Opening a restaurant is a two-part problem. The first part is a construction project. The second part is a team formation. Both have to land on the same day. The last 90 days is where the two parts converge, and where most openings either come together or fall apart.
I have run this window twelve times. The rhythm is consistent enough that I can tell you week by week what should be happening, what the failure signals look like, and what to do when the plan starts sliding. Here is the calendar.
T-90 to T-75: the plan meets the space
At T-90 the buildout is roughly 50 percent done. The bones of the space are visible. This is the first moment where you can walk the floor and see the actual flow. Almost every time, something in the design does not match how the team will actually work. A prep station in the wrong spot. An expo pass that is too short. A dry-storage door that opens into a traffic lane.
Catch these now. Every change gets more expensive the closer you get to opening. A prep-station move at T-90 costs a few thousand and a week. The same move at T-30 costs ten times that and can push your opening.
Meanwhile the general manager, hired at T-90 or earlier, is not at the site yet. They are at your best existing unit, learning your SOPs from your best operator. Four full weeks. This is the investment that pays for itself twenty times over in the first month. A general manager who has never run a shift on your brand cannot lead an opening.
T-75 to T-60: the culinary lead and the training crew
Culinary lead is hired by T-75. If you are running a brand with a strong menu identity, this hire is being pulled from an existing unit. If it is a market where you need someone with local network, they are hired externally and sent to an existing unit for two weeks of brand immersion before they land at the new site.
The training crew gets named now. Three to five people, pulled from your best existing units, who will come in from T-30 to T-14 and be the operating spine of the new team. Get their names on the schedule. Get their travel booked. Get their pay for the assignment agreed on in writing. Volunteers work worse than assigned people because volunteers change their minds in week three of the assignment.
Fig. 1 · Headcount in the space, week by week.
T-60 to T-45: the equipment arrives
Equipment install is the phase where everything you ordered six months ago shows up and either works or does not. Somewhere between 5 and 15 percent of what arrives will have a problem. A wrong voltage. A missing part. Damage in shipping. A gasket that never got included. Build a punch list from the moment the first crate is opened and hunt these down.
Health department pre-inspection happens in this window if your jurisdiction allows it. A pre-inspection is the single highest ROI thirty minutes you will spend in the whole opening. The inspector walks the space with you before your final inspection and tells you what will fail. You fix those things. Then the final passes on the first try. Every jurisdiction I have opened in offers this. Not every operator asks.
T-45 to T-30: the first local hires
Local hiring for the front of house and back of house begins at T-45. Offers go out at T-30. Paid training in the shell starts at T-25.
Interview differently for an opening than you would for an existing unit. In an existing unit you can afford to hire someone who will learn on the job over 90 days, because the team around them is strong. In an opening team, everyone is a new hire. There is no strong team around anyone. So you overhire for people who have opened before, for people who have worked at brands you respect, and for people who can teach as they learn. The best line cook at soft open is often the one who was showing the other two how to plate a dish they just learned themselves.
T-30 to T-20: everyone in the space
This is the shift. The building is largely done. The equipment is in and mostly working. The general manager, the culinary lead, the training crew, and the local hires are all on payroll and in the space every day.
The daily rhythm is a stand-up at 9am, three hours of SOP walks, lunch, three hours of menu cook and plate practice, a stand-down at 6pm. Every day. Same shape. This is where the team learns the space and each other. This is also where the training crew starts to disappear into the background and the general manager starts running the floor.
By T-20 the general manager should be running the stand-up. If the training crew is still running it, the hand-off is going to fail and week three is going to hurt.
T-20 to T-10: the punch list
The punch list is a running document with every unfinished thing in the space. It starts around T-30 with maybe 200 items. By T-20 it should be under 100. By T-10, under 30. By T-3, under 10.
The punch list is where opening days go to die. Every unresolved item is a small fire the team will have to fight in week one instead of serving guests. A soap dispenser that keeps clogging. A dishwasher that leaks. A hostess-stand printer that has to be reset every four hours. Individually, none of them are big. Together they consume the manager's attention and pull them off the floor.
The rule I use: no item stays on the punch list two days in a row without an owner and a completion date. If an item has been sitting for a week, escalate it. If it is still there at T-3, decide whether it is a reason to delay opening.
T-10 to T-3: the three soft opens
Three soft opens, escalating in difficulty. Team-only run at T-10. Friends and family at T-6. Full local run at T-3. Each one is a real service with real timing pressure. Each one produces a fix list. The fix list gets worked between soft opens.
The single most important number to watch across all three is ticket time from fire to plate. If soft open one is running 22 minute tickets and soft open three is running 14 minute tickets, you are ready. If soft open three is still running 22 minute tickets, you are not, and you should not open on the scheduled day.
T-3 to T-1: the final walk
Final health inspection. Final fire inspection. Final building sign-off. Health almost always fails somewhere the first time. Have your punch list ready to work overnight.
The general manager walks the space alone at 10pm on T-1. Kitchen. Line. Expo. Front of house. Bathrooms. Storage. If anything is wrong, they fix it or defer opening by a day. This walk is the last chance to catch what everyone else missed because they were looking at the whole and not the pieces.
T-0: opening day
Opening day is a soft ramp. Doors open at normal time. No opening event. No press. Reservations only, at 60 percent of capacity. The general manager runs expo. The chef runs the line. The training crew fills the gaps.
By the end of service, the team has run one full service together. Not a soft open. A real service, at real pace, with real guests paying real money. That is the whole point of opening at 60 percent volume. You get the first real service in a way the team can absorb.
What I would do differently
Three things I have learned over twelve openings that I did not know on the first three:
- The pre-inspection with health. I skipped this on my second opening because the inspector said we did not need one. We failed the real inspection on 14 items and lost three days to remediation. Every opening since then has started with a pre-inspection.
- Paying the training crew a premium for the assignment. On my early openings I paid them at their normal rate and hoped for goodwill. Goodwill runs out in week three. Now they get a documented premium (usually 15 to 25 percent uplift plus travel and per diem), signed in writing, contingent on staying through the hand-off. Attrition dropped to near zero.
- Slowing marketing. On my fourth opening the marketing team ran a launch event with 400 confirmed guests in week one. The team, ready for 200. We did the 400. The reviews from that week followed the unit for a year. Now the marketing plan starts in week three, not week one.
The point
The 90 days before opening is where the plan meets reality. If the plan is real and the team is real and the space is real, the last 90 days is a series of small decisions that all point at the same date. If any of those three are not real, no amount of hustle in the last 90 days will save it.
Run the calendar. Watch the punch list. Do the three soft opens. Open at the volume you can serve. The rest is the machine, and the machine works.