the restaurant soft opening checklist: what to test before you open.

the short answer.

A restaurant soft opening is a controlled test of your kitchen, bar, front of house and technology under real conditions, run before you take public money. Test seven things: ticket times by station, the POS build and modifier logic, server flow from table to pass, bar speed on your slowest drink, dish pit and bussing capacity, payment and closeout, and the door. Everything else is decoration.

Run it in sequence, not all at once. One night at a third capacity, one at two thirds, one at full. Invite people who will tell you the truth. Write down every failure as it happens, because by the third hour nobody remembers what broke at seven.

what is a restaurant soft opening actually for?

It is not a thank-you party for the people who helped you build. It is a stress test with forgiving guests. The point is to break things in a room where breaking things costs you a comped plate instead of a review.

The failures you are hunting are the ones that only appear under load. A ticket printer that works fine with four tables jams at twenty. A pass with room for six plates backs up when the kitchen fires three tables at once. An expo who knows the menu cold still cannot read the rail when it holds fourteen tickets. None of this shows up in a tasting. It shows up when the room is full and the clock is running.

So design the nights to create load, not to create goodwill. If every guest arrives at six and nobody ever waits, you have learned nothing except that your food tastes good. You already knew that.

the week before: what has to be true.

Walk the building with a legal pad before you invite anyone. Health permit posted. TABC in hand if you pour. Hood certified, ansul tagged, extinguishers current. Walk-in holding temperature with the door opening every ninety seconds, not with the door shut overnight. Hot water at the dish pit at the temperature the sticker says.

Then the quieter things. Does the music system reach the patio at a level that does not drown the bar. Do the restrooms have a lock that latches. Is there a hook on the back of the stall door, because the absence of one is the single most common complaint nobody writes down. Does the host stand have a pen that works and a view of the front door at the same time.

Build the POS last and build it twice. Every item, every modifier, every routing rule. Then ring a fake check for the hardest order on the menu, the one with three substitutions and a split payment, and watch where the ticket prints. Half of opening week chaos is a modifier that fires to the wrong station. We fix more opening-week problems in the POS build than on the line, which is why systems + ops work starts in the menu tree and not in a binder.

night one: a third of the room, a quarter of the menu.

Seat forty percent of your tables at most. Run a cut menu, maybe eight items chosen to touch every station. Charge nothing or charge something nominal. Tell guests plainly that the kitchen will be slow and the point is to find out how slow.

Put one person on a clock with nothing else to do. They time every ticket from fire to pass and write the number on a sheet with the table and the station. That sheet is the most useful document you will produce all week. It tells you which station is your bottleneck, and the bottleneck is never where the chef thinks it is.

Watch the floor, not the food. Where do servers stop walking. Where do two people meet and have to turn sideways. How many steps from the server well to table twelve, and how many times a night does that trip happen. A kitchen that plates beautifully still loses if the runner has to cross the bar line to deliver it. The room and the line are one machine, which is the argument we make in how commercial kitchen design shapes what guests feel.

End the night with a sit-down while everyone is still standing in their aprons. Twenty minutes, every position speaks, one person writes. Do not solve anything yet. Just collect.

night two of the restaurant soft opening: push it until something breaks.

Now you add load. Two thirds of the room, full menu, and here is the part most operators skip: stack the arrivals. Seat half your covers in a twenty-minute window. Real service does that to you whether you plan for it or not, and you want to meet it with friends in the chairs.

This is the night you learn your real capacity. Not the number of seats. The number of covers you can turn with the hands you have and the equipment you own before quality drops. A four-burner range and one fryer sets a ceiling no amount of hustle clears. Better to know the ceiling now and write a smaller menu than to discover it on a Thursday in week two with a full book.

Test the bar separately and on purpose. Order ten of your slowest cocktail at once and time it. If it takes four minutes a drink and a table of six orders them, the bar is down for twenty-four minutes while tickets pile up. That is a menu decision, not a bartender problem. Batch it, simplify it, or price it like the labor it costs.

Run a closeout. The full one. Count the drawer, run the reports, do the tip pool math, break down the line, and see what time the last person actually leaves. Labor on the schedule and labor on the clock are different numbers, and the gap shows up here first.

what do you do with everything you found?

Sort every failure into three buckets before you leave the building: fix tonight, fix before opening, live with it for now. Most operators try to fix all of it and fix none of it. A ticket printer in the wrong place is a tonight fix. A dish pit that cannot keep up at full covers might be a six-month equipment conversation, and in the meantime it is a staffing decision.

Then change the menu. This is the hard one. If one dish produced a third of your long tickets and a quarter of your mistakes, it comes off, or it gets rebuilt with prep that moves to the morning. Nobody wants to cut a dish they love three days before opening. Everyone who does it is glad they did. A plate that costs you eight minutes on the pass costs more than it makes, and the arithmetic is the same whether the plate is $14 or $34.

Write the fixes down as standards, not as notes. The runner order for a four-top. The two-minute drink. Where the mise sits at the sauté station. People remember a drill, they forget a conversation, and in two weeks you will have hired three more people who were not in the room.

Last, decide what opening week looks like. Shorter hours, a tighter menu, fewer tables than you own. Hold capacity back on purpose for the first ten days. The room that runs clean at seventy percent builds a reputation the room that drowns at a hundred never recovers.

who should be in the chairs for a restaurant soft opening?

Not only friends and family. They love you, and love makes a terrible test instrument. They will wait forty minutes and say it was great, and you will learn nothing.

Mix the room. A third people who know food and will tell you the truth. A third neighbors who have never heard of you and will behave like actual guests, ordering the thing you did not expect and asking for substitutions. A third industry people from other rooms, because they notice the dish pit and the restroom and the thing that is about to break.

Give them a one-page card at the table with four questions and a pen. How long did you wait for your first drink. What did you order and would you order it again. What did you notice that we did not. Would you come back and pay for this. Four questions get answered. Twelve get ignored.

the questions operators ask.

How long should a restaurant soft opening last?

Three to five service nights is enough for most independent rooms. Fewer than three and you never reach the load that exposes real problems. More than five and you are burning payroll on food you are not charging for, and the team gets tired before opening day. Space the nights so you have a day between to actually fix what you found.

Should we charge guests during a soft opening?

Charge something on at least one night. Free food changes behavior. People order less, complain less, and wait patiently, which is exactly what you do not want to measure. A reduced-price night also forces you to run the full payment and closeout process, which is where plenty of technology problems surface.

What is the difference between a soft opening and a friends and family night?

A friends and family night is a dress rehearsal with a kind audience, usually one service and often free. A soft opening is a sequence of nights with increasing load, a real payment process, and a written record of what failed. You can do both. Run the friends and family night first, then push harder on the nights that follow.

What should we do if the kitchen cannot keep up during the soft opening?

Cut the menu before you add staff. An extra cook on a four-burner range does not create burners. Identify the two or three items producing the longest tickets, move their prep to the morning, rebuild them, or take them off. Then reopen the question of staffing once the menu matches the equipment you actually own.