On The Line

The Chef and Operator Partnership That Actually Works

Almost every chef-operator conflict traces to one root: decision rights were never written down. Here is the split that lets both do their best work, and the weekly cadence that keeps the partnership alive.

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

The short version

The chef owns the plate. The operator owns the P&L. Both jointly own the guest, and that overlap is where the partnership either works or breaks.
Write the decision rights down in a one-page matrix. Most chef-operator conflict evaporates when both parties can point to a document.
The chef and operator are peers, not manager and report. Making one report to the other collapses the partnership.
One 45-minute weekly working meeting is the entire infrastructure. Miss it for a month and the drift begins.
The partnership works when both parties share the same numbers, the same guest feedback, and the same operating rhythm.

The single most productive working relationship in a restaurant group is the one between the chef and the operator. It is also the one that goes wrong most often. Every group I have worked inside or advised has some version of the same story: a talented chef, a competent operator, both trying to do good work, and a slow accumulation of unresolved decisions that turn into resentment, then into avoidance, then into one of them leaving.

The story is almost never about personality. It is almost always about decision rights that were never written down.

I spent five years inside a $30M Bay Area restaurant group where the chef-operator partnership was the load-bearing wall of the whole operation. When we got the partnership right, we scaled from three to five locations, held Michelin recognition, and pushed 215 people through a consistent standard. When we got it wrong, the friction cost us plates, staff, and a couple of unnecessary quarters. This piece is what I would tell any group standing up this partnership today.

Why the partnership breaks

The partnership breaks in one of three ways, and all three have the same root cause.

First, the chef and the operator both think they own a decision. A menu change is coming. The chef assumes she owns it because it is a plate. The operator assumes he owns it because it is a P&L input. Both push. One of them concedes reluctantly, and the concession accumulates as debt in the relationship.

Second, one of them assumes the other owns a decision that actually needs joint input. A kitchen hiring decision goes through without the operator being consulted. A pricing decision goes through without the chef being consulted. The excluded party finds out after the fact, feels stepped over, and either escalates or checks out.

Third, both parties think the other has a shared understanding of the numbers, and neither does. The chef is working from a food cost estimate. The operator is working from an actual food cost that is 4 points higher. Both are making decisions off the wrong data, and neither knows it until quarter-end when the P&L surfaces the gap.

All three of these are decision rights failures. All three are solvable with a one-page document. Almost nobody writes the document.

The decision rights split that holds under pressure

The version I ran, and the one I would recommend to any group at any scale, splits decisions into three buckets: chef-owned, operator-owned, and jointly owned. The rule is that anything jointly owned requires both signatures. Anything owned by one party is theirs, and the other party is a required input but not a decision maker.

Fig. 1 · One page. Written. Signed. Referenced when tension rises.

What the chef owns outright

Anything that is a food or plate decision. Recipe development. Plate standard and photo card. Kitchen staffing including sous chefs, line cooks, and prep. Purchasing specification (what protein, what supplier grade, what cut). Menu writing (what dishes go on the menu). These are the chef's calls. The operator is an input, not a decision maker.

What the operator owns outright

Anything that is a business or P&L decision. Menu pricing. Menu mix (which dishes get pushed on the floor). Unit-level P&L. Front-of-house staffing. Vendor contracts and payment terms. Expansion pace and unit real estate. Marketing spend and channel. These are the operator's calls. The chef is an input, not a decision maker.

What they jointly own

Anything that affects both the plate and the P&L in a material way. Kitchen leadership hires above sous chef. Capital projects (new equipment, new kitchen buildout). New unit opening. Recipe changes that materially move cost or complexity. Both signatures required. If they cannot agree, the decision escalates to the owner or the group principal, and both know that in advance.

The one-page decision rights matrix is not a bureaucratic tool. It is the document that keeps both parties respected and both parties productive. Skip it and you are hoping for a partnership. Sign it and you have built one.

The weekly cadence that keeps the partnership alive

Decision rights on paper is necessary. It is not sufficient. What holds the partnership together on a week-to-week basis is a short, standing, working meeting between the two parties. No delegates. No extra attendees. Forty-five minutes, once a week, same time.

