The Empty Chair Experiment: Running a Three-Week Absence Audit on Your Own Company

There is a question every owner eventually asks in the shower or on the drive home, and almost none of them answer honestly: if I disappeared for three weeks, what would break? The instinctive answer is "nothing, we'd be fine." The accurate answer, in most companies under $20 million in revenue, is considerably less reassuring.

The problem is not lack of effort or intelligence on your team. It is that founder dependence is invisible from the inside. You cannot see the bottleneck you are standing in. What follows is a structured exercise — call it the Empty Chair Experiment — designed to convert that vague suspicion into a documented, scored diagnosis.

Why Three Weeks, Specifically

One week proves nothing. Any competent team can coast on goodwill, backlog, and momentum for five business days. Two weeks starts to strain, but clever people paper over the gaps. Three weeks is where the mask slips. Payroll questions surface. A key customer asks for something nonstandard. A vendor dispute escalates. A hiring decision cannot wait. Three weeks is long enough that the business must govern itself, not merely endure.

The simulation is hypothetical — you are not actually leaving. You are auditing as if you were, because the discipline of an imminent departure forces honesty that a survey never will.

Step One: The Decision Log

For one week, carry a small notebook or open a running note on your phone. Every time someone asks you a question, write it down. Do not filter. At the end of the week, sort every entry into three columns:

1. Anyone could answer this. (The answer exists in a document, a dashboard, or common sense.)
2. Someone specific could answer this, but they came to me instead. (A routing problem.)
3. Only I could answer this. (A dependence problem.)

Column three is your exposure. Typical entries: pricing exceptions beyond a certain threshold, whether to extend credit to a slow-paying account, which candidate to hire, how to respond to a partner's contract language, whether a customer's complaint warrants a concession. Each one is a decision the company cannot currently make without you.

The count matters less than the type. If column three is dominated by judgment calls — pricing, credit, people, contracts — your business has no codified decision rights. That is fixable. If it is dominated by trivia, your team has a confidence problem, which is cheaper to fix but requires you to change your own behavior first.

Step Two: The Relationship Map

Draw a simple diagram. In the center, write your name. Around it, list every external relationship where you are the sole contact: the bank, your largest three customers, the commercial landlord, the insurance broker, the CPA, the key vendor whose rep takes your calls on the first ring.

Now ask the uncomfortable question: for each of those names, does anyone else in the company have a real relationship — one where a phone call would be answered and a favor could be asked? In most founder-led businesses, the answer covers a short list. Your best customer may like your operations manager fine, but when there is a dispute at renewal time, the conversation runs through you.

This is not a character flaw; it is a design flaw. Relationships that only you hold are assets the business does not actually own. Assign a secondary contact to each one, require that person to join the next two meetings, and let the counterpart know — explicitly — that this person speaks for the company.

Step Three: The Head-Dump Inventory

This is the least glamorous and most revealing step. Sit down for ninety minutes and list every process, fact, or preference that exists only in your head. The discounting logic you apply but have never written down. The reason you always schedule deliveries before 10 a.m. The history behind the handshake arrangement with the vendor in Ohio. The password to the state filing portal. The real reason Customer X left two years ago.

Each item is a single point of failure wearing the disguise of institutional knowledge. You do not need to document all of it this week. You need to rank it: what would cost the most money or the most sleep if it walked out the door with you?

Scoring the Audit

Score your company on three axes, one to five each:

• Decision autonomy: Can the team make the routine calls that keep revenue flowing? (Five means yes, without checking in.)
• Relationship redundancy: Do key external relationships have at least two faces attached?
• Knowledge transfer: Are critical processes and context written down where a competent newcomer could find them?

Below nine total, a three-week absence would cause measurable damage — lost deals, delayed payroll decisions, a customer who felt abandoned. Nine to twelve means you would survive but bleed. Above twelve, you have an actual company rather than a job with employees.

What to Do With the Results

The audit produces a to-do list, and the list is almost always the same three items in different clothes: delegate decision rights with explicit dollar thresholds, attach a second face to every key relationship, and write down what only you know. None of it requires new software or a consultant. All of it requires you to tolerate the discomfort of being slightly less necessary.

That discomfort is the point. The most valuable outcome of the Empty Chair Experiment is not continuity planning — though that alone justifies the exercise. It is discovering how much of your company's daily friction is manufactured by your own indispensability. Fix that, and you get back the one asset no owner can buy: the ability to step away without the whole thing wobbling.