Restaurant permits and requirements: the numbers before and after

A lease signed in January and the first sale in September

Five months of rent paid without selling a single plate: that is the real cost of chasing permits after signing, and in the case of the 180-square-meter unit whose exhaust hood vented into a shared interior courtyard the bill passed 96,000 dollars between dead rent and bridge financing. The condominium association held veto rights over the extraction system, and the horizontal property bylaws were available to anyone who asked for them before signing. Nobody asked. At 6,400 dollars of monthly rent, every month of finished construction without a license eats the equivalent of 21,000 dollars in sales at a 12% operating margin, which is the realistic ceiling for casual dining in an established district. The correct sequence flips the order: the regulatory file first, the signature afterward. A complete file gets resolved within the legal deadline; an incomplete one enters a cycle of objections and replies that stretches the calendar by two or three, and you caused that cycle.

The incomplete file multiplies the calendar, not the clerk

A note on scale helps here: according to the International Franchise Association, franchised QSR output reached USD 321.8 billion in 2025 from 305.3 billion in 2024, a 5.4% gain, and that growth is built by groups opening on schedule, not by operators refiling paperwork. Each round of objections at a planning office or health authority burns twenty to forty business days, and three of them pile up when the sealed extraction drawing, the current zoning certificate or the electrical load calculation is missing. Multiply that by your monthly rent and you have the exact price of improvising. Four out of ten sites evaluated under protocol collapse before signing, and each of those rejections is worth more than an average opening because it prevents an operation doomed from the lease. Partners struggle with this: paying for territorial analysis only to end up saying no looks like spending without return until the return gets quantified.

Pre-feasibility is measured by the sites you reject

The average SBA 7(a) loan ran near 542,000 dollars in fiscal 2024, across 57,362 operations totaling more than 31.1 billion, according to the U.S. Small Business Administration. One site rejected in time protects that entire debt structure. The cheap unit with an extraction problem is not cheap, and the expensive one with clean zoning and a 60 kVA electrical service already installed turns out, across the cash flow of the first twelve months, to be the cheaper of the two. Translate the benchmarks to your scale before applying them, because a licensing deadline weighs differently depending on how many locations absorb it. Small restaurant, one site, rent below 3,000 dollars: pre-feasibility means three weeks of your own time plus reading the condominium bylaws, and the risk that kills you is zoning, not electrical capacity. Mid-size operation, two or three sites with rent between 5,000 and 9,000: commission the extraction study before the letter of intent, because one dead month costs more than the full study.

How to read these numbers in YOUR operation?

Group with three or more openings a year: permits stop being a project and become a process with an owner, a traceability matrix and its own calendar.

Diego F. Parra sets a concrete threshold for that third scenario: if the group opens three sites annually, the regulatory role pays for itself by avoiding a single dead month. A multi-unit franchisee runs five locations on average, up from 4.8 in 2011, and 82% of franchised QSRs sit under multi-unit control, against 72% for table-service restaurants, according to FRANdata. That concentration rewrites the economics of the regulatory file entirely. Whoever files one license pays the whole learning curve on that single opening; whoever files five amortizes templates, technical memoranda, drafting vendors and the relationship with the local authority across the portfolio. The second site gets filed with the first one's paperwork corrected, the third with both corrections already folded in, and the curve between signature and first sale drops measurably.

Multi-unit turns the permit into a repeatable asset

So the mistake an expanding group makes is not filing slowly, it is filing every opening as if it were the first and throwing away what it learned. A fund evaluating restaurant investment does not ask about the concept in the first meeting, it asks about opening risk, and there the regulatory file does more for valuation than the menu ever will. With total U.S. franchise output above 936.4 billion dollars in 2025, 4.4% over the 896.9 billion of 2024 according to the International Franchise Association, available capital is broad and what is scarce is the operator who can prove control of the calendar. Present a traceability matrix of the file, the technical handling of extraction installations resolved and documented territorial pre-feasibility, and the conversation shifts to valuation before you show the menu. Consider the counterfactual: if two groups seek capital with identical projected EBITDA and one cannot say when it opens, the risk discount gets applied by the fund, not by the market.

Royalties drive the deadline when a franchise is involved

When the project is franchised, each month of permit delay gets paid twice, because the royalty structure runs on sales that do not yet exist and the lease does not wait either. U.S. restaurant royalties run from 4% to 8% of gross sales, rising to 6%-10% in coffee and dessert concepts, according to Toast; the general average sits at 6.7% of gross revenue, within a 4% to 12% range, according to Franzy. A location billing 90,000 dollars a month hands over between 3,600 and 7,200 in royalty, and that figure only starts making sense once break-even has been crossed. Delaying the first sale by four months does not shrink the royalty, it shrinks the months available to amortize the initial investment inside the franchise agreement term. The term does not stretch; yours gets shorter. The market figures in this material come from public, verifiable sources: the International Franchise Association for sector output in 2024 and 2025, FRANdata for multi-unit structure, the U.S.

Where these benchmarks come from and how far they reach?

Small Business Administration for fiscal 2024 7(a) lending, Toast and Franzy for royalty ranges, and ACODRES via Infobae for the Colombian dining sector's recovery, with sales growing close to 7% in the first half of 2025 after the 2024 drop.

Now the limits, which matter just as much. None of those sources publishes licensing timelines by municipality, because the procedure is local and its duration depends on the authority receiving it. What this material contributes on calendar and pre-feasibility is consultant reading over public data, not a proprietary sample of case files. Always check against the actual timelines in your jurisdiction before committing to an opening date. The difference is not how fast the reviewer works, it is the quality of the file that arrives. A complete application clears within the statutory window; an incomplete one enters a deficiency-and-response loop that doubles or triples the calendar, and you caused that loop, not the agency.

Where a group that opens well pulls away?

Site feasibility carries a counterintuitive payoff that partners resist: its return is measured in the sites it REJECTS.

Four of every ten evaluated locations die before signature under protocol, and each rejection is worth more than an average opening, because it kills an operation that was doomed from the lease. In a group opening three or more units a year, restaurant permits and requirements stop being a project and become a process with an owner, a template and standard cycle times. The first opening teaches; the second should cost half the management hours. If your third opening took as long as your first, there is no process, only repetition. The costliest judgment error I see in restaurant groups is treating the permit file as sunk cost instead of a transferable asset. That file is what gets audited in an equity sale, what lowers an insurance premium, and what a franchisee pays for so they never repeat the work.

Where a group that opens well pulls away — in practice?

Documenting it well is capital, not bureaucracy. And a genuine concession: for years I told operators to close leases fast so they would not lose the site, because competition for square metres in hot districts is real.

My proportions were wrong. More money is lost signing a regulatory landmine in three days than losing two good sites while the third gets validated.