Written for the operator putting their own money at risk: the one who buys the machines, restocks them, absorbs the breakdowns and collects whatever they sell. A vending operator moving upmarket, a roaster looking for a channel of their own, a service engineer going independent, or a small investor. Five business archetypes modelled with the same economic engine at 2026 Spanish costs, with the owner's salary paid before the cash flows are discounted and their own hours counted as route capacity, not as free labour.
A decision report, not an overview of the vending trade. Every published figure comes out of a single economic engine that is executed and validated, and every decisive statement carries its rule, its source and its date — including the gaps that could not be closed, which are declared as gaps rather than filled in.
This is the fracture that organises the whole report. A machine serving twenty cups a day and one serving sixty cost exactly the same to restock, and restocking is not a support function: it is the cost that does not fall with volume. That is why the report separates, from page one, four pairs of quantities the trade publishes mixed together — installed fleet versus productive fleet, cost per cup versus cost per visit, contribution before route versus after route, technical uptime versus commercial availability — and never merges them again. Once you see why a machine can throw off positive contribution while the operation loses money, you have the report in your head.
Scattered small and mid-sized offices, a concentrated large corporate headquarters, a high-traffic public site awarded by tender, a premium coffee corner, and the employer-paid service. All five are solved with the same discount rate, the same cost build-up and the same horizon, so their results are comparable with each other. The report determines which of them remunerates the capital and which amounts to buying yourself a job, how much each lever weighs in the sensitivity tornado, and at exactly which fleet size the result changes sign.
The trade still repeats that a cup out of a machine costs «a few cents». That was true in 2019. The report rebuilds the cost line by line — premium beans including grinder loss, milk powder, cup, stirrer, water, filter and imputed electricity — at Spanish B2B list prices, then carries it through to real contribution by deducting VAT, the site commission and the card-payment fee. Two uncomfortable facts surface: how much the packaging weighs against the coffee itself, and that the commission taken by whoever provides the space costs more than the beans.
The weighted scorecards used in the trade produce «7 out of 10» sites that destroy margin. The report replaces them with thresholds you can apply on Monday morning, in two distinct families: an economic one that depends on the price the location lets you charge, and a freshness one that does not depend on price at all and that almost nobody measures — the one that turns the criterion into an exclusion rather than a margin adjustment. With the full conversion chain from the building's headcount to the cup actually served, and the four field questions to ask on the site visit, because none of them appear on any property listing.
The legal chapter is written so your adviser can check it in a minute: every statement carries its rule and its article. And it reaches a conclusion that contradicts what machine suppliers and trade consultancies publish: there is a registration the route operator does not need and that many pay for, and a regulation repealed in 2022 that the trade still cites. What does bite — allergen labelling, charging separately for the cup, the IAE business tax heading with its two rates, 10 % IVA on coffee against 21 % on soft drinks, and the verifiable invoicing obligation — enters the engine with its cost attached.
In hospitals, universities and government offices the operator does not set the price: it is written into the tender, with the published, dated cap list the report reproduces. And that channel does not take a percentage of sales: it charges a fixed fee per machine, bid upwards in the tender itself, which a weak site pays exactly as a strong one does. The report models that combination — a ceiling on revenue and a floor on cost in the same document — and sets out what it takes to make it work and what happens in the year it does not.
This report publishes no savings figure for telemetry, and explains why: the percentages circulating in the trade are not in the sources they are attributed to. That was checked by opening those pages in full. What it publishes instead is the trail of where each figure breaks down, the real cost of telemetry and cashless per machine per month, and the maximum size of the prize at stake, so that anyone about to sign a monthly fee knows what promise they are paying against. Including the tax trap that can drag an entire route into a regime that does not apply to it today.
Any investment is judged against its alternative, not against zero. Two of them here are not variants of the plan but refutations of it: the free-loan contract, under which a roaster hands over the machine at no cost in exchange for a supply exclusive — verified across four players with their published thresholds — and the market in going routes, which sells you takings already built for less than new hardware costs. The report quantifies both before resolving the lesser trade-offs, because either one on its own shifts the recommendation.
