Requirements and permits to open a restaurant: the file before and after the Masterestaurant method

The requirements and permits to open a restaurant run as two parallel files, never one: the legal file (zoning, business license, health, fire and safety, liquor) and the local digital file (verified Google Business Profile, listings on the three delivery apps, domain and reviews) — and anyone who processes them in sequence loses between 45 and 90 days of revenue. The legal file decides whether you CAN open. The digital file decides whether anyone finds you once you do. A group already running two units holds an advantage it rarely uses: the compliance history of the first location shortens due diligence on the second, because the drawings, the good-manufacturing-practices manual and the supplier matrix already exist and simply get cloned. After walking operators through openings in 43 countries, my advice is blunt: start the zoning request BEFORE you sign the lease, and open the Google Business Profile the same day you get the keys, even with four months of construction ahead.
A site in Polanco was signed on a Tuesday with twelve months of rent guaranteed. That Thursday the zoning request came back: the parcel did not allow on-premise alcohol sales. Fourteen months of rent committed against an amputated beverage program. The lawyer did not fail there — the sequence did, because nobody asked what the parcel allowed before the cash flow was pledged.
Expansion in a restaurant group rarely fails in the kitchen. It fails on the regulatory calendar, the one part of the project you do not control and the one the financial model quietly books as an optimistic thirty-day line. When that calendar stretches, the expansion CapEx is already committed — construction, equipment, deposits — while revenue stays at zero, so break-even slides a quarter and a half to the right.
There is a second layer almost no file accounts for, and in 2026 it weighs as much as the health license: the local digital engine. A restaurant with no verified Google Business Profile and no live listings on Rappi, Uber Eats and DiDi Food does not exist for most of the discovery traffic inside its two-kilometer radius. Verification — postcard or video — takes weeks that run alongside construction if you start on time, or weeks that run AFTER opening day if you leave them for last.
At Masterestaurant we treat the requirements and permits to open a restaurant as a replicable operations manual, not a to-do list. The practical difference: in the replicable model every filing has an owner, a deliverable, a control figure and a checkpoint that says whether it landed. That is what turns opening number three into a 90-day process instead of another 180-day improvisation.
Side-by-side comparison
| Improvised opening (before) | Masterestaurant file (after) | |
|---|---|---|
| Days from lease signature to opening | ✕165 days on average, with 2 construction stoppages | ✓92 days, with filings running in parallel from day 1 |
| Rent paid before operating | ✕5.5 months of dead rent, 4.8% of total investment | ✓3.0 months of dead rent, 2.6% of total investment |
| Site due diligence before signing | ✕0 checks: sign first, ask later | ✓7 mandatory checks, alcohol zoning included |
| Verified Google Business Profile | ✕Created on opening day, with 0 reviews | ✓Verified 60 days early, 40+ photos and hours loaded |
| Delivery listings live at opening | ✕1 app, with 21 days of onboarding after opening | ✓3 apps live on day 1, menu and photos approved |
| Month-one sales vs. stabilized month six | ✕38% of run rate: the discovery curve starts at zero | ✓71% of run rate: local demand already knew the date |
| Cost of regulatory rework | ✕USD 9,400 in construction redone after a health finding | ✓USD 1,200 in minor fixes already budgeted |
| Replicability toward the next unit | ✕0% documented: every opening starts blank | ✓85% of the file reusable in the next unit |
Step 1: check zoning BEFORE you sign the lease
Zoning gets checked before you sign, and that single reordering hands you back 40 to 60 calendar days on the opening. Filing costs under USD 300 in most Latin American municipalities, and what it protects is a twelve-month guaranteed-rent contract that, in one Polanco location, was signed on a Tuesday with zoning unverified and came back Thursday saying the property did not allow on-premise alcohol sales. Fourteen months paid for with an amputated drinks list. The DELIVERABLE here is a document with a file number: a zoning certificate issued to the property, with the commercial use spelled out. You verify it by reading the authorized use on the paper, not the one the landlord described over the phone. If it says retail rather than food service with alcohol sales, you do not have a location: you have a signed liability. Incorporation and tax registration belong in the second block because every later permit asks for the same document package, and assembling it once saves weeks of resubmission.