The agenda is fixed:

Last week's numbers. Food cost, labor as a percent of sales, top three sellers, top three margin drags, guest complaint themes. Both parties look at the same dashboard.
Menu changes coming. What the chef is developing, when it lands on the menu, what the cost estimate is, what the operator needs to price it.
Staffing decisions. Kitchen and floor. Any hire, any promotion, any conversation the other party needs to know about.
Guest feedback themes. Complaints from the log, patterns forming, any table incident either party needs to hear about.
One decision each. Every meeting, each party brings one decision they need the other's input on before making. Not a discussion. An input, then the decision.

Meetings that follow this agenda run in 45 minutes. Meetings that do not follow it drift to 90 minutes and cover less. The standing agenda is what makes the meeting a working tool rather than a check-in.

Where the tension lives

Even with the decision rights signed and the weekly meeting running, three areas produce recurring tension. The healthy partnerships surface the tension in the weekly meeting. The unhealthy ones let it accumulate.

Menu changes with cost implications

The chef wants to add a new dish. It is beautiful. The recipe cost is 42 percent. The rest of the menu runs at 30 percent. The operator has to push back on the number, and the chef has to hear the pushback without taking it personally. The partnership works when both parties come to the meeting knowing that the number will get discussed. It breaks when the operator raises the cost as an attack on the chef's creativity, or the chef treats the cost concern as a lack of vision.

Pricing decisions during inflation

Food cost rose 6 percent this quarter. The operator wants to raise menu prices 4 percent. The chef worries about the guest experience. Both are right. The partnership works when the operator lays out which items can absorb a price increase and which cannot, the chef weighs in on the guest impact, and both sign off on a specific set of price changes rather than a blanket increase.

Kitchen hiring during a growth phase

A new unit is opening in six weeks. The chef wants to promote a sous chef from an existing unit. The operator worries about the hole that leaves in the existing unit's kitchen. Both are right. The partnership works when the two parties look together at the bench strength across every kitchen and jointly decide who moves and who backfills, in the same meeting.

What actually holds when the partnership is working

When the chef-operator partnership is working, three things become visible in the operation.

The plate holds across shifts. Because the chef owns the standard and the operator owns the operating rhythm, the plate that leaves the pass on a full Saturday matches the plate that left on a slow Tuesday. Neither party is fighting the other's discipline.

The numbers hold across quarters. Because the operator has the chef's read on menu items before pricing decisions, prices track cost. Because the chef has the operator's read on mix, the menu evolves toward items that carry margin without losing character.

The team below both of them stops feeling like a battleground. Sous chefs and general managers know who to go to for what. They are not being pulled between two leaders who disagree on the direction. That alone is worth a percent or two of retention across the year.

Scaling the partnership past two seats

The chef-operator partnership works well at one restaurant. At three, it starts to strain. At five or more, it either evolves into a partnership of leadership teams or it breaks. The pattern is predictable and the fix is well understood.

What changes at scale is that the chef can no longer be in every kitchen and the operator can no longer be in every dining room. Both need lieutenants who can carry the standard and the numbers into rooms neither leader is in. That means an executive sous chef under the chef and a director of operations under the operator. Both new seats need to be introduced to the weekly partnership meeting cadence.

The pattern that works is a monthly operating review that expands the weekly meeting into a four-person conversation: chef, operator, executive sous, director of ops. Same standing agenda, longer format, focused on rolling patterns across the group rather than acute decisions. The weekly meeting stays a two-person meeting. The monthly review is where the group operates as a team.

The failure mode at scale is the chef and operator continuing to meet weekly, in a two-person conversation, on decisions that now actually affect four people. The executive sous and the director of ops feel bypassed, decisions get made without their input, and both of them start making their own calls on the floor that contradict what the two leaders decided. Two months of that and the group looks like it has four leaders pulling different directions.

The partnership does not scale by adding people to the weekly meeting. It scales by holding the weekly meeting at two people and building a monthly meeting that includes the lieutenants who now carry the decisions.

Mistakes I made and would not repeat

Three mistakes from my own operator seat, worth naming plainly.

