The first week inside a broken operation is the most valuable week you will ever get, and most operators waste it. They walk in with answers instead of questions. They fix the visible thing in front of them on day two and then spend the next 60 days trying to figure out why nothing else moved. The picture never forms because the questions were never asked.

The last full turnaround I ran was at a $30M Michelin-recognized Bay Area group. Three of five locations were bleeding. I came in as Fractional Head of Operations and Turnaround Lead. In eleven months we recovered roughly $4.9M in operating profit across the three underperforming units. Everything downstream of month two ran off the questions I asked in the first five days. I want to write those questions down, because they are more portable than the story is.

Why questions come before conclusions

The temptation in week one is to skip straight to a plan. The board wants a plan. The general managers want a plan. You want a plan, because a plan is what you were hired to produce. Resist it. A plan built off week-one instincts is a plan built off the last operating system you fixed, not this one.

Numbers on a P&L are lagging indicators of an operating system. They tell you what happened. They do not tell you why. The why lives in the walk-in cooler, in the schedule that was built off the last schedule, in the decision the closing manager did not have the authority to make. If you skip the operating layer and go straight to the numbers, you will change something visible and break something invisible. I have done this. It cost me two weeks of clean signal the first time.

Week one is a probe, not a test. Every question you ask is designed to reveal, not to grade. The moment your questions start to feel like an audit, the answers stop being useful.
The four question stacks of week one STACK 1 General Manager Decision rights Real cadence What they own What blocks them STACK 2 Line Staff What breaks first Waste patterns Equipment failures One-day fixes STACK 3 Walk-In Count gap FIFO discipline Buried inventory Prep sheet reality STACK 4 The P&L Line volatility Silent lines Missing lines Peer benchmark gap All four stacks are asked in the same week. Skip any one and the picture will lie.

Fig. 1 · The four categories of week-one questions.

Stack 1: Questions for the general manager

The general manager sit-down happens on day one or day two at each location. I ask for 60 minutes. I take notes on paper. I never bring a laptop to this meeting. The list below is what actually gets asked, in roughly this order.

Decision-rights questions

  • If a truck arrives with a short-shipped case of chicken, who decides whether to accept it or refuse it? If the answer is "I call the area director," you have found a decision right that should live at the unit and does not.
  • If a line cook does not show up for a Friday dinner, what can you do about it without asking permission? The range of legal answers tells you the operating slack the general manager has.
  • Up to what dollar amount can you comp a guest without approval? If the number is $10 in a restaurant with a $75 average check, you have found the reason service recovery is not happening.
  • Can you hire a dishwasher without permission? This one surprises people. In many struggling groups the answer is no, and the average time-to-hire is 21 days for a role that should be filled in 3.

Cadence questions

  • When was the last time you sat down and read a full P&L for this location? A shocking number of general managers will pause here. A pause is data.
  • What number do you look at every morning before you unlock the door? If they cannot name one, there is no morning cadence. If they can name three, they are trying to look at everything, which means they are looking at nothing.
  • When do you next meet with your area director, and what is the agenda? A standing meeting with no standing agenda is not a meeting. It is a check-in.

Ownership questions

  • What are you most proud of in this restaurant right now? This softens the room and it also tells you what the general manager thinks is working, which is often something the P&L will disagree with.
  • What is the one thing you would fix if I gave you Monday off and full authority? This is the most productive question in the entire interview. Write the answer down verbatim. You will come back to it in week six.

Stack 2: Questions for line staff

Line staff know the operating system better than anyone in the building. They will tell you the truth if you ask about the work rather than about the people. I do these conversations informally, standing near where they are already working, five to ten minutes at a time. Never in an office. Never with the general manager present.

The four questions

  1. What breaks first when we get slammed? A cook will tell you within thirty seconds. A dishwasher will tell you within ten. The answer is almost always a piece of equipment, a station design, or a prep decision made too early in the day.
  2. What do you throw away every night? Not "how much waste do we have." Ask what specifically hits the bin. Line cooks know exactly. They will name the two prep items that always overshoot and the garnish nobody uses.
  3. What was different about the last general manager who worked here? This is a lookback question that surfaces cultural signal without asking anyone to critique current leadership. The answers are useful whether the last general manager was better or worse.
  4. What would you fix tomorrow if it were up to you? Same question I ask the general manager, and often the answers rhyme. When they rhyme, you have found something real.
Ask about the work, not about the people. Line staff will tell you the truth about the operating system. They will not tell you the truth about their coworkers, and you should not be asking them to.