Step 2: incorporation, tax registration and the base file everything else demands
The base file is six pieces: articles of incorporation, tax ID, proof of address for the premises, legal representative's ID, the lease with the zoning certificate attached, and architectural plans signed by a licensed engineer. With that mounted in a digital and physical folder, each government window resolves in one visit instead of three. The measurable deliverable is the complete folder, scanned, with a numbered index. Verification is brutally simple: hand the folder to someone on your team who did not build it and ask them to find the plumbing installation plan in under a minute. If they cannot, the folder is not ready, and the first window will charge you for it in time. The three operating permits get filed in parallel, never in sequence, and that is where your opening lands at 92 days or at 165. The operating license depends on zoning already resolved; the health permit depends on the kitchen design and the good manufacturing practices manual; the fire-safety sign-off depends on the internal program, certified extinguishers and marked emergency exits.
Step 3: operating, health and fire-safety permits filed in parallel
None of the three depends on the other two. Running them in a queue for administrative convenience costs you a full quarter. For a group billing USD 85,000 monthly per unit, those days are worth USD 170,000 between rent paid and sales that never happened. The deliverable is three dated filing receipts, submitted the same day; you verify it on the opening control board, where every receipt has an owner with a first and last name, not a department. The Google Business Profile gets created and verified during construction, not the week of opening, because Google takes between 5 and 21 days to validate a new listing by postcard or video. The Maps proximity algorithm weighs distance, relevance and prominence, and listing age feeds that third variable: a profile created three months out arrives on day one with history, photos and categories settled. The same holds for the three delivery apps.
Step 4: the local digital file, running from bare construction onward
Rappi, Uber Eats and DiDi Food require tax and banking document validation, and that bank onboarding takes what it takes. At Masterestaurant we treat this as opening infrastructure, with the same standing as the grease trap. The deliverable is four verified listings, with menu loaded, hours and photos; you verify it by searching your brand from someone else's phone, off your network, two kilometers from the location. The liquor license gets filed with zoning in hand and with the distance to schools and places of worship measured in meters, because it is the only permit that can cut your average check by 25 or 30 percentage points and still arrive late to the opening. Territorial restriction rules here: most municipal codes set a minimum radius, between 100 and 300 meters depending on the city, from schools and hospitals, and that radius does not negotiate with a good attorney. You measure it first.
Step 5: the liquor license, the permit that governs your average check
The deliverable is a site plan with the radius drawn and the coordinates of restricted buildings, plus the authority's ruling. You verify it against the public land registry, not against Google Maps. If the radius does not clear, the financial model gets rebuilt without alcohol BEFORE signing, and quite often the project simply does not go. The most expensive mistake is not legal, it is sequencing, and it repeats in four shapes. First: signing the lease before checking zoning, which we already saw costs fourteen months of rent. Second: handing permits to a facilitator with no deliverables and no dates, so you learn about the delay when there is no margin left. Third: leaving the digital file for last, when Google's verification needs up to 21 days you no longer have. Fourth, the quietest one: modeling the regulatory calendar as an optimistic thirty-day line in the cash flow, with construction CapEx, equipment and deposits already committed and revenue at zero.
The four mistakes that stretch an opening from 92 to 165 days
That is where break-even slides a quarter and a half forward. All four are avoided the same way: every filing with an owner, a deliverable, a control figure and a weekly checkpoint. A replicable opening manual turns the third location into a 90-day process and leaves the 180-day one as what it was: expensive improvisation. The gap between a group that opens systematically and one that opens by feel shows up in franchising's own behavior, where the average multi-unit franchisee runs 5 locations against 4.8 in 2011 (FRANdata), and 82% of franchised QSRs already sit under multi-unit control (FRANdata) precisely because the file is standardized. Diego F. Parra keeps pressing a point operators dislike: the manual is not a handsome PDF, it is the list of file numbers, real timelines measured in your city, and the names of the people who answer. The deliverable is an opening master sheet with columns for filing, owner, actual days from the last location, file number and status.
