Why do coffee shops fail? The recurring causes
Short answer: very few coffee shops fail because the coffee was bad. They fail because the drinks were priced from what the competition charges rather than from what they cost to make, because fixed costs were committed against optimistic volumes, because labour and cash timing were underestimated, and because there was no clear reason for a customer to walk past somewhere else to get there.
Each cause below is a mechanism rather than a statistic: how a business gets into trouble, and what the early signal looks like, so that an operator can check their own books against it. Most are survivable if they are caught. What makes them dangerous is that several produce no visible symptom until the cash runs out.
Undercosting: pricing from the street rather than from the books
This is the deepest of the causes, because it makes every other problem harder to solve.
The usual method for setting a price is to look at what the shops nearby charge and land somewhere in that range. It feels like market discipline. It is the adoption of somebody else's cost base — their rent, their milk contract, their staffing model, their equipment, their volume — as the basis for your own prices.
If your cost per drink is higher than theirs for any of many ordinary reasons, you have set a price that does not cover your own costs: the business is unprofitable at every volume. Selling more drinks makes the loss larger, because every additional customer consumes coffee, milk, a cup and staffed time and returns less than those cost.
Operators in this position almost always diagnose it as a volume problem, because that is what it feels like from behind the counter. They market harder, extend hours, add a loyalty offer — all of which increases throughput and accelerates the loss.
The fix is not complicated but it does take an afternoon: cost the actual drinks you actually sell, line by line, including the things nobody counts — purge shots, dial-in waste, milk poured away, remakes, the cup and lid. How to cost and price a cup of coffee properly sets out the method. Until that is done, no pricing decision is anything more than a guess dressed as a benchmark.
Rent and fixed-cost load: committing to your best month
The second recurring cause is a lease signed on the strength of a forecast.
The mechanism is simple and unforgiving. A prospective operator models the business at the volume they believe it will reach, finds that the rent works at that volume, and signs. The volume arrives later than expected, or never quite arrives, and the rent is due regardless from the first month.
Rent is only the most visible of these commitments. Business rates or local taxes, insurance, utility standing charges, equipment finance, a service contract, waste collection, software subscriptions and the base rota all behave the same way: indifferent to trade.
Two consequences follow:
Fixed costs do not fall on quiet days. A wet Tuesday in February costs almost exactly what a busy Saturday in July costs to keep the doors open. Only the variable lines — coffee, milk, cups, and the marginal hours you can actually flex — move with trade. This is why quiet periods do damage disproportionate to their share of the year.
The break-even volume is a fact, not a target. Every business has a daily number of drinks below which the day loses money, and many operators have never calculated theirs. It comes entirely from your own books, and it tells you whether the premises you are considering is viable at realistic trade rather than at hoped-for trade.
The honest test before signing a lease: model the business at a volume clearly below your expectation, not at it. If the lease only works when the forecast is met, the forecast is now a fixed cost too.
Labour: underestimated in two directions
Labour gets underestimated twice, and the second one is the one people miss.
The cost of the hours. The true cost of an hour of staffed time is not the hourly wage. It includes employer contributions, holiday, sick cover, any pension obligation, training time, and the hours that are staffed but not productive — opening, closing, cleaning, the quiet afternoon that still needs someone behind the counter. A rota built on the bare wage rate understates what staffing the week costs.
The recruitment and training burden. Hospitality turnover means the same role gets filled repeatedly. Each cycle costs advertising or agency time, the manager's hours spent interviewing, the trainer's hours, the trainee's unproductive hours, and a period of below-standard output while someone learns. That is not an annual overhead — it is a cost that recurs every time somebody leaves, and it lands on whoever is already busiest.
The operators who get into difficulty here are usually not paying too much per hour. They are running a rota that assumes full staffing, no absence and no vacancies, and then absorbing every gap through the owner's own unpaid hours. That works until the owner runs out.
