01 / Read the unit
Here is a quick way to understand any building without reading its business plan. Ask one question: what does it sell, and in what unit?
A ten-year office lease needs a lawyer and a good memory. A hotel night needs a front desk, a housekeeper and a revenue manager who reprices the room before breakfast. A coworking desk by the month needs a community manager and a coffee machine that never breaks. A parking space by the minute needs an app and nobody at all.
Same square metres. Completely different machines.
That is what an operating model is. Not a mission statement and not an org chart. It is the machinery a building needs to keep reselling its unit. The smaller the unit, the bigger the machine.
02 / Forty years of shrinking
Commercial real estate has spent two generations cutting its unit into smaller pieces.
In Britain the standard institutional office lease used to run twenty-five years, with rent reviews that only went up. Then in 1989 a man named Mark Dixon opened a business centre in Brussels and sold offices by the month, and later by the day. Regus became IWG, with thousands of locations in around 120 countries. WeWork took the desk and sold it by the month to people who had never signed a lease. Then came the day pass. The office went from a decade to an afternoon.
Retail went from the fixed lease to turnover rent to the pop-up that pays by the week. Parking went from the annual permit to the app that bills by the minute. Storage went from the warehouse to the box. The gym went from the club fee to the class. The hotel was always the night, and now there are platforms that sell the room by the hour, for reasons that are not always about sleep.
Every time the unit shrank, the same three things happened.
The customer started deciding more often. A ten-year tenant chooses you once. A hotel guest chooses you every night.
The owner started carrying the empty time. With a long lease, vacancy is the tenant’s problem until the break clause. With a night, every empty room is yours by eleven the next morning.
And the building needed someone to run it. You can own a long lease from a spreadsheet in another country. You cannot sell a night without a person awake at three in the morning.
Smaller units are more work, more risk, and more information. They are also, when they work, more money per square metre, which is why capital keeps cutting.
03 / The exception
Now look at housing.
In 1865 Octavia Hill’s workers collected the rent in London every week, at the door. In 1894 Johanna ter Meulen brought the method to the Jordaan, and her woningopzichteressen did the same: weekly rent, weekly visit, weekly conversation about the leak and the job and the debt. The unit of sale in Dutch social housing was the week.
Then the week became the month. The month became the open-ended tenancy. And on 1 July 2024 the Netherlands made the permanent contract the norm again by law, with temporary contracts allowed only for a few groups such as students. The same day, the Wet betaalbare huur brought rent regulation to the middle segment.
So while every other building type cut its unit into smaller pieces, housing made its unit as large as a unit can be: indefinite, and in much of the market, price-capped.
I want to be clear that this is not a complaint. Security of tenure is the whole point of a home. Nobody should have to re-buy their kitchen every Monday. Of all the units in real estate, the home is the one that should be large.
But follow the mechanics, because they do not care about our intentions.
The largest unit in real estate is sold to the person who uses the building most. A hotel asks its guest every night whether it is still worth it. An office landlord gets asked every five years or so. A housing block asks its resident once, at the door, and then hears the answer at the exit.
Housing is the only business in which the customer feedback form is a removal van.
04 / What the big unit does to a building
A large unit is not just a contract. It quietly designs the whole operation.
Because the resident decides rarely, nobody has to win him back often, so nobody is employed to try. Because the vacancy risk sits in long gaps between tenancies, the operating budget is built for maintenance, not for use. Because nothing is resold in small pieces, nothing is measured in small pieces. Ask a hotel what its lobby earned last Tuesday and it can tell you. Ask a residential owner what the common room did last year and you get the cleaning invoice.
And then there is the most revealing space in any housing scheme: the one with no unit at all.
The roof terrace. The shared lounge. The guest room nobody can find the key for. The bike storage that became a bin for things nobody wanted to throw away. These spaces were drawn with real ambition and real money. But nobody sells them in any unit, so nobody resells them, so nobody runs them, so nobody knows whether anyone uses them. A space without a unit of sale has no operator. A space without an operator becomes furniture.
That is where buildings start to decay. Not in the apartments, which have a unit and a tenant who cares. In the square metres that belong to everyone and are sold to no one.
05 / Keep the home big. Make everything around it small.
So here is the operating model I think residential buildings are heading towards, whether or not anyone plans it.
Two layers.
