The 90-Day Check-In After Handoff

The handoff itself, I have written about elsewhere. You spent a year rebuilding an operation, you installed the operating rhythm that was supposed to hold, you trained a general manager to run it, and you walked out on a Friday afternoon feeling like the job was done. The last thing you told the owner was that you would be back in 90 days to check.

Now it is 90 days later. And this visit, the one nobody puts on the calendar with any real weight, is the single highest-return day in the whole engagement. What you find on the 90-day visit tells you whether the turnaround was a real transfer of capability or a temporary rental of your attention. What you do with what you find determines whether the numbers you moved stay moved.

I have done this visit at every turnaround engagement I have run in the last decade, including the three-location Bay Area group I keep referencing. What follows is the exact structure. It is boring on purpose. Boring is what makes it repeatable, and repeatable is what makes gains hold.

Why 90 days is the right window

Not 30, not 60, not 180. The reason is not superstition. It is the natural cadence of decay.

At 30 days after handoff, the general manager and the team are still running on the muscle memory you built. The standing meeting still runs because it just ran last week. The dashboard still gets opened because the habit is not old enough to be broken. Nothing has slipped yet. There is nothing to look at.

At 60 days, drift is starting, but it is inside the normal variance of any operating month. A missed meeting, a dashboard that got skipped one morning, a labor variance that ran hot for a week. You cannot tell the difference between drift and noise yet.

At 90 days, three full monthly cycles have run. The patterns are visible. What held has held. What slipped has slipped. And the general manager, importantly, has now spent a full quarter with no external accountability. Everything you see is the operating system running on its own.

Wait until 180 days and two things go wrong. The drift becomes the new baseline in everyone's mind, and correcting it now feels like a change instead of a return. And the general manager has now spent six months without a check-in, which sends the wrong signal: that the handoff was actually goodbye.

Ninety days is the earliest point at which the operating system is running on its own, and the latest point at which drift is still cheap to correct. Do not miss the window.

What to do before you walk in

Half of a good check-in happens before you get to the site. You want to walk in with a hypothesis, so that the onsite time is spent testing what you already suspect rather than starting from zero.

Pull the last 13 weeks of daily P&L

Same view you built during the original engagement. Read it the same way. Where is labor variance now versus at handoff. Where is food cost. Where is comps and voids. Any line that has moved more than two points from the handoff baseline is your first onsite question.

Ask for the last 12 weeks of standing meeting notes

If they exist, read them. If they do not exist, you already know something important. The meeting either drifted or died. Do not send an email asking why. That question happens in person.

Ask the CFO or bookkeeper for the current AP aging

This is the single fastest read on whether cash discipline is still in place. If the aging report has crept over 45 days average when handoff was at 28, something on the revenue side or something on the vendor-management side has moved.

Check whether the dashboards are still being opened

Most modern operations dashboards have simple usage analytics. If the general manager opened the labor dashboard 62 times in the first 30 days post-handoff and 8 times in the last 30 days, that is your leading indicator. The tool is not being used, so the number the tool tracks is not being managed.

The two-day onsite structure

The visit itself is two days. Anything less is too shallow. Anything more and you start becoming operational again, which is the wrong role.

Half day of silent observation

Same shift you observed on day one of the original engagement. Friday dinner if that is where the operating system flexes. Sunday brunch if that is the pressure test. You are watching, not coaching. Take notes on your phone or a small notebook. Do not answer questions from staff unless asked directly, and if you are asked, redirect to the general manager.

Sit in on the weekly operating meeting as a guest

Not the head of the table. A chair against the wall if there is one. Do not contribute unless the general manager pulls you in. What you are watching for: does the standing agenda still exist. Do the general manager and their area director still walk through labor variance, food cost, top three items, and one thing to fix this week. Is the meeting still 45 minutes or has it drifted to 90. Are numbers being reviewed, or is the meeting a status update.

