How to Start a Restaurant: The Complete Step-by-Step Guide

Most people who want to start a restaurant begin with the fun part (the concept, the menu, the name) and reach the hard parts in the wrong order. They sign a lease before they know their seat count. They buy equipment before the hood is approved. They open before the kitchen can produce the covers the dining room can seat.

This guide is the whole sequence, in the order the work actually has to happen, with the current numbers attached. It covers all ten stages from concept to opening day, and it goes deeper on the ones where decisions are hardest to reverse. Where a stage deserves a full treatment of its own (the business plan, the floor plan, the soft opening) you will find a short summary here and a link to the detail.

The Bottom Line

• US restaurant and foodservice sales are forecast to reach $1.55 trillion in 2026 across more than a million outlets, with the industry expected to add over 100,000 jobs (National Restaurant Association).
• Demand is real but margins are tight: 42% of operators reported their restaurant was not profitable and 60% reported softer customer traffic, with more than 9 in 10 citing food, labour, insurance, energy and swipe fees as significant challenges.
• Budget $238,000-$604,000 for a fast-casual location and $482,000-$1.28 million for full service as 2026 reference ranges, and add three to six months of operating cash on top of the build.
• The order of operations matters more than the speed. Concept, then plan, then funding, then lease: never lease first.
• Expect 9 to 18 months from serious concept work to opening day. Permits and construction are the stages that slip.

What does it actually take to start a restaurant?

Six things, and you need all of them: a concept somebody wants, a site that can produce it at a price that covers the rent, enough capital to reach steady trading, the licences to operate legally, a team that can execute, and the discipline to measure what happens afterwards.

The market context is worth holding onto because it cuts both ways. Industry demand is strong: sales are forecast at $1.55 trillion for 2026 and more than 7 in 10 consumers say they would use restaurants more often if they had more disposable income. But the same National Restaurant Association research found 42% of operators were not profitable, which tells you that demand alone does not produce a viable restaurant. Execution and cost control decide it.

The good news for first-time owners is that the failure story is overstated. A University of California, Berkeley study found about 17% of full-service restaurants closed within their first year: a long way from the “90% fail” line that circulates constantly and has no traceable data behind it. The real reasons restaurants close are specific, documented and mostly avoidable, and why restaurants fail goes through them with the warning signs that appear months in advance. Read it early rather than late; most of what it describes is cheaper to prevent at the planning stage than to fix in year two.

How much does it cost to start a restaurant?

Most independent restaurants open for somewhere between $175,000 and $1.5 million. The spread is driven by three things: square footage, how much construction the space needs, and whether you serve alcohol.

Useful 2026 reference ranges:

• Fast casual: roughly $238,000-$604,000.
• Full service: roughly $482,000-$1.28 million.
• A small build in a space that was already a restaurant, no liquor licence: commonly $150,000-$400,000.

The single biggest variable is the condition of the space. Taking over a working restaurant where the hood, grease trap and most of the plumbing already exist is dramatically cheaper than converting a retail unit or a cold shell, where everything gets installed from scratch and the permit path is longest.

The cost line that sinks the most new restaurants is not in the build at all. It is working capital. You will pay rent, payroll and suppliers for weeks before revenue stabilises, and training week has full payroll with no sales attached. Budget three to six months of operating expenses as a separate pot and do not spend it on a nicer fit-out.

Cost pressure is also compounding year over year. Industry tracking compiled in this set of restaurant industry statistics has food and beverage costs up 21.8% and labour costs up 18.3% over a recent three-year period, which means a budget built on figures more than a year old is already wrong.

Step 1: Define the concept and the customer it serves

A concept is not a cuisine. It is a specific answer to who eats here, how often, at what price, at what time of day, and what they would otherwise have done instead.

Settle these before anything else, because every later decision inherits them:

• Service format. Quick service, fast casual, full service, fine dining. This sets your staffing model, your square feet per seat, and your check average.
• Target check average and covers per day. These two numbers multiplied out are your revenue model. If that number cannot cover rent, payroll, food cost and debt service, the concept is wrong regardless of how good the food is.
• Daypart strategy. A dinner-only concept pays rent for 24 hours and sells for four. Breakfast, lunch, retail or outbound orders change the arithmetic substantially.
• What makes it worth leaving the house for. Be honest. “Good food, good service” is not a differentiator; every competitor claims it.

Test the concept cheaply before you commit capital. Pop-ups, farmers markets, a residency in someone else’s kitchen and catering orders all generate demand data for a fraction of the cost of a lease.

Step 2: Write the business plan

The plan is where the concept gets stress-tested against arithmetic, and it is also the document a lender, investor or landlord will judge you on. Nine sections cover it, following the structure the US Small Business Administration uses: executive summary, company description, market analysis, organisation and management, service or product line, marketing and sales, funding request, financial projections, and appendix.

