MM Packaging was valued at nine times earnings. Every water dispense owner should study what the seller kept.

If I owned a bottled dispense estate today, I would price the water apart from the machine. MM Group has just shown me why.

On 9 November it agreed to sell 70% of its packaging business to KPS Capital Partners. The enterprise value is 2.4 billion euros. MM Group then signed a long-term cartonboard supply agreement into the business it sold.

What KPS bought, and what MM kept

KPS Capital Partners takes 70% of a new holding company. MM Group keeps 30% and stays on as a minority partner. The business changing hands employs about 8,500 people. MM Group published the multiple itself, at about nine times this year's expected operating earnings. The measure behind it strips out interest, tax and depreciation. Sellers almost never print that number. Printing it tells every rival what the asset is worth.

One clause in the announcement does the real work. A long-term cartonboard supply agreement will keep board flowing from MM Group into the business it just sold. MM Group sold the network that converts and sells, and kept the business that supplies it. Set the two earnings engines side by side. The converting network earns on service, speed and customer relationships. The board mill earns on volume and on price per tonne. One name, two quite different businesses.

The three formats, and what each one makes you buy

Every dispense format bills a customer for a machine. What differs is what the operator has to buy to keep that machine working.

With a bottled water dispense cooler (BWD), the thing on the invoice every month is water. Standard sizes are 15 litres and 19 litres. A point of use cooler (POU) runs off the mains instead. Its recurring purchase is a filter and a sanitisation kit. The newest format, an instant tap system (ITS), is fitted inside the kitchen unit itself. One spout delivers hot, cold and fizzy water. Here the operator buys the unit once and sells it on.

Only one of those three formats makes the operator buy a product the customer then drinks. Follow the invoice. A bottled account generates a delivery note every few weeks. A mains-fed account generates one service visit a quarter. The first has a supplier standing behind it. Behind the second is a van and a service diary.

Inside a German bottled account the water is the bigger half

Our German model splits a bottled account into its two lines. The water is the bigger half of the monthly revenue, and the machine rental is the smaller one. That gap is wide enough to survive any rounding. It holds across the quieter sites as well as the busy ones.

Throughput decides it. Each cooler gets through enough bottles in a month to outweigh what the machine rents for. So the operator's largest recurring line is a product it buys in, rather than an asset it owns. That changes who the business is exposed to. A bottled operator carries far more supplier risk than a mains-fed one.

The same model carries a second reading, and it points the other way. A rented mains-fed unit in Germany out-earns a whole bottled account by roughly a third. Bottled throughput per machine has also been sliding for years. It sits well below where it was in 2008. Both readings are true at once, which is the honest position.

A bottled account holds two revenue lines, and one of them is sellable alone

Two revenue lines sit inside a bottled account, and only one of them needs a route. The water needs trucks, a depot and a delivery round. The equipment needs a contract and an engineer. A mains-fed account collapses into a single equipment-and-service line, so there is nothing to separate.

The split is what makes a bottled book sellable in pieces. That is the structure MM Group has just put a price on. A specialist could buy the route and the customer list. The seller could keep supplying the water into it under contract. The arithmetic is simple. A logistics buyer values a route on drops per mile. A water business values the same customers on litres a year. Two buyers, two different numbers, one account.

None of this makes bottled the easy option. Someone still has to drive the water to the door, and that costs money. Where a kitchen is already fitted out, the tap usually wins on cost and on space. The point is narrower. A bottled estate can be split and sold in two, and a mains-fed estate cannot.

Converting to mains-fed deletes the supply leg entirely

Convert a site to mains-fed water and the supply leg disappears. Nobody buys water for that customer again. The revenue does not vanish with it. It moves across into rental and service, where the cash is steadier and smaller per visit.

That trade is usually worth making. The pipe is what does it. Once the machine is joined to the building's own supply, the only recurring purchase is a filter. So the conversion swaps a variable product margin for a fixed service one. Most operators take that deal gladly, and most of the time they are right to.

There is a cost on the other side of it, though. Operators who convert hard are also shrinking the part of their business a logistics buyer would pay for. Convert everything and you end up with a single class of buyer. A mixed book has two. Almost nobody prices that difference, because almost nobody has had to.

Our read on who sells the route first

Here is where we land. We think the first European dispense deal of this shape sells the route and keeps the water. Think about the morning after completion. An operator that owns a bottling plant still has a product to supply. An operator with no plant has only the route, so selling it leaves an empty company behind.

Nine times expected earnings is the public mark now sitting on the table. That is what a buyer paid for a converting network with a supply contract attached to it. A route-only dispense book has to clear a similar bar for the split to be worth doing. Below that, the owner is better off holding both lines and running them harder.

The deal will not complete until 2027. Until then this is a structure rather than a result. Operators have that long to work out which half of each account they would rather own. The other half is the one somebody else may pay more for.

P.S. Even a well-run operator can misread a market it has never invoiced in. The 2026 Zenith Water Dispense Market Reports exist to settle the question a board asks before funding an entry: who serves this market, in which format, and at what size. Thirty-one European markets each have their own file. Whether what lands on your desk is Turkey or the UAE, the file is already written. The same holds when the market in question is Mexico, Japan or South Korea. Each file carries the unit base and the revenue behind it, out to 2030. Hospitality is split out from workplaces, so bars and restaurants never hide inside an office total. The database underneath these files is the largest in the trade. Some buyers want only the spreadsheet. Others ask for the analysis alongside it, and that is included. https://waterdispenseinsights.com/reports