The Rescue Job

Deciding When to Save a Location and When to Close It

The hardest call in a multi-unit turnaround is the one you keep deferring. Here is the framework that forces the decision, and the emotional resistance that keeps operators from making it.

Kiyong (Eric) Lee·Operations Executive·August 2026·9 min read

The short version

The close-or-save decision has three axes: unit contribution, lease overhang, and brand impact on the rest of the group.
Give the unit 4 to 6 months of actual turnaround work, not calendar time. If contribution has not moved, the site is telling you something.
Emotional resistance is a real variable. Name it, put a decision date on the calendar, and bring in an outside voice.
Contribution positive and absorption negative is a keep. Contribution negative is a close.
A bad location costs more than its losses. It costs brand equity that the other locations have to earn back.

In every multi-unit turnaround I have worked on, there is one location that becomes the argument. Everyone knows the numbers. Everyone can see the trend line. And nobody wants to be the person who says out loud that this one is not coming back.

The close-or-save decision is the hardest call in the whole rescue job. It is hard because it looks like a financial decision and behaves like an emotional one. It is hard because the answer is almost always in front of you but the courage to name it is not. And it is hard because the cost of getting it wrong compounds. Every month you keep a location that should be closed, you are burning capital, brand equity, and the trust of the general managers at the healthier units who are watching.

This is the framework I use to force the decision. It has three parts. All three matter. Only one of them is about the money.

The three axes

Every close-or-save call sits at the intersection of three things: the unit's own economics, the time you have on the lease, and the effect this location has on everything else in the group.

Fig. 1 · Three axes. The math is simpler than the emotion.

Axis 1: Contribution versus absorption

The first question is the cleanest, and it is the one operators most often confuse. There are two versions of "does this unit make money" and they mean different things.

Contribution

Contribution is what the unit produces after its own direct costs: labor, food, unit-level rent, utilities, unit-level management, unit-level supplies. If contribution is positive, the unit is generating cash the group can use, even if that cash does not cover a fair share of corporate overhead. A contribution-positive unit is worth keeping open unless something else is broken.

Absorption

Absorption is contribution minus the unit's fair share of corporate overhead: shared marketing, corporate salaries, back-office systems, the accountant, the insurance. A unit can be positive on contribution and negative on absorption. That unit is not a failure. It is a unit that is helping the group cover overhead, and if you close it, the overhead does not go away. It just gets reallocated across fewer units, making them look worse.

The rule

Contribution positive is a keep. Contribution negative is a close. Absorption negative on its own is not a close, and mistaking absorption for the deciding number is how operators close units that were actually helping the group.

The unit is not losing money because it exists. The unit is losing money if it costs more to run than it brings in. Those are different sentences, and the second one is the one that decides.

Axis 2: Lease overhang and the payback window

The lease is your effective payback clock. Every credible turnaround investment has a payback window. If the remaining lease term is shorter than the payback window, the math cannot work, and no amount of operational excellence changes it.

Here is the arithmetic. Suppose a location is losing $12K a month and you believe you can turn it to $8K a month of contribution with $180K of capital investment (kitchen equipment, refresh, some training). That is a $20K per month swing, which pays back the investment in roughly nine months. Add a six-month operational ramp and you need at least 15 months of lease term for the math to work. You should probably want 24 months to make the risk-adjusted math work.

If the remaining lease is 14 months, close. If it is 30 months, save is on the table. If it is 18 months, you have a real conversation about whether to negotiate an extension in exchange for a rent concession, which is the third path operators forget about.

The three lease paths

Ride out the term and close at the end. Sometimes the cheapest option is to let the lease expire naturally, harvest whatever contribution you can in the remaining months, and not renew.
Negotiate a buyout. If the landlord will accept a lump payment less than the remaining lease liability plus expected losses, buyout is usually cheaper than continuing to operate.
Extend and invest. If the site fundamentals are strong and only the lease term is short, extending the lease in exchange for turnaround capital can rewrite the whole equation.

Axis 3: Brand impact on the rest of the group

This is the axis operators most often ignore in the spreadsheet and pay for later. A location does not exist alone. It is a public-facing member of a brand family, and its quality feeds into how guests, staff, and prospective hires think about every other unit in the group.

A weak location damages the group in ways that do not show up on that location's P&L:

Reviews. A one-star review at your weakest location shows up in search results for your strongest one. Guests do not always parse which specific address they visited when they leave a review.
Staff turnover contagion. Line cooks and servers at the weak location talk to line cooks and servers at the strong locations. Bad morale is transmissible.
Hiring signal. Prospective hires read the whole brand, not one address. A visibly struggling location makes the group harder to recruit into.
Guest confidence. A regular who has a bad experience at the weak location will hesitate before booking any location for the next occasion that matters to them.

None of this shows up as a line item. All of it costs the group real money. When you evaluate a location for closure, ask this question directly: if a guest, a hire, or a food writer only ever visited this location, would they be more or less likely to visit any other location in the group? If the answer is less likely, brand impact is negative, and it is a vote for close.

Time-boxing the turnaround attempt

The framework above assumes you have made a real turnaround attempt. Do not close a unit before you have tried to save it, and do not let the "trying to save it" phase run indefinitely.

A real attempt is four to six months of actual operational work: schedule redesign, food cost recovery, general manager coaching, service-level rebuild. Calendar time does not count. If a unit has been "in turnaround" for a year with no meaningful operational intervention, that is a year of losses without a real signal.

Set the decision date on the calendar at the start of the attempt. Month five, month six. Whatever the window is, put it on the calendar and hold the date. If you do not, the decision drifts, and the drift is where the losses compound.