The regulatory file as a replicable asset, not a to-do list
You verify it when someone else executes it without calling you. The opening is ready when you can display, on the wall and in the folder, eight live and unexpired documents. One: the zoning certificate with the correct commercial use. Two: a current operating license. Three: the health permit with its establishment number. Four: fire-safety sign-off with the date of the last drill. Five: the liquor license, if it applies, with authorized hours printed. Six: food-handling certificates for the kitchen staff. Seven: the waste collection contract and the grease trap log. Eight: the four verified digital listings with menus loaded. Put an expiration date on each one in the general manager's calendar the same day you receive it, because a permit that lapses unnoticed closes the restaurant just as efficiently as one you never filed. That review is yours to run, not the facilitator's. The gap between 92 and 165 days has nothing to do with how fast the authority moves.
Where the opening is actually won?
It is SEQUENCE. Zoning conditions everything else, so starting it before the lease returns forty to sixty calendar days, and those days are worth exactly one month of rent plus one month of lost sales.
A group billing USD 85,000 per unit per month leaves USD 170,000 on the table over a filing that costs under USD 300 to start. The local digital file is infrastructure, not marketing. Google needs 5 to 21 days to verify a new listing, and the Maps proximity algorithm weighs prominence alongside distance and relevance: a listing created three months before opening arrives with history, while the one created that morning competes from the bottom of the local pack. Same story on the apps — Rappi, Uber Eats and DiDi Food onboarding runs 10 to 25 business days depending on city and tax verification. I got this wrong for years. I treated the liquor permit as a closing item, something you handle once the kitchen is built.
Where the opening is actually won — in practice?
It works the other way around. In many jurisdictions the permit hangs on minimum distances to schools, temples and hospitals, so the site decision already settled the matter, and you are merely discovering late a restriction that predated your first visit.
A beverage program typically carries 22% to 30% of contribution margin in a full-service restaurant; losing it is not a setback, it is a different business model. On menus: if your file includes a digital menu, KEEP the printed one. The QR handles delivery, accessibility, price changes and analytics on what guests actually browse; the printed menu governs service pace, dish narrative and suggestive selling, which is where average check lives. Both, each with its role. Operations that dropped print to save on printing have given back 6 to 9 points of average check the following quarter. For a group leader, the metric that organizes all of this is MTIE — the margin of tolerance to expansion surprises — which is simply how many months of rent and payroll your cash can absorb if the opening slips.
Where the opening is actually won — key points
If your MTIE is two months and your average regulatory calendar is five, permits are not your problem: capital structure is, and no expediter fixes that over the phone.
Before and after, criterion by criterion
What the improvised opening bringsBefore
- Lease signed before anyone reads the parcel's zoning, its hour limits and its alcohol restrictions.
- Filings in sequence — zoning, then health, then fire and safety — stacking calendars instead of overlapping them.
- Construction built on drawings the health inspector questions afterward, with grease traps and staff flows solved wrong.
- Digital presence starting on opening day: unverified listing, zero reviews, zero approved photos.
- No homologated supplier matrix, so first-quarter food cost climbs to 38-41%.
- Every new location repeats the same mistakes, because nobody wrote down the previous ones.
What the Masterestaurant file installsMasterestaurant
- Seven-point due diligence BEFORE signature, zoning certificate in hand and an adaptation budget already closed.
- Critical path in parallel: zoning, health, fire safety and liquor advance simultaneously, each with an owner and a date.
- Drawings reviewed against health code before a single wall comes down, with the practices manual ready for inspection.
- Google Business Profile verified 60 days before opening, 40 photos, hours, menu loaded and three delivery apps in onboarding.
- Matrix of 12 homologated suppliers with closed pricing, holding food cost under 32% from week one.
- A master file that clones itself: the next unit reuses 85% of the paperwork and the replicable operations manual.
