A buyer does not buy a workforce. He buys what keeps running on the day you are no longer there, and what keeps running if a few people leave. Team and succession are therefore not one value driver alongside the rest, but the question that runs through all the others: is this company a collection of people who happen to complement each other well, or is it a process that people carry out and that can also continue without those specific people.
AI changes the weight of that question, not by replacing staff as a prediction, but because today already parts of the work can proceed without the current staffing. Some tasks a system can take over without oversight. Other tasks a system can prepare or execute while a human approves or rejects, with reason. Still other work remains human work, because it requires judgment, negotiation, or relationships that are not recorded. That division differs in every company, and that is where the difference lies between companies that already see this driver improving and companies where nothing changes.
A buyer asks himself a number of questions that all come down to dependency. Does revenue run through relationships that one person has built, or through a process that is documented and transferable. Is knowledge of customers, prices, and exceptions recorded somewhere, or does it only exist in people's heads. Can a key role fail without output stalling. And, specific to the current shift: which part of the work now done by people is already described in enough detail that a system can take it over or prepare it. A company where this has been worked out and documented weighs differently than a company where no one knows.
This also affects how a buyer assesses profit. A saving that arises because one person happens to work efficiently counts differently than a saving that is built into the process itself and thus transfers upon sale; this connection with profit quality is further explained at why a buyer weighs profit quality differently now. And it affects the question of who provides leadership after the owner: a buyer looks at the management team not to see whether it is pleasant, but to see whether the company has a governance layer that functions without the founder, which is explained further at why a buyer looks at the management team.
The owner dependency index makes this concrete: how much of the revenue, the decisions, and the customer contacts run through the owner or through one irreplaceable employee. You measure that against a few tangible things. Can a quote, an exception to pricing policy, or a complaint handling be done by someone other than the regular person, without the customer noticing the difference. Is the knowledge needed for those tasks stored in a document, a system, or only in someone's head. Was there, in the past twelve months, a period in which a key person was absent, and what happened to output during that period.
Beyond that, there is the question of takeover by AI, separate from the question of transfer to colleagues. For each task it can be established whether it can be fully taken over by a system, partly with human oversight that approves or rejects, or whether it remains human work. That distinction is not the same question as 'who can take this over if someone leaves', but it is a question that changes the answer to it: work that a system can already prepare is less dependent on the specific person currently doing it. Figuring this out precisely, task by task, is what FTE TO AI's work scan does.
The shift is uneven. In one company, customer communication is already largely prepared by a system and checked by a human; in another company, all customer knowledge still resides in the head of one account manager. That difference does not come from the sector or the size, but from whether the work has ever been written down at a level a system can follow. Companies where processes, role definitions, and decision rules have been documented see owner dependency decline faster than companies where that still resides in experience; this documentation is connected to how processes and systems are structured, explained further at what AI does to processes and systems.
What is not at issue here is a judgment about who stays and who leaves. What an employer does with its staff falls under its own statutory requirements for labor law and employee participation, and that is for the employer and its advisors to determine. What is measured here is something else: how much freed-up capacity and which FTE deployment arises if certain tasks are prepared or taken over by a system, and what that means for how dependent the company is on the people currently keeping it running.
The owner dependency index and the two-year calendar in the value scan show which steps reduce dependency on specific people before a sales process begins, in connection with the other drivers a buyer weighs, including market position at what AI does to market position and the legal documentation of knowledge and contracts at what AI does to legal hygiene. How these eight drivers together determine the value of a company is explained at how the value of your company is determined.
Anyone who wants to know where this stands today can take the free value check: eight short questions, one per driver, with a picture of which driver is putting the most downward pressure on the company's price today. The full value scan, with evidence per driver and the two-year calendar, is under construction.