Trying to win the pricing argument in the moment

Early on, I would push a pricing decision through in a heated conversation because I knew the number and the chef did not. I won the moment. I lost the partnership for a week. The right move is to bring the pricing case to the weekly meeting with the data prepared, let the chef read it, and let the decision happen with respect on both sides. Speed on a pricing call is worth less than trust in a partnership.

Under-communicating hiring intent

I once made an operator-side hire (a general manager) without giving the chef a heads-up. Not a decision the chef had rights over. But a decision that affected the chef's day. I should have mentioned it in the weekly meeting the week before. Not asking permission. Just sharing the intent. That single omission cost me a month of chill on other topics.

Treating the meeting as skippable when the week was busy

The weekly meeting is the whole infrastructure. Skipping it because the week is busy is exactly backwards. The busy weeks are the ones that need the meeting most. Skip the meeting for two weeks and you will spend the next two months repairing decisions that could have been made cleanly if you had held the meeting.

The point

The chef and operator partnership is the load-bearing wall of any restaurant group above a single unit. Almost every failure of that partnership is a failure to write decision rights down and hold a weekly working meeting. Both are cheap. Both are avoidable failure modes. Both are avoided by operators who take the partnership seriously as a piece of operating infrastructure, not as a personality question.

The chef will do her best work when she owns the plate and knows the operator is not going to second-guess a recipe. The operator will do his best work when he owns the P&L and knows the chef is not going to unilaterally raise the food cost of the menu. When both of those are true, the group has the foundation to scale without breaking. Everything else is downstream of that.

Standing up a chef-operator partnership in a group that has grown past one seat?

I have spent the last decade inside multi-unit F&B operations, including a $30M Michelin-recognized Bay Area group where the chef-operator partnership was the whole game. If you are trying to write the decision rights matrix, I am happy to compare notes.

Get in touch

Questions people ask

Who owns the menu, the chef or the operator?

The chef owns the menu. The operator owns the mix. That means the chef decides what a dish is, how it is plated, and what the standard is. The operator decides which dishes stay on the menu based on margin, mix, and prep complexity. The chef writes. The operator prunes. Both agree in a weekly working meeting, not in a fight.

What causes chef-operator conflict?

Almost every conflict traces to one root: decision rights were never written down. The chef assumes they own something the operator also thinks they own. When a decision has to be made under pressure, both push and one loses. Two conflicts later, the partnership is strained. Writing the decision rights down in a one-page matrix eliminates 80 percent of the friction.

Should the chef report to the operator or the other way around?

Neither. In a healthy restaurant group at any real scale, the chef and the operator are peers, both reporting to the owner or the group principal. The chef leads the kitchen and the plate. The operator leads the P&L and the units. Making one report to the other collapses the partnership into a hierarchy and the losing party either checks out or leaves inside a year.

How do you split decision rights between chef and operator?

The chef owns recipe, plate standard, kitchen staffing, purchasing spec, and menu writing. The operator owns pricing, menu mix, unit-level P&L, front-of-house staffing, vendor contracts, and expansion. Both jointly own hiring for kitchen leadership, capital projects, and any change that affects both the plate and the price. That is the split that works in practice.

What is a good weekly cadence for the chef and operator?

One 45-minute working meeting a week, standing agenda, both parties present, no other attendees. Menu changes coming, cost issues surfacing, staffing decisions, guest feedback themes. This meeting is the entire partnership infrastructure. Skip it for two weeks and the drift starts. Skip it for a quarter and the partnership breaks.

Who decides when to raise menu prices?

The operator owns the pricing decision because the operator owns the P&L. The chef is a required input because the chef owns the recipe cost and knows which items can absorb a price increase and which cannot. In practice this is a joint decision with the operator holding the final call. If a chef unilaterally raises prices, or an operator raises prices without the chef's read, the partnership is already broken.

Can a chef also be the operator?

In a single unit at low volume, sometimes. Above two units or above roughly $3M in annual revenue, almost never well. The two roles pull attention in different directions. A chef running operations makes food decisions with a P&L bias and operations decisions with a food bias. Neither optimizes. Separating the seats and building the partnership between them produces better food and better numbers.