Several decisive parameters of this business are simply not published in Spain: how many stops a route driver makes in a day, how many cups a machine actually serves, what share of the menu contains milk, and — the one that weighs most — how many of their own hours the owner puts in. The report designs a six-month pilot that is not built to make money but to produce those four numbers of your own and replace the model's assumptions with them, with the decision rule for each outcome and the exact owner-workload threshold above which the verdict changes sign.
Download a free sample with the cover and the full table of contents of the report.
The sample includes the cover and the contents, so you can see the full scope of the report. The analysis, the economic model and the verdicts are in the complete edition.
For whoever will run the route at their own risk: you buy the machines, sign the site agreements, restock, absorb the breakdowns and collect whatever the machine sells. The typical profile is a small business or an entrepreneur deciding alone, with no analysis department behind them: a vending operator moving upmarket, a roaster looking for a channel of their own, a service engineer going independent, or a small investor. It is not a vending manual or a sector overview: it is the calculation of whether that decision returns the money.
A scattered fleet of small and mid-sized offices served by tabletop machines; a concentrated large corporate headquarters with floor-standing equipment; a high-traffic public site awarded by tender; a premium coffee corner with few units and top-of-range equipment; and the subsidised service where the employer pays and the employee does not. Five, because these are the five configurations in which the same business produces different results with the same demand and the same prices. All five are solved with the same engine, the same discount rate and the same horizon, which is what makes them comparable.
It gives a verdict per archetype, with the figure in front of it, applying four cascading thresholds that are published before the results so you can check they were not tuned after the fact: not viable by structure, self-employment, viable subject to a named condition, and viable outright. The report separates the business that sustains itself and pays whoever runs it from the one that also returns the capital, and does not soften the difference. It also states what would turn each provisional verdict into a firm one.
In both directions, and that is one of the report's axes. The owner is charged a market salary at full employer cost before the cash flows are discounted — because the most expensive mistake in this trade is counting the owner's work as free and publishing positive NPVs for businesses that are really just a job — while at the same time being credited as the route's first restocking capacity, with a declared workload, so that hiring only starts above it, and in fractions of a full-time role rather than whole-person steps. The report calculates the exact workload threshold above which the result changes sign.
It includes the recalculable model in CSV and JSON — assumptions and results — plus the traceability register with all 149 assumptions it uses, each with its value, its unit, its source, its consultation date and its confidence grade, with the weak ones flagged as such. The sensitivity tornado tells you which variables actually move the verdict and over what realistic range they have to move, and the fleet and price matrices give you the exact point at which the decision flips.
Yes, and you are probably the reader who gets most out of it, because you can replace the model's assumptions with your own numbers and rerun the arithmetic. The report gives you the cost build-up updated to 2026 prices, the cost-per-visit arithmetic to work out which sites in your fleet are subtracting, the acceptance and rejection thresholds for the next ones, the new-versus-refurbished analysis, and the comparison against the two alternatives that compete with buying hardware: the free-loan machine and buying a going route.
Scope is Spain, with data closed in August 2026 and Spanish B2B list prices from that year. The legal chapter carries each statement with its rule and its article so your adviser can verify it in a minute: the food business register and the retail regime, the IAE business tax heading with its two rates, VAT rates by product, allergen and payment-method obligations, verifiable invoicing and the free-loan contract regime. Several of those conclusions contradict what the trade itself publishes, which is why each comes with the text quoted in full.
A 183-page PDF, downloadable immediately after purchase, under a single licence, plus the model in CSV and JSON. Available in Spanish, English, German and French. You can see the cover and the full table of contents in the free sample before deciding.
Five archetypes each with its own verdict, the real cost build-up per cup at 2026 prices, the thresholds for accepting or rejecting a location, the legal framework verified rule by rule, the public channel with its capped price and its fixed fee, and the model delivered in CSV and JSON so you can redo the numbers with your own assumptions.
Buy the report · €990