Price the brand before the route. A $400m exit just valued an asset the water dispense trade carries at zero

If I ran a bottled water dispense book in Europe, I would have the brand valued this winter. On its own, away from the route. A deal an American company signed last week is why.

A.O. Smith agreed to sell 81% of its Chinese water business for about 400 million dollars. That is roughly 358 million euros at the 27 October rate. A.O. Smith is selling the operation and renting out the badge.

Two prices sat inside one transaction

Most write-ups of the deal printed one number. There were two in it.

The operating business goes to CPE, an alternative asset manager. A.O. Smith keeps 19% of it and licenses its brand to the buyer under a long-term agreement. Counting the present value of those licence payments, the company expects total value above 500 million dollars.

So the badge on its own carries more than 100 million dollars of value.

The split works because demand never lived in the factory. A buyer of an operation pays for plant, people, stock and the customer list it already serves. A licensee pays for something older than all of it. The shopper who asks for the name was already there. One business can sell its operation to one party. It can license its brand to another the same day.

Why a royalty holds up when sales fall

China was shrinking when the deal was struck. Sales there fell 17% in local currency in the third quarter. The reporting segment that holds China saw its margin slip to 5.4%, from 7.4% a year before. This was a business of real size. It put 674.9 million dollars of sales through the group in 2025.

Work out who pays the fixed costs. A plant and a sales force get paid whether volumes rise or not. So a 17% volume fall hits profit far harder than it hits revenue.

A royalty behaves differently. It is a slice of someone else's turnover with no plant behind it. It falls when volume falls and it never falls faster. The licence is the one thing A.O. Smith kept that carries no fixed costs at all.

Dispense handles named water and never prices it apart

This trade holds the same two assets and almost never separates them. Delivered bottles of 15 and 19 litres feed what this trade shortens to BWD, for bottled water dispense. A point of use machine, POU for short, draws straight from the mains. Instant tap systems, written ITS, do that job at a counter.

The cause sits in what gets invoiced. A bottled operator sells a named water, and the customer reads that name on the bill every month. A mains-fed operator sells a service against water the building already owns. There is no product to put a name on.

That difference decides what can be licensed. A name the customer already asks for can be lent out. A company in a market your vans never reach can use it. A service contract cannot travel. It only exists where the vans go. Zenith's country models carry dispense water volume by brand and by operator. The two lists are often different companies.

Only bottled gives an operator something to license

That is the strongest card bottled holds and the trade undersells it badly. A bottled book contains an asset that can leave the route and keep earning. The water has a label and a source of its own. Neither needs a van to exist.

The label travels and the van does not. In Croatia our model puts most bottled dispense volume behind a short list of named spring brands. The businesses that deliver that water often do not own those brands. That separation is already ordinary in eastern Europe. The A.O. Smith deal has now put a public price on it.

A mains-fed estate has nothing to license. The water belongs to the building.

Two buyers exist for one bottled book

A buyer of a dispense book pays for contracted income. The rental and the delivery round are what produce it. A licensee pays for recognition instead. That is a different thing, bought with different money, and it is rarely in the same data room.

Each cash flow has a different ceiling. Contracted income stops at the number of sites a service can physically reach. Royalty income stops at how far the name is known, and that can run much further. A bottled operator with a brand worth naming has two sale processes available and usually runs one.

Mains-fed rentals run three to four times a bottled unit across the countries in our set. On a rent roll that reads like the better asset. On a licence it is worth nothing, because there is no name on the water to lend.

Our call, and the figure that would break it

We think a European dispense group licenses a water brand inside two years. It will be into a market the group does not serve itself. The structure is public now and the template is simple to copy. Roughly a fifth of the consideration in this deal is the name rather than the business.

Our own data points the same way. Bottled still carries close to nine in ten dispense units in several eastern markets. The branded water is where the volume sits.

One figure would break the call. A licence only pays where the drinker already knows the name. So watch the royalty terms between now and completion. That falls in the first half of 2027. If the terms are cut before closing, the badge was worth less than the headline said. A separately licensed name works only where the customer chose it. That is a bottled fact before it is a strategy.

P.S. Licensing a name into a market means knowing who already pours there. There is a brand table in every country file of the 2026 Zenith Water Dispense Market Reports. Pick a European market and the file is already built. Turkey and the UAE sit in the same set. Mexico, Japan and South Korea need a word from you first. Who owns which brand and who delivers it both get a row, out to 2030. You can see what a hotel buys without the office numbers mixed in. The sheet is what most teams actually use. The written study exists for whoever needs the reasoning. Built from direct interviews with local operators and data partnerships in every market we cover. Open your market's brand list