A recreation business lives on guest contact, occupancy and availability. The hours sit in reservations, planning, communication with guests before and during their stay, maintenance of accommodations, and in the administrative flow around bookings, invoices and seasonal rosters. Part of that work peaks with the season, part runs throughout the year. Anyone assessing the value of such a business looks not only at occupancy rate and margin, but at the question of how much of that work is tied to individuals and how much is embedded in the process itself.
The outcome is driven by circumstances that differ per business: the number of booking channels and whether they run separately or through one system, the extent to which guest communication follows a standard or is reinvented case by case, and whether maintenance and purchasing run on fixed arrangements or on the improvisation of one person who knows where everything is.
AI is taking over work, and it happens in parts. Three categories run through everything: work a system can handle independently, work where a system makes a proposal and a person approves or rejects it with reason, and work that remains human work because it requires guest contact, negotiation or on-the-spot judgment.
For reservations and standard communication with guests -- confirmations, reminders, frequently asked questions about arrival and facilities -- the first category is within reach in most cases: the system handles it independently. For seasonal pricing and availability across channels, much of the work falls into the second category: the system calculates and advises, a person assesses and adjusts. For complaints, custom requests from guests and the on-the-spot assessment of maintenance quality, most of the work remains human work, because the situation is different each time and the judgment cannot be captured in rules.
The difference between businesses lies not in the sector but in how things are set up. A business where bookings, guest communication and maintenance logs are already in one system can now largely run that first and second category through it. A business where the same information sits in the head of the owner or one fixed employee cannot hand that work over to a system, simply because it was never recorded. That is not a matter of willingness, but of what has been built up.
A buyer does not buy the number of people, but what continues to run without those people. That changes the weight of the value drivers assessed at a sale. The owner-dependency index reacts most sharply: if seasonal planning, supplier arrangements and dealings with regular guests all rest with one person, value drops regardless of revenue. What succession readiness means for the price shows how that dependency is concretely translated into a discount or a premium at sale.
But profit quality also changes in character. A saving that arises because one external party cheaply carries out a task counts differently than a saving that is anchored in the process itself -- for example because guest communication runs through a system as standard, independent of who is on duty that day. The first saving disappears as soon as the relationship with that party changes; the second holds up when the business is sold. Buyers weigh that difference, even when it is not visible in the profit-and-loss statement.
This shift is not limited to recreation. In construction, comparable recording of planning and calculation determines what determines the value of a construction business now that AI is taking over work, in healthcare it runs along roster planning and record-keeping as described in what determines the value of a healthcare organization now that AI is taking over work, and in the cleaning sector it depends on planning and quality control, set out in what determines the value of a business in the cleaning sector now that AI is taking over work. The pattern is always the same: where work has been recorded, it shifts; where it sits in someone's head, it stays put and weighs down the price.
Where this shift touches on decisions about staff, separate legal requirements apply; this is not personnel advice and not a basis for a dismissal decision. This concerns facts about work: what a system can take over, what requires human oversight, and what remains human work.
The underlying question -- which work in this business can genuinely be taken over by AI -- is answered per task with the work scan from FTE TO AI. For anyone who first wants to know whether the administrative and legal side of a sale is in order, which legal matters need to be in order for a sale sets out the requirements.
Anyone who wants to know now where value is being pushed down in their own business can take the free value check: eight short questions, one per value driver, giving a picture of which driver is weighing most heavily on the price today. The full value scan, with maturity scores per driver, the owner-dependency index and a two-year calendar toward the exit moment, is under construction.