A let office is not a full office.

Occupancy is the first number every operator reports and the last one I trust on its own. It tells you how full a location looks. It does not tell you how full it is, and the gap between the two can be large enough to change what a business is worth.

I learned this while leading the acquisition of another operator. The seller reported high occupancy across the private offices and was proud of it. On the face of it, with reason: almost every room was let. The P&L told a different story. When I rebuilt occupancy desk by desk instead of room by room, it was materially lower than the figure in the sale pack.

Nobody had lied. Both figures were true. They were answers to two different questions.

Two ways to count the same offices

Private offices can be measured in rooms or in desks. Counted in rooms, an office is either let or it is not, and a let office counts as full.

That works when every office is let to a team that fills it. It stops working the moment a six-desk office is let to a team of four. The room is let. Two of its desks are empty. Counted in rooms, the report still shows six.

I call this hidden vacancy: desks inside let offices that nobody is paying for. It does not appear in an occupancy figure measured by rooms. It appears only in revenue, later, and usually without an explanation attached.

How I found it

I have run locations. When a location reads almost full and the P&L does not look like a location that is almost full, something is wrong, and it is almost never the P&L.

So I stopped reading the summary and went through the agreements one by one. For each let office I set its monthly revenue against the number of desks in the room. The pattern showed quickly. A six-desk office let to three or four people. An eight-desk office let to five or six. Not every office had been sold this way, but enough had to open a real gap between the occupancy the seller reported and the revenue the location produced.

The seller had been selling rooms. The aim was to fill every room, and by that measure it was working. A team was offered whichever office was free and asked to pay only for the desks it used, at a premium price per desk. Each deal looked good on its own. Together they left a large part of the inventory unbilled, month after month.

Why the premium does not close the gap

The logic on the sales floor is easy to follow. List price is €350 per desk. A team smaller than the room agrees to €400 per desk, about fourteen percent above list. The room is let, the price per desk sold is above list, and occupancy goes up. Everyone is pleased.

Now measure what the room earns against the desks it takes out of inventory.

Every configuration was sold above list, and every one earns below it. The four-desk office let to two gives up forty-three percent of what the room should produce. Even the eight-desk office let to six earns fourteen percent less than it would have at list.

The arithmetic has a simple threshold. At a fourteen percent premium, an office needs seven of every eight desks paid for before it earns what it would at list. Below that line, the premium is a discount on the room, whatever the agreement says.

This is what makes hidden vacancy so hard to see from the inside. Every figure the sales team watches improves. Occupancy rises. Average price per desk sold rises. The only number that falls is revenue per desk built, and it is the easiest one to miss.

Who carries it

The operator carries it first, as margin that never arrives. The cost of a room does not shrink because two of its desks are empty. Rent, cleaning and energy are paid on the whole room.

An owner carries it next, wherever the asset runs on a management agreement or a revenue share. The owner is paid from revenue, and the occupancy report gives no warning that revenue is thinner than it should be.

A buyer carries it last. A business valued on reported occupancy is priced partly on desks that do not pay.

In every case the knowledge arrives late. Agreements run for months or years. By the time the pattern shows in contribution, it has been built into the location one signature at a time. Unwinding it takes about as long as it took to create. The only faster route is an aggressive price increase, and customers feel that in their experience of your service. Controlled from the first agreement, it never builds up at all.

Why it happens

Rarely through dishonesty. Occupancy is the number a landlord understands, the number an investor asks for first, and the number a general manager is usually measured on. If the incentive is to let rooms, rooms get let. The fastest way to let a room is to accept the team in front of you, whatever its size.

The single figure hides a second thing as well. Occupancy blends committed space, meaning agreements of twelve months or more, usually priced at a discount in exchange for certainty, with rolling membership at or near list price. Rolling membership carries more margin and more churn, and it depends on how well the location is run. Two locations reporting the same occupancy can be running very different businesses underneath it.

What I would measure instead

Keep occupancy by room, because the market expects it. Report occupancy by desk beside it, and revenue per desk built for every location. The gap between the first two is your hidden vacancy. The third tells you what it costs.

Then price the room. The way I have always done it starts from the average price per desk the business model needs. The room price is that figure multiplied by the desks in the room. A six-desk office is priced as six desks, whether four people move in or six. The deal can be built around that number. The number itself does not move.

Setting this up for a new location is straightforward. Correcting it inside a portfolio that already carries years of agreements is not, and that is where the real difficulty sits.

None of this is sophisticated. It takes a willingness to distrust the number everyone else takes at face value. In the transaction I described, it was the difference between the location in the sale pack and the location I was actually buying.

← All notes