If your concept sits at the edge of the restaurant category, the emphasis shifts: a bakery plan leans harder on production capacity and wholesale accounts than on covers and table turns, and the essentials of a bakery business plan show where those differences fall.

The financial section is the one that matters most and gets the least attention. It needs a startup cost budget, a month-by-month cash flow projection through at least the first 18 months, a break-even analysis, and the assumptions behind every number stated openly so a reader can challenge them.

Why you need a great restaurant business plan covers what each section has to prove and how to write the parts lenders actually read first. Do not skip the plan on the grounds that you are self-funding; the exercise of projecting cash flow is what reveals an unviable concept while it is still only a document.

Step 3: Secure the funding

Know your total number before you start raising, and raise for the total rather than the build. Running out of money three months after opening is the most common way a solvent-looking restaurant dies.

Common sources, roughly in order of how often they appear in independent restaurants:

• Personal capital and money from family. Almost always part of the mix, and usually what lenders require to see before they participate.
• SBA-backed loans. The 7(a) programme is the most common route for restaurants. Expect to need a business plan, a personal guarantee and often 10-30% down.
• Equipment financing and leasing. Secured against the equipment itself, so it is easier to obtain than general-purpose debt, and it preserves cash.
• Investors. Faster and more flexible than debt, at the cost of equity and some control.
• Landlord contribution. A tenant improvement allowance negotiated into the lease is effectively funding, and it is the most overlooked source on this list.

Step 4: Choose the site and negotiate the lease

Location is the decision with the longest consequences and the one most often made emotionally. Judge a site on evidence: actual foot and vehicle traffic counted at the hours you intend to trade, the surrounding demographics against your target check average, parking and access, visibility, what the neighbouring businesses pull in, and what happened to the last occupant of that unit and why.

On the lease itself, the terms that matter as much as the rent:

• Total occupancy cost, not base rent. Add common area maintenance, taxes, insurance and any percentage rent.
• Term and renewal options. A short term with options is better than a long term without them.
• Tenant improvement allowance and who owns the improvements at the end.
• Exclusivity and use clauses, so the landlord cannot put a direct competitor in the next unit.
• A permit contingency. Never sign a lease that starts paying rent before your use permit and licences are approved. This single clause has saved more restaurants than any other.

Have a restaurant-experienced attorney read it. The fee is trivial against a five-year obligation.

Step 5: Register the business and get the licences

Form the legal entity first (most independent restaurants use an LLC or S-corporation) then get the federal EIN, then work through the permits. Requirements vary by state, county and city, so confirm locally rather than working from a generic list.

The usual set:

• Business licence and a certificate of occupancy
• Food service establishment permit and health department approval
• Food handler and manager certifications for staff
• Liquor licence, if applicable: the longest lead time on this list by far, often 6-12 months, and in some jurisdictions capped in number and expensive to buy on the secondary market
• Sign permit, music licensing, and a resale or seller’s permit
• Employer registrations for payroll taxes and workers’ compensation

Start the liquor licence and the health department process before construction finishes, not after. Both can hold your opening date hostage.

Step 6: Design the floor plan and buy the equipment

The layout decides your capacity, your labour cost per cover and your ticket times, permanently. Fix the kitchen’s position first, because venting, gas and plumbing are the expensive things to move, then fit the dining room into what remains.

As a starting point: allow 12-15 square feet per seat for full service and 18-20 for fine dining, split the space roughly 60% front of house to 40% back of house, and respect the ADA minimum of 36 inches clear width on accessible routes. How to create a restaurant floor plan has the full dimension set, the service-path method and the mistakes that quietly cost you covers.

On equipment, buy the cooking line new where reliability and warranty matter, and buy used for stainless prep tables, shelving and refrigeration you can inspect. Lease what dates quickly. And confirm the hood and fire suppression specification with your inspector before ordering anything, because an undersized hood means the whole line gets re-planned.

Step 7: Build and price the menu

Design the menu against the kitchen you are building, not the one you wish you had. Every additional cooking method adds equipment, every additional ingredient adds a storage slot and a spoilage risk, and every additional item adds ticket time. If you are starting from first principles on structure and sections, what a restaurant menu is covers how one is put together before you get to pricing it.

The discipline that keeps menus profitable:

• Cost every single dish to the gram, including trim waste and garnish, and price from food cost percentage rather than from what competitors charge.
• Keep the item count honest. A shorter menu produces faster tickets, less waste, tighter inventory and better consistency. Long menus are usually a decision nobody made.
• Share ingredients across dishes so each delivery turns over quickly.
• Re-cost quarterly. With food costs having risen more than 20% over a recent three-year span, a menu priced on last year’s invoices is losing margin on every cover.