Stack 3: Questions of the walk-in

The walk-in is the operating system in physical form. You cannot fake a walk-in. Every count discipline problem, every FIFO drift, every prep decision made off vibes rather than off numbers, it all lives on the shelves.

Day three or four of week one, I walk into every walk-in cooler and every dry storage at every location. Same week. Camera on. I am not counting yet, I am reading. The count comes in week two.

What to look at

  • The gap between what is on the shelf and what the inventory system says is on the shelf. Spot-check three high-value items. The gap will not be zero. The size of the gap is your count discipline in a number.
  • What has been pulled to the front and what has been buried. If the freshest product is in front and the oldest is buried, your FIFO discipline is dead. That single fact predicts your food cost variance more reliably than any P&L line will.
  • The prep sheets stuck to the walk-in wall. Are they from this week, or from last month. Is anyone actually writing on them. A dead prep sheet is a dead prep culture.
  • What is in the reach-in versus what is on the line. Overprepped items on the line tell you the prep cook is guessing at demand. That guess is where the waste is born.

Stack 4: Questions of the P&L

By day four you have enough operational context to read the numbers usefully. Not before. Reading the P&L cold, without the operational picture, produces confident wrong conclusions.

The volatility questions

Pull 90 days of daily P&L, not the monthly rollup. The rollup will lie to you because it averages the volatility away, and the volatility is the story.

  • Which lines swing the most from day to day at the same location? Labor is almost always the biggest swinger. A labor cost that ranges from 26 percent to 42 percent inside one week is not a labor problem. It is a scheduling problem.
  • Which lines never move at all? A line that reads the same number every week is either a beautifully controlled process or it is not being measured. In turnarounds it is the second one.
  • Which lines never appear when they should? Comps and voids should have their own line. Waste should have its own line. If they do not, they are being buried in cost of goods sold and you have no visibility.

The benchmark questions

  • What is the healthy peer benchmark for this concept? If you do not know, find out before you finish week one. You cannot judge a P&L in isolation.
  • Where is the biggest gap versus benchmark? That is your first stabilization target. Not the biggest cost line. The biggest gap.

The Friday-dinner sit

Day five, or day six if you can stretch it, sit at a two-top in the dining room during Friday dinner. Order food. Watch the room. Time the ticket. Watch the expo. Watch how the general manager moves. Watch the moment a table takes eight minutes for a drink refill.

This is the moment the numbers become people and the people become a picture. Every week-one diagnostic I have run has had one Friday-dinner moment where a piece of the puzzle clicked. In one location it was watching the closing manager stand next to a busser trying to solve a comp problem that they clearly did not have authority to solve. That was the shape of the whole operating system, and I saw it in ninety seconds from a corner table.

What I would not do in week one

Three things I have learned to leave alone until the diagnostic is done:

  1. Do not rebuild the schedule. You will want to. The schedule is almost always broken and it is right in front of you. Wait. If you rebuild it in week one, you contaminate the labor variance signal for weeks and you cannot tell what was the system versus what was your change.
  2. Do not fire anyone. Firing in week one teaches the entire region to hide problems from you. Hidden problems are the ones you cannot fix. The rare exception is a safety issue or an integrity issue, and those are self-evident.
  3. Do not announce a plan. When someone asks, "What are you going to change," the correct answer is, "I do not know yet. I am learning." This answer buys you diagnostic credibility that a confident wrong answer will burn in three days.

The point

The first-week diagnostic is a discipline, not a personality. Anyone can walk into a broken restaurant on day one. Very few operators can walk in and resist the urge to fix the first visible thing they see. The ones who can are the ones who end up with a real picture by day seven, and everything downstream of that picture is faster, cleaner, and holds better after they leave.

Ask decision-rights questions of the general manager. Ask friction questions of the line staff. Ask count-discipline questions of the walk-in. Ask volatility questions of the P&L. Sit at a Friday-dinner table and let the operating system show you what it is under load. Write it all down in a notebook. Type it up at the end of each day into a running document that becomes your operating hypothesis for week two.

The plan comes in week two. Week one is for the questions. If you get the questions right, the plan almost writes itself.