Side-by-side comparison
| Improvised opening (before) | Masterestaurant file (after) | |
|---|---|---|
| Days from lease signature to opening | ✕165 days on average, with 2 construction stoppages | ✓92 days, with filings running in parallel from day 1 |
| Rent paid before operating | ✕5.5 months of dead rent, 4.8% of total investment | ✓3.0 months of dead rent, 2.6% of total investment |
| Site due diligence before signing | ✕0 checks: sign first, ask later | ✓7 mandatory checks, alcohol zoning included |
| Verified Google Business Profile | ✕Created on opening day, with 0 reviews | ✓Verified 60 days early, 40+ photos and hours loaded |
| Delivery listings live at opening | ✕1 app, with 21 days of onboarding after opening | ✓3 apps live on day 1, menu and photos approved |
| Month-one sales vs. stabilized month six | ✕38% of run rate: the discovery curve starts at zero | ✓71% of run rate: local demand already knew the date |
| Cost of regulatory rework | ✕USD 9,400 in construction redone after a health finding | ✓USD 1,200 in minor fixes already budgeted |
| Replicability toward the next unit | ✕0% documented: every opening starts blank | ✓85% of the file reusable in the next unit |
The figures that settle the decision
“We signed the third location with the lesson learned: zoning certificate requested before the contract, and the Google Business Profile opened the day we got the keys, with the space still in shell condition. We reached opening day with 63 reviews redirected from our other two units through a geotargeted campaign and with all three apps live. Month one closed at USD 61,000 against the USD 23,000 our second location did in its first month, and food cost started at 30.4% because the supplier matrix was already signed. It cost us USD 1,900 in expediting fees and saved two months of rent.”
The file in five steps, with a deliverable and a checkpoint
Do not start without four things: incorporation papers and tax ID of the operating entity, a financial model with break-even calculated, an UNSIGNED letter of intent for the site, and USD 3,000 to USD 6,000 reserved for expediting and fees. Deliverable: a shared folder holding those four documents with a named owner for each. Control figure: if your MTIE — the months of rent and payroll your cash survives with no revenue — sits below 4, do not sign yet. Common error here: using the existing location's entity to open the new one, which contaminates both tax histories and complicates any future sale of a unit. Verification: your accountant confirms in writing that the corporate structure carries an additional unit without changing tax regime.
Request the zoning or land-use certificate for the exact parcel in writing, then clear seven points: permitted use, maximum occupancy, authorized hours, on-premise alcohol sales, minimum distance to schools and hospitals, required parking spaces and allowed exhaust discharge. Deliverable: the official PDF plus a one-page memo answering all seven YES or NO. Control figure: typical cost between USD 60 and USD 300, turnaround 5 to 15 business days. Common error: accepting the landlord's word that 'there has always been a restaurant here,' which proves nothing once the prior use has lapsed. Checkpoint: if a single point comes back NO and it touches the beverage program or the exhaust hood, renegotiate the price or walk away from the site.
Build a board with five filings moving at once: business license, health notice or license, fire-and-safety clearance, liquor permit and exterior signage. Each one gets an owner, a target date and its enabling document. Deliverable: a single-page Gantt where no filing waits on another unless a real dependency exists, and the health filing starts when construction hits 60%. Control figure: 45 to 75 calendar days for the full package in most Latin American cities; if an expediter promises 20, be suspicious. Common error: waiting for construction to finish before requesting health inspection, which adds three or four dead weeks. Weekly checkpoint: no filing sits more than 10 days without a logged movement.
Submit the kitchen layout for health-code review BEFORE construction: separated clean and dirty flows, grease trap sized to occupancy, hood height and material, three-compartment sink, hands-free handwash station and elevated storage. Deliverable: drawings stamped by the responsible sanitary engineer plus the grease-trap sizing calculation. Control figure: an average health finding costs USD 2,500 to USD 9,400 in redone construction, against the USD 400 to USD 900 a prior review costs. Common error: sizing the trap by square meters instead of liters per cover. Verification: run a private pre-inspection with an accredited verifier at 80% construction, then correct against the findings list.
The day you receive keys, create the Google Business Profile with the exact address, correct primary category and hours, then start video or postcard verification. Upload at least 40 photos, load the menu and publish the opening date. In parallel, begin onboarding with Rappi, Uber Eats and DiDi Food, which require tax registration, a bank account under the operating entity and approved product photography. Deliverable: verified listing, three delivery listings in active status and a geotargeted 3 km campaign ready to switch on. Control figure: Google verification 5 to 21 days; app onboarding 10 to 25 business days. Common error: registering the listing at the corporate address, a mismatch that forces reverification later and erases accumulated history.
Before the first purchase order, close written pricing with twelve key suppliers and calculate theoretical food cost dish by dish; the ceiling is 32%, and if your menu average lands at 35% you have a menu-engineering problem rather than a purchasing one. Deliverable: a supplier matrix with price, lead time and substitute, plus costed recipes for the ten dishes that will carry 70% of sales. Control figure: weighted theoretical food cost below 32%, and actual-versus-theoretical variance under 2 points at month one. Common error: buying month one at list price with no negotiated volume and then normalizing that cost. Checkpoint: run a full inventory on day 30 and compare; a variance above 3 points points to portioning or waste, not to the supplier.