Three private conversations, in this order

Start with the general manager. Three questions. What is holding better than you expected? What has been harder than you expected? What do you wish we had installed before I left that we did not? Listen more than you talk. The third question is where the gold is.

Then, one line-level employee who was there during the turnaround. A closing cook, a bartender, a shift lead. One question. Is the place running better or worse than 90 days ago, and why do you think so? Do not lead them. Their answer will tell you what the general manager will not.

Finally, the owner or CEO. This is the setup for the memo, not the memo itself. Preview what you saw. What held, what slipped, what you would do. Let them react. Some of what they react to will change what you write.

What holds, what slips

After doing this visit maybe 30 times across multiple engagements, the pattern is startlingly consistent. Some things hold. Some things slip. Almost always the same things.

Notice the shape of the pattern. The habits that hold have a physical artifact attached. A count sheet. A waste log. A line-check clipboard. A vendor invoice with a memo line. You touch a thing, and the thing reminds you to do the habit.

The habits that slip live only in calendars or dashboards. A standing meeting is just a recurring calendar event that anyone with authority can cancel. A dashboard is just a URL that a busy operator can decide not to open today. The absence of a physical artifact is the drift risk. If you had one lesson to take from years of check-ins, this would be it: attach every operating habit to something you can physically hold or physically walk past. The habits that live in software slip. The habits attached to paper on a clipboard hold.

What you fix, what you leave, what you flag

Here is where operators most often get the visit wrong. You will see problems. Some of them will be problems you spent 60 hours fixing in month three of the turnaround. The urge to grab the clipboard and fix it in real time will be enormous. Do not do it.

Fix nothing yourself onsite. Every problem you fix during the check-in teaches the team that you will still catch things. That teaches them not to catch things themselves. This is the single fastest way to undo a year of capability transfer.

Leave the things that look broken but are actually a workable new equilibrium. Sometimes the general manager has changed something you built, and their change is not as elegant as yours was, but it works for them and their team runs it. Leave it. Your job on the check-in is not to protect your design. It is to protect the operating capability.

Flag the three things that most matter to the owner. Not fifteen. Three. Two things that need attention in the next 30 days, one thing that needs a decision from the owner about direction. The check-in memo is the deliverable, and its whole job is to sharpen the owner's attention on the right three things. For more on framing the end of the engagement itself, see Handing Back a Fixed Operation So the Gains Hold.

The check-in memo

One page. Five sections. Sent within 48 hours of the visit.

1. What held. Two or three sentences. The specific rituals or numbers that are still in place. Name them.
2. What slipped. Two or three sentences. Specific, with a number where possible. Not "the meeting is drifting" but "the weekly meeting has grown from 45 to 82 minutes and now covers status rather than variance."
3. What I would fix now. Two or three specific recommendations, sized so the general manager can execute them without additional resources. This is advice, not instruction.
4. What I would leave. One or two things that look broken but are working. This section prevents the owner from over-correcting after they read the memo.
5. What I am flagging for you. One decision that needs to happen at the owner level. Usually a capital allocation, a headcount, or a strategic direction question.

What the visit actually costs and what it buys

Two days onsite. Two half-days of prep. One 48-hour turnaround on the memo. Maybe 20 hours of work total. Priced in advance as part of the original engagement scope, so nobody has to negotiate whether it happens.

What it buys is the difference between a turnaround that held and a turnaround that decayed. Across the engagements where I have consistently done the 90-day check-in, gains have held into year two roughly 80 percent of the time. Across the engagements where I skipped it or delayed it past 120 days, the number is closer to 50 percent. That gap is the visit, and it is the reason I now write it into every scope of work at the outset.

You did not spend a year rebuilding an operation so you could watch it drift back over the next quarter. The 90-day visit is the discipline that connects the work you did to the outcome you promised. Put it on the calendar the day you sign the engagement. Show up on the day you promised. Write the memo inside 48 hours. Then trust the operator you built to run the thing they now own.