Before you shortlist a water cooler market, count the columns on your screen
By Zenith Water Dispense Team ยท
Most water dispense entry decisions rest on two columns: market size and penetration. Neither says what the revenue is made of.

If I were choosing the next water dispense market to enter, I would not open with its size. I would open a blank sheet and argue about the column headings first. Get those wrong and the shortlist is already wrong.
Germany gave me a clean illustration this year. Its mineral water sales, including the regulated Heilwasser category, reached 10.2 billion litres in 2025. That is a rise of 2.4%, or 128.8 litres a head. The Verband Deutscher Mineralbrunnen published the figures on 27 January 2026. The headline reads as steady growth. Inside it, still water grew 6.1%, faster than the category itself. One number said the market was fine. A second number said the market was moving.
What a market screen usually holds
Most screens I am shown hold two columns. Market size, then penetration. Both are real numbers and both are easy to find. Neither one tells you what the revenue in that market is made of. Size and penetration describe how many machines a market could hold. They say nothing about what one machine earns.
Two markets the size column calls identical
Take Italy and France. On total dispense revenue in 2025 they sit within a few per cent of each other. On a size screen they are the same opportunity.
They are not the same business. Italy's revenue is led by bottled water dispense (BWD, coolers fed from 15 or 19 litre bottles). France's is led by point of use (POU, water coolers plumbed into the mains). Italy has a working instant tap market, worth well over a tenth of its dispense revenue. In France the instant tap and sparkling segment is a rounding error beside it. ITS means counter-top and undercounter units. They deliver boiling, chilled or sparkling water on demand.
An operator who wins in Italy wins on route density and bottle logistics. An operator who wins in France wins on install capacity and plumbing. The cost bases do not resemble each other. A size column cannot tell them apart.
Four columns that change the answer
Rental against outright sale comes first. Rental mix decides when a cost saving reaches the profit line. A sales-led segment books the margin this quarter. A rental-led fleet buries the same saving inside depreciation for years.
Bought-in water as a share of revenue comes second. Where bought-in water dominates revenue, the operator is a distributor wearing an equipment badge. Where that share is lowest, the operator mostly rents hardware. Buyers price those two businesses on different multiples.
Specification rate comes third. Sparkling capability is the clearest example, and it moves by country rather than by format. In several converted western markets the mains-fed fleet carries sparkling more often than the tap fleet does. Germany is the sharpest case of that inversion. A market can look premium and still specify cheap.
Throughput per unit comes fourth. Throughput separates markets more sharply than penetration does. Litres through one machine each month is the column that predicts whether a bottled book is worth buying. It is also the one most screens leave out.
Bottled is the cheapest way to be wrong about a market
There is a reason to keep a bottled option on the screen even when the plan is conversion. A bottled install is the only way to test a market and walk away whole. Nothing is plumbed. Nothing is sunk into a wall. If the site or the country does not work, the cooler comes back on the next van. It then goes somewhere else and earns again. A plumbed install does not travel. That makes bottled the low-risk instrument for an unproven market. That is a strategic use of the format. Legacy has nothing to do with it.
Build the screen first, then rank
Build the screen before you look at any market. Put those four columns next to size and penetration. Then rank, and watch which markets move. The candidates that fall once composition is added are the ones a two-column screen would have sold you.
Investors can ask the same question in reverse. If a target's own market screen holds two columns, that is a finding about the management team. A buyer can read a company's strategy off the spreadsheet it uses to pick its next market.
What this sets up for 2027
The next two years reward this work. Regulation is landing unevenly across Europe, and format mix decides where it bites hardest. An operator planning a 2027 entry needs the composition columns. Scale on its own will not carry that decision. So does an owner planning a 2027 sale, because the buyer will arrive with the better sheet. Building the screen costs an afternoon. Getting it wrong costs a market.
🤝 Have a market question that does not fit a template?
Send it to us in plain words, however rough. We will come back with what it would take to answer it properly, and what that would cost.
P.S. Column headings are easier to argue about when somebody has already built the sheet. Revenue composition is modelled country by country in the 2026 Zenith Water Dispense Market Reports. Rental against sale, the water line, specification rates and throughput each carry their own row. Ask for any European country and the file exists. Five more sit outside Europe: South Korea, Japan, the UAE, Turkey, Mexico. Operators are named and their shares sized, with the forecast carried to 2030. Teams wanting the narrative as well as the numbers get both under one licence. Built on the world's largest water dispense database. See what your screen is missing: https://waterdispenseinsights.com/reports