How toOpen aRestaurant

Each section opens with a thesis— the whole argument of the page in one line. If the thesis doesn’t apply to you, skip the page. Two short paragraphs follow, then the substance sits in a list or a framed block.

Figures like 28–35% are starting points to test against your own numbers, never laws. Every one is illustrative. Where a page touches money, law or employment, it says so and points you at a professional.

Deciding not to open, or to wait a year, is a good outcome. This guide is written to help you get that answer cheaply.

Choosing chairs, sketching logos, browsing listings — it all feels like progress. None of it is. It is also cheap and enjoyable, which is exactly why founders lose three months there.

This is the only stage where changing your mind is free. Everything after it — lease, build, hiring — is expensive and hard to reverse. Spend the time here.

Restaurants are demanding in three specific ways. The hours land on evenings and weekends. Cash leaves months before it arrives. The work is relentlessly about people — staff who quit, guests who complain, suppliers who are late.

None of that is a warning. Plenty of people find those three things energising rather than draining. It makes this a question of fit, not nerve.

“Modern Italian” is a category, not a concept. It tells you nothing about menu size, price, hours, room, or who walks in. A concept decides all of it at once, which is why getting specific saves money at every later step.

Five expensive decisions, made by one sentence.

Location advice is vague because good locations are specific. The version that works is boring and physical: go and stand there with a timer.

Six counts costs about two hours and replaces the most expensive assumption in the plan. If your concept is a lunch counter and the lunch count is thin, you just saved yourself a five-year lease. A ceiling worth holding: rent plus service charges no more than 8–10% of projected sales. Above that, every slow month gets loud.

A cover is one guest served. A turn is how many times a seat is used in a service. Multiply seats by turns by days by average check and you have a week.

Run it three times. The middle column is what you hope for. The left column decides whether you sign.

Empty streets are the worry, not busy ones. You are not asking whether the area is served — you are asking which specific gap is open.

Copy the busiest place and you become its second-best version.

Eat at five places within a ten-minute walk, at the daypart you plan to own. Take notes at the table; you will forget by the second one. Budget $60.

Prime cost is food plus labor — it moves with sales. Fixed costs are rent, utilities, insurance, software, accounting and your own salary — they arrive whether anyone shows up.

Working targets for full service: food 28–35%, labor 30–35%, rent inside 8–10%. Starting points to test against your own menu, not laws.

Opening cost depends so heavily on the building that any single figure misleads. A former restaurant with working extraction and a landlord contribution costs a fraction of a bare shell on the same street.

So do not start with a total. Start with the lines, price what you can, get three written bids for what you cannot, and add contingency. Every funding route trades money for something else. Savings cost your safety net. Bank and SBA lending is cheapest but slowest. Leasing preserves cash and costs more over time. A landlord contribution is the most overlooked and most negotiable. Friends and family is fastest and changes relationships — put it in writing either way.

There is a wide gap between friends saying the food is good and strangers paying for it twice. You can close it in a few weeks for very little money — and everything you learn makes the expensive decisions better.

Measure three things: what people ordered, what they paid without flinching, and how many asked when you are doing it again.