs, break-even, and the capital you need including a 20% contingency and three months of fixed costs.
- Shop the competition properly: eat at the three busiest places near your site, collect their menus and note prices, average stay and who the guests are.
- Recost your ten best sellers with real invoice prices and real yields, then check each against a target food-cost percentage and its contribution in euros.
- Take your draft lease to someone who reads commercial leases for a living, and set a rent ceiling as a share of forecast sales before you negotiate.
- Start a weekly numbers ritual: count high-value stock, calculate food, drink and labour cost as a share of sales, and investigate any gap of two points or more.
Where the book falls short
The book is American and it shows. The financing chapter is built around US government-backed loans, the licensing section around state liquor law, and the figures are in dollars, ounces and pounds, so every worked example needs translating. The marketing chapter predates social media and reads like it. For Europe the ratios also shift: labour weighs heavier on sales in Belgium or the Netherlands than in Fields's New York, menu prices include VAT so every cost percentage must be computed on ex-VAT revenue, and the book says almost nothing about VAT timing or cash flow, which is where many EU independents actually run out of money. None of that invalidates the method; you just have to bring your own numbers.
Our verdict
Buy it if you are opening, buying or refinancing a restaurant and want to understand the spreadsheet your accountant will build. It is the most patient explanation of restaurant arithmetic we have read, and the lessons outlast its dated examples.
Frequently asked questions
Is Restaurant Success by the Numbers still relevant for a European restaurant?
The method is; the examples are not. Break-even, food-cost pricing, working capital and lease reading work the same in Antwerp as in New York, but you need to swap in European wages, VAT-exclusive prices and your own bank's loan terms.
Do I need an accounting background to read it?
No. Roger Fields writes for first-time owners and explains every calculation with a worked example. If you can divide two numbers you can follow the whole book.
What is the single most useful number in the book?
Break-even: your fixed costs divided by the share of each sales euro left after variable costs. It tells you the daily sales you need before any profit, and lets you test a rent, a price or a hire in a minute.
Is it only about opening a restaurant?
Mostly, but the last chapter on par levels, weekly cost of sales, portion control and theft applies to any running venue, and it is the part busy owners tend to need most.
This is our own reading of the book, not a substitute for it. Buy the book from your local bookshop.