And with AI?
Standardize and replicate processes to scale and franchise with control. Diego F. Parra is an expert in AI applied to restaurants.
Free tools to apply this now
Ecosystem tools for this opening
The regulatory file organizes the calendar, but the decision to open is made with numbers. These three Masterestaurant tools answer the three questions a restaurant group leader needs settled before signing: whether the model closes, whether cash survives a delay, and whether the unit is replicable.
Use them in that order. Model first, cash flow second, scaling plan only after — because a beautiful model sitting on short cash is the most expensive way to learn what MTIE means.
Questions that always come up
What are the requirements and permits to open a restaurant in 2026?
What are the requirements and permits to open a restaurant in 2026?
Five blocks: land-use or zoning certificate for the parcel, municipal business license, health notice or license with a good-practices manual, fire-and-safety clearance, and a liquor permit if you sell alcohol. Add corporate tax registration, employer registration with social security, and an exterior signage license. In 2026 the local digital package — verified Google Business Profile and delivery listings — behaves as an operating requirement even though no authority demands it.
How long does a restaurant health permit take?
How long does a restaurant health permit take?
Between 15 and 45 business days from filing, depending on whether your jurisdiction works by notice of operation — nearly immediate, with inspection afterward — or by prior license with a scheduled visit. The timeline blows up when inspection finds construction issues: an undersized grease trap and crossed clean-dirty flows are the two most frequent, and each adds three or four weeks of correction plus reinspection.
Can I sign the lease before holding the permits?
Can I sign the lease before holding the permits?
You can, but negotiate two clauses first: a rent-free period covering construction and filings, three to four months, and a termination condition letting you exit without penalty if the authority denies land use or the liquor permit. Without those clauses you are financing the landlord's regulatory risk. And request the zoning certificate before signing even if it costs two weeks of waiting; it is the cheapest check in the whole due diligence.
Should I open with a QR menu only and skip printing costs?
Should I open with a QR menu only and skip printing costs?
No. Keep both, each with its job. The printed menu governs service pace, dish narrative and suggestive selling, which is where average check gets built; the QR handles delivery, accessibility, price changes without reprinting, and analytics on what guests browse. Dropping print to save a few hundred dollars a year typically costs 6 to 9 points of average check the following quarter.
What carries over from one opening to the next inside a restaurant group?
What carries over from one opening to the next inside a restaurant group?
Roughly 85% of the file: good-practices manual, standard kitchen drawings, homologated supplier matrix, costed recipes, application templates, pre-inspection checklist and the delivery-app onboarding script. What never carries over is site due diligence, because zoning, distances and occupancy change by exact address. That is precisely the point of a replicable operations manual: it frees the team's time for what cannot be repeated.
Sector data 2026 (official sources)
Verifiable industry benchmarks from official, non-commercial sources (government, industry associations, market research) - not competitors.
| Metric | Benchmark 2026 | Source |
|---|---|---|
| Empleo de las franquicias en EE.UU. proyectado para 2026 | cerca de 8,9 millones de empleos (+150.000, +1,8%) | FRANdata / IFA — Franchising Economic Outlook 2026 |
| Volumen medio por unidad (AUV) de Jack in the Box | 1.913.335 USD (12 meses a sep. 2025) | Jack in the Box — FDD 2025 |
| Volumen medio por unidad (AUV) de Chick-fil-A | cerca de 7,5 millones USD | Restaurant Business — AUV ranking 2025 |
| Volumen medio por unidad (AUV) de Raising Cane's | cerca de 6,5 millones USD | Restaurant Business — AUV ranking 2025 |
| Volumen medio por unidad (AUV) de Wingstop | 2,13 millones USD (FDD 2025) | Restaurant Business / Wingstop FDD 2025 |
| Inversión inicial total de una franquicia Taco Bell (FDD) | 1.584.750 a 3.980.200 USD | Taco Bell — FDD Item 7 |
Related content
Grow your restaurant with the Masterestaurant method
Applied in +8.400 restaurants across 43 countries.