The emotional resistance

Here is the part that framework alone does not fix. Closing a restaurant is emotionally hard, and it is emotionally hard even for operators who have done it before.

You know the team. You know the general manager who has been there since the opening. You know the regulars who have watched the space go through three menus. You remember the walk-through with the founder before the buildout, or the first night of soft open, or the first time the espresso machine actually worked. All of that history sits in the room with the decision.

The emotional resistance is not something to shame yourself out of. It is a real variable. Name it. Say out loud that closing this unit is going to feel like a personal failure, and that the feeling is not evidence about whether the decision is right. Then get an outside voice into the room. Someone who was not there for the opening. Someone who can look at the three axes without the story getting in the way.

On the Bay Area group, we had one location that everyone had a story about. The founder had proposed to his wife across the street. The general manager had been there for seven years. The regulars were in the walls. When we ran the framework without flinching, contribution was slightly negative, the lease had 16 months left, and the brand impact was mixed. Two axes leaning close. We closed it, redirected the capital into extending the strongest location, and the group went from three units to five inside a year. The closed location was the funding source for the growth. That is not what it felt like at the time.

Closing a unit is not the end of the group. It is often the beginning of the group actually being able to grow. The capital tied up in the wrong location cannot fund the right one.

How to actually close a location well

Once the decision is made, the execution matters. A badly closed location damages the group more than a well-closed one. Two things to get right:

Take care of the team first

Every displaced staff member gets a redeployment offer to another location if possible, at their current rate. If not possible, they get real severance, a written reference, and help with the next placement. The team at the closed location knows every other team in the group. How you close this one is how the other teams learn what happens when things get hard.

Communicate to guests before they hear it from a sign on the door

Email the regulars. Post to social. Tell the neighborhood. Frame it straight. A restaurant that closes with dignity leaves the door open for future units in the same market. A restaurant that closes by locking the door and posting a notice leaves scars that follow the brand.

Manage the last thirty days as a real operation

The month between the decision and the final day of service is where operators most often let discipline slip. Do not. The unit continues to serve real guests, and every one of those last services will be remembered longer than any of the previous six months. Keep the standing meetings. Keep the line checks. Do not skimp on repairs the unit still needs to run cleanly. The team is watching how you close as much as they were watching how you tried to save.

The point

The close-or-save decision is one of the highest-impact calls in a multi-unit turnaround. Made well, it frees capital and attention for the units that can grow. Deferred, it drains both.

The framework is not complicated. Contribution positive or negative. Lease overhang long or short. Brand impact positive, neutral, or negative. Two out of three leaning close is a close. All three is a close this quarter. Time-box the turnaround attempt at four to six months of real work and hold the decision date on the calendar.

The hard part is not the math. The hard part is the emotional weight of the decision, and the way that weight quietly extends the timeline until the losses have compounded past the point where a save was ever possible. Name the resistance. Bring in an outside voice. Hold the date.

The location you should close is the one that has been sitting in your gut for six months. You already know. The framework is for the courage to act on what you know.

Staring at a location you cannot decide about?

I take on a small number of multi-unit rescue engagements each year, mostly restaurant and hospitality groups in the $10M to $60M revenue range. If you have a save-or-close call you keep deferring, I am happy to trade notes.

Get in touch

Questions people ask

When should you close an underperforming restaurant location?

When the unit cannot cover its direct operating costs at realistic sales, when the lease overhang is shorter than the payback period on any credible turnaround investment, and when the location's brand impact on the rest of the group is negative rather than neutral. If any two of those three are true, close. If all three are true, close now.

What is contribution versus absorption in a restaurant P&L?

Contribution is what the location produces after its own direct costs (labor, food, rent, utilities, unit-level management). Absorption is contribution minus the location's fair share of corporate overhead. A unit can be positive on contribution and negative on absorption. That unit is worth keeping if the group needs the contribution to cover overhead. A unit negative on contribution is losing money on every cover it serves and should be closed.

How long should you give a turnaround location before closing?

Between 4 and 6 months of actual turnaround work, not calendar time. If the unit has not shown a clear contribution improvement by month four of a real intervention, the market or the location is telling you something. Extending past six months usually reflects the operator's emotional attachment, not the unit's real prospects.

How does lease overhang affect the close decision?

The remaining lease term is your effective payback window. If you have three years left on the lease and a credible turnaround requires an 18-month investment window to reach payback, the math can work. If you have 14 months left and a 24-month payback, the math cannot work. Also consider whether the lease can be assigned, subleased, or bought out at a reasonable cost. Sometimes the buyout cheaper than continuing to bleed for the remaining term.

Why is closing a restaurant location so hard emotionally?

Because you know the team. You know the guests. You remember the opening. The people making the decision at the top of the org almost always have relationships that make closing feel like a personal failure. The way through this is to name the emotional resistance as a real factor, get an outside voice into the decision, and set a decision date on the calendar before the emotional resistance has time to expand.

What brand damage does keeping a bad location cause?

Bad locations produce bad guest experiences, bad reviews, and staff turnover that spills into other units. Guests do not always distinguish between locations of the same brand. A one-star review at your worst unit gets read by someone considering a reservation at your best one. Keeping a bad location open costs the group brand equity that the other units have to pay for.

Can you turn around a location the market has already rejected?

Rarely. If the trade area does not have the density, income, or dining occasion to support the concept, operational excellence cannot fix that. The site was miscast. In those cases the honest move is to close, absorb the loss, and redeploy the capital into a site that fits the concept. Trying to run the same concept better in a wrong location is a common way to burn 18 more months of losses.