Step 8: Hire and train the team

Hire the chef or kitchen manager first, and ideally before the menu is finalised so they have ownership of it. Then the front-of-house manager, then the line, then the floor.

Allow two to four weeks between the first hires and opening for training, and budget full payroll for it with no revenue coming in. Run the kitchen through the real menu repeatedly until ticket times are consistent, and have the floor team practise a full service with staff sitting as guests. Written systems (opening and closing checklists, recipe cards, service standards) are what make the second month as good as the first week.

Step 9: Set up suppliers, tech and delivery

Line up a primary broadline supplier plus specialists, and get standing order days and delivery windows agreed in writing before opening week. Negotiate payment terms; net 14 or net 30 is a meaningful working capital improvement over paying on delivery.

The technology that needs to be live and tested before you open:

• POS, with the full menu built, modifiers configured and staff trained on it
• Payment processing, and know your effective rate including the fees that do not appear on the headline
• Accounting and payroll
• Inventory and ordering
• Reservations or waitlist management, if your format needs it
• Online ordering, and a plan for how outbound orders get packed and dispatched without colliding with dine-in service

That last point deserves more thought than it usually gets. If any meaningful share of your revenue will leave the building (takeout, online orders, catering drops, wholesale accounts) then packing space, staging and dispatch are part of the operation from day one, not something to improvise later.

Step 10: Run a soft opening, then open properly

Open quietly before you open loudly. A soft opening runs real service at reduced capacity with a limited menu and an invited or restricted guest list, so that the things that are going to go wrong go wrong in front of a forgiving audience.

It is the only realistic way to test ticket times under load, find the layout problems, calibrate staffing levels and fix the menu items that do not work at volume. What a restaurant soft opening is, and whether you need one covers the formats, who to invite and what each version is good for.

Then hold the grand opening once service is consistent: not before. A grand opening that brings a full house into a kitchen that is not ready converts marketing spend into bad reviews that outlive the promotion by years.

What to measure once you are open

The first ninety days are where you find out which of your assumptions were wrong. Track weekly, not monthly, because monthly is too slow to act on:

• Prime cost (food plus labour as a percentage of sales). The number most operators manage the business by.
• Food cost percentage by category, against your menu costings.
• Labour cost percentage and hours per cover.
• Covers, average check and table turn times by daypart.
• RevPASH: revenue per available seat hour, which measures how hard your seats work rather than just how full the room looked. What RevPASH is and why it matters explains the formula and what it exposes that average check and daily sales both hide.
• Cash position, projected forward weekly. Profit and cash are not the same thing, and restaurants die of the second.

Growing beyond the first location

Do not think about location two until location one has been consistently profitable for a full year and runs well when you are not in the building. The systems that make expansion possible (documented recipes, training programmes, repeatable layout, reliable supplier terms) are built during year one or not at all.

When growth does come, the main routes are a second company-owned unit, licensing, or franchising. Franchising moves the capital burden to franchisees but turns you into a different business with legal obligations attached, including a federally mandated disclosure document. How to franchise a restaurant walks through what that actually involves before you commit to it.

How long does it take to start a restaurant?

Nine to eighteen months is realistic from serious concept work to opening day. A rough distribution:

Permits and construction are where timelines slip, and they slip together. Build float into the schedule and, more importantly, into the cash plan: every month of delay after rent starts is a month of expense with no revenue against it.

Frequently asked questions

How much money do you need to open a small restaurant?

For a small space that was previously a restaurant and does not serve alcohol, $150,000-$400,000 is a realistic range covering the build, equipment, licences and opening inventory. Add three to six months of operating expenses on top as working capital. A larger build, a cold-shell conversion or a liquor licence moves the figure substantially higher.

Do I need restaurant experience to start a restaurant?

It is not legally required, but it is the single best predictor of whether you will survive the first two years. If you do not have it, buy it: hire an experienced chef or general manager early and give them real authority, and spend time working service in someone else’s restaurant before you commit your capital.

What is the most profitable type of restaurant?

Formats with low labour per cover, short menus and fast turns tend to show the best margins: quick service, fast casual, pizza, coffee and bar-led concepts where beverage mix carries the gross margin. But format matters less than execution: a well-run full-service restaurant will out-earn a badly-run fast-casual one every time.

What is the first thing to do when starting a restaurant?

Define the concept and build the revenue model for it: target check average times realistic covers per day, tested against the rent and payroll that concept requires. Everything else follows from that arithmetic, and doing it first is what stops you signing a lease a viable business could never pay for.

Is it cheaper to buy an existing restaurant or start one?

Buying an existing restaurant is usually cheaper and faster to open, because the buildout, licences and equipment already exist and there may be a customer base attached. The risk is inheriting the previous operator’s problems: a bad lease, deferred maintenance, a poor local reputation or the underlying reason they are selling. Have the books, the lease and the equipment independently reviewed before you commit.