Specification errors: equipment sized to the wrong hour
Equipment mistakes rarely close a business on their own, but they quietly remove the revenue that would have made everything else work.
The recurring error is sizing to average demand. Average demand is a comfortable number — total drinks divided by trading hours, which almost any machine can meet. But a coffee shop does not trade at its average. It trades in peaks, and the peak is where the money is.
A bar that cannot clear its busiest hour loses that hour every single trading day. The customers who see the queue and keep walking do not appear anywhere in your figures; they are invisible revenue, and their absence gets misread as low footfall. Worse, the busiest hour is precisely the one whose customers are most time-constrained and least willing to wait.
The related errors are the same mistake in different clothes: one grinder where the menu needs two, a milk workflow that becomes the bottleneck once most orders are milk drinks, a layout that makes staff cross each other at peak, and a machine that has to leave the counter for routine service and takes trading hours with it.
Size to the peak you actually have, measured rather than estimated — how to size a coffee machine to peak demand sets out how. And consider the whole cost of the equipment decision rather than the acquisition figure alone: the total cost of ownership of a commercial coffee machine.
No differentiation: being one of several similar offers
A great many coffee shops are, from the customer's point of view, interchangeable with two others within a few minutes' walk: similar menu, similar prices, similar room, similar quality.
When nothing distinguishes the options, the customer picks on convenience and habit — the nearest one to their route wins, and everyone else lives on the remainder. That is fragile, because it leaves you exposed to a new opening, a changed bus route or an office moving, and gives nobody a reason to make an effort to reach you.
Differentiation does not require being remarkable. It requires having a specific answer to why would someone walk past another coffee shop to come here? Quality is one valid answer. So are speed, the room itself, the people behind the counter, the food, the opening hours, or being genuinely the most convenient option for a defined group of people. What does not work is having no answer and hoping for footfall. What makes a coffee shop succeed takes this from the other direction.
Cash management: profitable on paper, insolvent in practice
Timing closes more businesses than margin does.
A coffee shop can be genuinely profitable across a year and still fail in a particular fortnight, because profit is an accounting result and cash is a calendar. Rent quarters, insurance renewals, tax payments, an equipment repair and a supplier account all fall on dates that have nothing to do with when your customers happen to come in. Seasonality does the rest: many venues have predictable troughs, and a trough that coincides with a large payment date is where businesses die.
The failure pattern is recognisable. Trade dips. A payment is deferred. Stock is bought in smaller, more expensive quantities because there is no cash to buy properly. Maintenance is postponed, which later produces a larger bill. Staff hours are cut, service slows, and the customer experience declines exactly when the business can least afford it. Each step is a rational response to the previous one, and together they form a spiral.
What prevents it is unglamorous: a rolling cash forecast covering the next several weeks, updated weekly, listing every known payment date against expected takings. It is a spreadsheet, it costs nothing, and it converts an emergency into a decision made in advance.
Owner dependence: a business that cannot run without one person
If the business only works when one particular person is present, two things are true, and both are problems.
It is fragile. Illness, family circumstances or simple exhaustion take out not an employee but the whole operating system. Standards, supplier relationships, recipes, opening routines, the knowledge of which regular takes what — if these live in one person's head, they leave the building when that person does.
It is close to unsellable. A buyer purchasing a business that depends on the departing owner is purchasing very little. That matters even to operators with no intention of selling: years of work build no transferable asset, and the owner has bought a demanding job rather than a business.
Owner dependence is rarely a decision. It accumulates, because doing a task yourself is always faster than teaching it, and because in the early years there is nobody else. Reversing it means writing things down, training a second person to run a full day unsupervised, and then actually being absent — which is the part most owners skip.
Drifting standards: the decline nobody decides on
The last cause is the quietest. No one chooses to lower standards. They erode.
The grinder goes a little longer between calibrations. The milk gets textured more roughly during the rush and it is fine, so it stays that way. Tables get wiped less carefully. The cleaning routine gets abbreviated on a short-staffed night, and the abbreviated version becomes the routine. A new starter is trained by someone who was trained by someone who had drifted.