The first is the home. Indefinite, protected, regulated where it should be. A large unit, on purpose. That layer is settled, and it should stay that way.
The second is everything around the home, sold in the smallest unit that makes sense. The guest room by the night. The workspace by the day. The shared car by the hour. Storage by the box. Laundry by the cycle. The chef’s kitchen by the evening. The event by the ticket. Care by the visit, for the residents who will need it.
The point of the small units is not the revenue. In the regulated segment service charges have to reflect cost, so this is not a margin play there, and anyone who pitches it as one should be sent back to the lawyer. The point is what small units do to a building. They make residents decide often again. Every booking is a vote. Every empty slot is a signal. And a building that sells something every day needs someone on site to sell it, which is how the operator comes back into housing without anyone having to argue for the word.
The big unit gives the resident security. The small units give the building a pulse.
06 / The unit map
Here is the instrument. It takes an afternoon and a set of drawings.
Take the plans of any residential scheme, built or planned. Next to every space that is not a private home, write four things. What is the unit of sale. Who decides to use it, and how often. Who carries the empty hour. Who is employed to resell it.
You will find three kinds of space. Spaces with a unit and an operator, which work. Spaces with a unit and no operator, which work until the first thing breaks. And spaces with no unit at all, which are a liability with a nice render.
For every space in the third group there are only two honest options. Give it a unit and someone to sell it, or make it smaller and give the square metres back to the homes.
The risk is real and it is the owner’s. Small units expose bad ideas quickly. The coworking corner nobody books, the guest room that sits empty, will show up in the numbers within a quarter instead of hiding in a service charge for a decade. That is not a flaw of the model. That is the model.
07 / The pulse sheet
The unit map tells you what every space is. It does not tell you whether anyone wants it. For that you need what the hotel has: a way to read choice.
Small units give you that for free, because every small unit leaves a record. A booking. A key swipe. A ticket. You do not have to ask residents how they feel about the guest room. You can see whether they chose it, whether they came back, and how many of them did.
I learned this in supermarkets before I learned it in buildings. Nobody in retail cares much what you say about the yoghurt. They care whether you buy it again.
So here is the framework. Four numbers per space, per month. One number per building, per year. It fits on one sheet.
Uptake. Of all residents, the share who used this space at least once this month. It tells you whether the space is relevant to the people who actually live here, rather than the people in the render.
Return. Of the residents who used it last month, the share who used it again this month. It tells you whether the experience was worth repeating. Uptake can be bought with an opening party. Return cannot.
Reach. The number of different residents who account for half of all use. If five people make half the bookings, you do not have a shared space. You have a private club with a service charge.
Idle. The share of available hours nobody used. The empty hour from section 02, finally with a number on it.
And once a year, for the whole building:
The van test. Split the residents who moved out or renewed this year into two groups: those who used at least one small unit in their last twelve months, and those who never did. Compare how long each group stayed. If the people who chose the building every week also stayed longer, the small layer has earned its line in the IC pack. If they did not, the layer is decoration, and you found out for the price of a spreadsheet instead of a decade.
The sheet reads itself in four patterns.
High uptake, low return: a good idea, badly run. Fix the operation, not the space.
Low uptake, high return: a niche that works. Keep it small. Do not scale it to please a brochure.
High return, narrow reach: a clique. Change the hours, the access or the host.
High idle everywhere: too much shared space. Give the square metres back to the homes.
Three rules keep it honest. Measure groups, never individuals: the sheet is about the building, not about who used the sauna on Tuesday. Use the first six months as the baseline, because the only fair benchmark for a building is the same building last quarter, and only after that the rest of the portfolio. And show the sheet to the residents. A hotel shows you its rating. A building can show its residents its pulse.
What the sheet does not measure is everything that happens without a booking. The conversation in the lift. The neighbour who notices the post piling up. For that, add one question a year: would you recommend living here to a friend? Behaviour first, opinion second. That order is the whole point. For decades housing has measured experience by asking. The small units let it measure by watching what people choose.
08 / What the hotel knows
A hotel learns three hundred and sixty-five times a year whether its rooms are any good. It is not smarter than a housing block. It just sells a smaller unit.
Keep the home exactly as large as it is. Cut everything around it into pieces small enough to be chosen every day. Then read the choices.
The building will start telling you the truth. Most owners have never heard it speak.