Essay · Executive Depth

The Fifth Chair

Autumn 2026 · Reading time · 16 minutes

Every consequential meeting has four people at the table and one person who decides the outcome. That person is not in the room.

I run several companies and I keep a fifth chair at every meeting that matters. Four people take their seats. Each carries a role, a set of interests, a way of speaking. The founder speaks. The operator speaks. The analyst speaks. The designer speaks. The decision does not come from any of them. It comes from a fifth chair, deliberately empty, into which each of the four is expected to step, one at a time, before the meeting closes. What emerges from that chair is not the average of the other four voices. It is a different voice altogether. Owners who learn to leave that chair open produce different companies than those who fill every seat with a defender.

Why the seat is empty on purpose

The temptation in any executive meeting is to fill every chair with a defender. Every function protects its own edge. Every voice pulls the decision toward the ground its speaker already stands on. This is not weakness, it is competence doing what competence is trained to do. The trouble is that a decision made from those four seats will always be a compromise between four competent defenders and a compromise between defenders is almost never the sharpest available move.

The fifth chair is the place from which the decision itself speaks. It does not belong to any function. It belongs to the situation. When I ask a colleague to sit in that chair, I ask her to speak not as her role but as the situation itself, momentarily using her mouth. What would this quarter, this market, this constraint ask of us if it could talk? The answer she produces from the fifth chair is often different from the answer she produced from her own seat three minutes earlier. That difference is where the value of the practice lives.

The chair is a role, not a person

Different meetings call for different occupants. In a hiring decision, the fifth chair is the future colleague you have not yet met, whose first week is being shaped right now by policies she will never see you write. In a pricing decision, the fifth chair is the customer being asked to pay, and the customer she will become in year three. In a partnership decision, the fifth chair is the version of the company you will be eighteen months from now, looking back on today with either gratitude or regret.

The fifth chair, in my daily practice, is almost always a person who is not in the room and may never see the meeting. That is precisely why the chair matters. The four people in the room have advocates. They are their own advocates. The fifth chair is the advocate for the party who is not present, and if we do not seat that advocate deliberately, no other seat produces her voice.

The chair a model cannot sit in

A useful test of the practice arrived when I started routing certain decisions through an AI model as a fifth voice. The model is very good at compression, at summarizing all four defenders, at proposing what a reasonable synthesis would say. It is almost useless as a fifth chair. A model does not sit in the situation, it patterns across situations. It answers the question the meeting asked. The fifth chair answers the question the meeting did not know it was asking, because it speaks from a party the meeting did not think to consult. A model can be a very fast fourth defender. It cannot be the empty seat.

This distinction has become one of the more useful lines I draw for owners who are trying to figure out where automation belongs in their decision loop. Wherever the answer is a compression of what the room already knows, delegate freely. Wherever the answer requires a party who is not in the room to speak, keep the seat for a human who is willing to accept the responsibility of standing in for someone else. Owners who confuse the two are already losing decisions and they do not yet know it.

The failure mode I see most often

The failure mode is not that leaders forget the fifth chair. The failure mode is that one of the four defenders occupies it in disguise. The founder speaks from her own seat, then repositions herself in the fifth chair and speaks again, still from her own interests, now given the rhetorical authority of the whole. It sounds like judgment. It is only repetition. The room senses this, though rarely names it, and trust in the fifth chair silently degrades. Six months later the meetings look inclusive and everyone in them has stopped bringing their real opinion.

A cleaner protocol is to name aloud who is sitting in the fifth chair, in what role, for how long. When we say, out loud, that for the next five minutes the fifth chair is occupied by a first-year customer who has never used the product before, the room automatically produces a different quality of speech. Even the founder cannot smuggle her own interests through the chair without being audibly at odds with the role she just accepted. Naming the occupant is the enforcement mechanism.

The chair I keep for my own team members

A less obvious use of the fifth chair, which took me several years to notice, is that it can be seated with team members who are in the room but are not senior enough to speak freely. A junior colleague sitting at the table often has the sharpest read on the decision, because she is closest to the work and least invested in the political geometry of the room. She rarely speaks. When she does speak, her comment is treated as one input among many and quickly folded into the discussion above her.

Seating her explicitly in the fifth chair for the last ten minutes of the meeting changes what she is able to say. She is no longer commenting as her role. She is commenting as the situation itself and the room is under an obligation to hear her at that register. I have watched decisions turn on the intervention of a colleague who would not have spoken in her own seat, precisely because the fifth chair gave her the temporary authority her position had not yet earned. Owners who use the chair this way build a specific kind of company, one in which the person closest to the work has a durable path into the conversation about it.

Why this is depth, not technique

The practice does not work without a specific psychological posture. The owner must be able to hold, for the duration of a meeting, the honest suspicion that her own view is one of four, not four of four. Most leaders can perform this posture. Fewer can hold it. The performance produces meetings that look inclusive and remain, in fact, dominated. The genuine posture produces meetings that occasionally embarrass the owner and it is precisely those meetings that produce the sharpest decisions of the year.

