How Much Does It Really Cost to Open a Restaurant? A Realistic Startup Budget

Every aspiring owner asks the same question first: how much is this going to cost? It's the right question, but it's usually answered badly — either with a single headline number pulled from a franchise brochure, or with a fit-out quote that ignores everything that happens after the doors open.

Both mistakes point the same direction: toward opening undercapitalised. I covered why that's the single most common reason restaurants fail in an earlier piece. This one is about the budget itself — what it typically costs, where estimates go wrong, and the line almost everyone underfunds.

The headline number — and why it's nearly meaningless alone

Industry surveys of restaurant owners put the median cost to open an independent, full-service restaurant at roughly $375,000, or around $113 per square foot. The middle 50% of respondents typically report costs between about $175,000 and $750,000, depending heavily on build-out complexity, location, and whether the space needs work from bare shell or is already fitted for food service.¹ A separate survey measuring cost per seat put the median at roughly $3,046 per cover for leased space.²

That spread — $175,000 to $750,000 for the same broad category of restaurant — is the point. A single "average cost" figure is close to useless for budgeting your own concept. What matters is building your own number from the categories below.

The line every budget underestimates: working capital

Every category above is a one-off. None of them is what actually determines whether the business survives its first year. That's working capital — cash held back specifically to cover trading losses while the restaurant builds a customer base, separate from the money spent getting the doors open.

In my own experience across seven franchise openings and an independent restaurant, the sites that struggled weren't usually the ones with the cheapest fit-out or the smallest kitchen. They were the ones where every available dollar went into the build, leaving nothing to absorb a slow first quarter, a delayed equipment delivery, or a supplier who wanted payment faster than expected. A modest fit-out with six months of reserve behind it is a stronger position than an impressive one with none.

Lease versus buy, build versus take over

Buying the property adds a large capital outlay but removes long-term rent risk — rarely the right call for a first restaurant. Leasing keeps upfront capital lower but means rent is a fixed cost for the life of the lease, which is exactly why rent above 10% of revenue is so dangerous: you can't renegotiate your way out of it once trading starts.

Taking over an existing restaurant space — ideally one with a working commercial kitchen already installed — is consistently the cheapest way to open. You inherit grease traps, extraction, and often equipment, which is where the bulk of build-out cost usually sits. A bare shell with no prior food service use is the most expensive starting point by a wide margin.

A worked example: a 60-seat casual dining restaurant

These are illustrative, round numbers to show how the categories stack up — not a quote for any specific site:

• Lease deposit & bond: $15,000
• Build-out (existing food service space): $120,000
• Kitchen equipment: $80,000
• Furniture & fit-out: $30,000
• POS, licensing, opening inventory, marketing: $25,000
• Subtotal — capital cost: $270,000
• Working capital reserve (4 months of a $35,000/month operating base): $140,000
• Total realistic budget: $410,000

Notice the reserve is more than half the capital cost. Most first-time budgets built by prospective owners have no equivalent line at all.

Financing it

Most independent restaurants are funded through some combination of owner equity, a bank or SBA-backed loan, equipment financing or leasing, and occasionally outside investors. Lenders typically want to see 20–30% of the total in owner equity before financing the rest, along with a credible cash-flow forecast — which is exactly where a properly built weekly P&L model, rather than a hopeful revenue guess, makes the difference between an approved application and a declined one.