Each step is imperceptible, which is the danger: nothing is ever bad enough to trigger a response. Customers do notice at some threshold, and their reaction is not to complain but to stop coming and never say why. By the time the numbers show it, the cause is months in the past.
The countermeasure is a standard that exists outside anyone's memory — a written specification for the drinks, a cleaning schedule that is signed rather than assumed, calibration at a defined interval, and someone tasting the coffee every day with the specification in front of them. Equipment can help here, since recipe-driven production holds a dose and a recipe whether or not anyone is watching, but the discipline is the point rather than the machine.
Is the trade structurally difficult, or are operators making avoidable mistakes?
Both, and it is worth separating them.
Barriers to entry are low. Opening a coffee shop requires no professional qualification and a fit-out modest compared with a restaurant. That is genuinely good — it is why the trade is open to people without capital or connections. It also means any street with demand attracts supply until the returns are ordinary. A crowded market is not evidence you did something wrong; it is the predictable consequence of a business anyone may open.
The margins are structurally thin and the fixed costs are structurally high. A venue with a rent, a rota and a coffee machine has committed most of its cost base before it sells anything. That combination punishes both undercosting and quiet periods more severely than it would in a business with lower fixed costs.
Passion for coffee is not a business model. Caring about extraction is a real advantage — it produces a better product and it is a reason for people to choose you. It is not a substitute for a costed menu, a survivable lease, a cash forecast or a rota that works. A great deal of the failure in this trade involves people who were extremely good at the coffee and had never calculated their break-even day.
What buying better equipment will not fix
Better equipment does not fix an undercosted menu. It does not fix a lease the business cannot service. It does not create differentiation on its own, and it does not manage cash. If any of those is your actual problem, capital expenditure makes it worse, because it adds a fixed commitment to a business that already has too many.
What equipment does is narrower and worth having: it removes the peak-hour constraint if that is what limits you, lowers recurring waste where waste is a real line, raises the quality floor when skilled staff are not on shift, and reduces trading hours lost to service if it can be maintained in place. Real gains, and the reason to specify carefully. Not a rescue.
The order of operations is: know your costs, know your break-even, know your constraint. Then buy the thing that relieves the constraint.
Frequently asked questions
Is undercosting really more dangerous than low sales? It is more insidious, because low sales are visible and undercosting is not. A shop with low sales knows it has a problem. A shop priced below its own cost sees busy periods and still runs out of money, then usually responds by trying to be busier — which deepens the loss. Costing the menu is the check that distinguishes the two.
How do I know whether my equipment is actually holding the business back? Watch your busiest hour with a clock. Count who arrives, who leaves without ordering, and where the queue stops moving — grinder, group, milk, till or hand-off. If the constraint is not in production, more machine capacity will not help.
Can a coffee shop survive on a high rent in a busy location? Sometimes, and the test is arithmetic rather than judgement. A prime site can carry a high fixed cost where the volume is genuinely there. The failure comes from paying prime rent on the assumption that volume will follow. Model the site below your forecast; if it only works at the forecast, the position is speculative.
We're profitable but always short of cash. Is that normal? It is common, and it is a timing problem rather than a margin problem. Build a rolling forecast of the coming weeks with every known payment date against realistic takings, updated weekly. Most cash crises are visible weeks ahead in that document and invisible in a profit and loss account.
What single review would tell me most about my own risk? Sit down with your own accounts and produce three things: the fully costed price of your five best-selling drinks, your break-even number of drinks per day, and a list of every cost that will be charged next month whether or not anybody comes in. Those three answers locate you against most of the causes above.
This article is part of the Vea Group knowledge base. Vea Group S.p.A. is an Italian designer and manufacturer of professional and premium coffee machines, with heritage dating to 1919 and production in Chignolo d'Isola, Italy, and Suzhou, China.