A Retail Permitting Playbook, Built Jurisdiction by Jurisdiction

The approval you cannot buy your way out of

Before anyone frames a wall, a jurisdiction has to say yes, and every store project waits on that. The building permit sits at the front of the critical path, it depends on a reviewer you do not employ working through a backlog you have no visibility into, and until it lands the crew you booked is standing down.

The delay shows up in public filings. On Chipotle's late-2023 earnings call, CFO Jack Hartung told investors that a restaurant build which used to take 15 to 16 months had stretched to 21 to 22, and he named permitting, inspection, and utility delays among the reasons. A chain that opens hundreds of units a year, off a standard prototype and with a practiced team, still lost the better part of half a year to the approval track. For an operator without that muscle, the number runs higher and harder to predict.

One country, thousands of separate processes

The United States has more than 19,000 incorporated municipalities, plus counties and special districts, and each one sets a code edition, local amendments, forms and fees, a portal or paper intake, and a path to a certificate of occupancy of its own. No national registry ties them together. A retailer opening across 30 states runs somewhere between 50 and 500 separate permitting processes at once, each with its own reviewer and queue, and its own idea of what counts as a complete application.

That work does not scale the way the rest of the program does. A prototype travels from market to market, and so does a national GC agreement. The permitting knowledge for Cook County does nothing for you in Maricopa, and the person who earned it may be on another project by the time you are back.

What the playbook holds

The teams that run permitting well treat each jurisdiction as a record they build once and keep. The first project in a city is expensive to permit, because you are learning the authority from scratch. The second should cost less, and the fifth less again, but only when what you learned the first time was written down somewhere the next project manager can find it.

A per-jurisdiction playbook holds the things that do not change between your projects in that city.

• The authority having jurisdiction for each permit type, since building, fire, health, sign, and utility approvals often answer to different offices
• The adopted code edition and the local amendments that depart from the model code
• How the jurisdiction takes an application, what it charges, and what it treats as a complete submission
• The realistic review-cycle length and how many comment rounds are normal, so the schedule reflects the city instead of a hope
• The certificate-of-occupancy procedure and the inspections that gate it
• The approvals that tend to surface late here, from fire-marshal sub-permits to health and alcohol licensing

The clock is set locally, and it moves

How long a permit takes belongs to the jurisdiction rather than your project, and it does not hold still. Dallas overhauled its permitting department and cut the median wait for a new commercial permit from 276 days in 2023 to 189 in 2024, a 32% improvement in a single year. A playbook still carrying the old number would have padded every Dallas schedule by three months it no longer needed.

So the figure you carry has to be dated. Refresh it when the city changes. Commit an opening date against last year's queue and you have set it on stale information, whichever way the queue has moved since.

The permits that surface after the building permit

Teams track the building permit closely, and the approvals that wreck a schedule are often the ones that never made it onto the list. A sign permit runs on its own track and its own clock. In some cities it outlasts the building permit it accompanies. Fire-marshal sub-permits, health-department sign-off on anything with a kitchen, alcohol licensing, and utility service each move on their own, and a condition buried in the certificate of occupancy can hold the doors shut after the space is finished.

In New York, before the Department of Buildings will issue a work permit, the project's applicant of record has to file a TR-1 naming every special and progress inspection the scope requires. Skip it and there is no permit, however ready the drawings are. Most jurisdictions have a version of that one form or step that stops everything, and the playbook is where you record it so the next project does not find it the hard way.

Build it once, and the program compounds

Whether a team dreads permitting or manages it comes down to where the knowledge lives. When the authority contacts, the code editions, the real cycle times, and the hidden-permit list sit in the same place as the schedule and the budget, permitting stops being a fresh scramble on every project and becomes a step the team has run before. RolloutIQ models permits as first-class records on a location rather than files buried in a project folder, which lets the second project in a jurisdiction start from what the first one learned.

You never fully control permitting, but a written jurisdiction record means the second project in a city inherits what the first one paid to learn.