The discipline of an owner is not the discipline of not speaking. It is the discipline of speaking in the correct register for the seat one currently occupies. The fifth chair has a register of its own: cooler than defender, warmer than judge, always accountable to a party who is not present to defend itself. Owners who develop that register keep more good people, close more good deals and lose fewer good decisions to the gravity of the loudest voice in the room.

How to install the practice

The mechanics are simpler than the psychology. Add one line to the agenda of every consequential meeting: Who sits in the fifth chair today? The answer is decided at the top of the meeting, not the bottom. Assign it explicitly to a person for the duration. Give that person permission to interrupt the four defenders whenever their view drifts too far from the interests of the absent party the fifth chair represents. Rotate the assignment across meetings so that no single person becomes the resident conscience of the room.

The first three or four meetings under this protocol are awkward. Defenders find their arguments interrupted. The person in the fifth chair overreaches or underreaches. By the sixth or seventh meeting, the room begins to speak differently even without the interruption, because everyone has internalized that a chair is watching them from a position they cannot occupy from inside their own function. That internalization is the change one is actually after. The chair, over time, becomes structural rather than ceremonial.

Why the chair matters more this decade than last

One of the quieter shifts of this decade is that the parties who are not in the room have gotten more numerous, more powerful and more distant. Ten years ago, the absent parties in a boardroom were a small set of stakeholders and customers. Today they include algorithms that will govern how a product is discovered, models that will rewrite the interface the customer actually experiences and regulators who will decide, in two years, what the whole industry is allowed to sell. Every one of them is affected by the meeting on the calendar today. None of them can send a representative.

The fifth chair is the only reliable mechanism I know for keeping these newer absent parties from being systematically underweighted in a company's decisions. Owners who leave the chair empty in the old sense, meaning simply unfilled, are effectively casting a vote for the interests already present in the room and against every party whose voice has to be constructed. In a decade whose most important stakeholders are almost never physically in the room, the chair has moved from being a useful practice to being a load-bearing one.

The three occupants I use most

Over the years three specific occupants of the fifth chair have earned recurring seats in my meetings. The first is the customer who has not bought yet, the person who has heard our name once, is skeptical for reasonable reasons and has three other options open in her browser tabs. When she sits in the fifth chair, our marketing decisions get sharper by an order of magnitude, because the room stops describing the product to itself and starts describing it to someone who has not yet made up her mind about us.

The second is the employee we will hire eighteen months from now, whose first week we are shaping today with policies that will feel to her either like a company that respects her attention or a company that consumes it. The third is the version of the business we will run in five years, looking back at this meeting and asking whether the compromise we are about to accept was one we can live with when the business is twice this size and there are twice as many stakeholders defending it. Any of the three, seated properly, will change the answer the room was about to produce.

Two habits that keep the chair from becoming a ritual

The risk with any practice that survives past its first year is that it becomes ritual. The chair is named, the assignment is made, the person occupies it and speaks in the expected register, and the room feels virtuous for having gone through the motion. Nothing about the decision actually changes, because the occupant is performing the role rather than genuinely stepping into the perspective. The ritualization is subtle and it usually announces itself as smoothness. A fifth-chair intervention that flows too easily is almost always a fifth-chair intervention that did not do its work.

Two habits keep the practice honest. The first is rotating the occupant across meetings so that no single person becomes fluent enough in the role to phone it in. The second is spending five minutes at the top of the meeting on why this particular occupant, for this particular decision, is the right choice. When the room has to argue briefly about who the fifth chair represents today, the person eventually seated arrives with a specific charge rather than a generic one and the intervention that follows is harder to reduce to ritual.

The chair after the meeting

The fifth chair does not disappear when the meeting ends. It follows the decision out into the week and continues to speak, quietly, as the decision meets reality. I keep a short list, in a private notebook, of the fifth-chair objections I overruled in the meetings where I chose to. Some of them turn out to have been correctly overruled. Some of them turn out to have been the sharpest signal in the room and by the time I recognize them, the decision they warned against has already begun to age poorly.

Reviewing that list quarterly is one of the more painful hours of my calendar and one of the more useful. It is the closest thing I have to a personal instrument for calibrating my own judgment against the parties I claim to be serving. Owners who never look back at the objections they overruled tend to accumulate a hidden pattern of overruling in favor of the same recurring interest, usually their own convenience and the accumulation eventually shows up in a way the market will not let them ignore.

What the practice will cost you

It will cost speed. A meeting with an active fifth chair takes fifteen to twenty percent longer than the same meeting without one. It will cost occasional comfort. Founders discover that their favorite move is often the move the fifth chair objects to most cleanly. It will cost the fiction that the loudest voice in the room is the smartest voice in the room. What you gain in return is a decision-quality curve that bends upward over quarters and years, not months, and a team that begins to trust the meeting itself as a decision-making instrument rather than as an arena in which the strongest personality wins.

The fifth chair is not a technique. It is a small architectural choice about who is allowed to speak for the parties who are not in the room. In a decade where the room is expanding to include employees who have not been hired yet, customers who do not exist yet and technologies whose first customer will be your grandchild, the fifth chair is the only chair honest enough to hold that future accountable to the meeting that is deciding for it. Every serious owner I know has made that choice, one way or the other, whether or not she has a name for the practice. Once you can point to the seat, the practice teaches itself.