Common barista mistakes — and how to fix them
Short answer: the mistakes that cost a bar most are rarely the exotic ones. They are not purging the group, distributing unevenly, dosing by eye, chasing a stopwatch instead of the taste, adding air to milk too late, overheating it, letting shots sit, working out of sequence, and blaming the machine before checking the coffee, the grinder and the water. All of them are free to fix.
Each entry follows the same shape: the mistake, what it does, how to tell, and what to do instead. The mechanisms behind the puck items sit in puck preparation, distribution and tamping and espresso channelling explained; this piece is about spotting and correcting behaviour.
What goes wrong before the coffee is even in the basket?
Not purging and wiping the group between shots. After a shot the shower screen holds spent grounds and the group holds cooled, standing water. Both end up in the first part of the next shot — the part carrying most of the flavour — giving a muddied cup with a stale, faintly ashy note nobody traces to a variable, because it isn't one.
How to tell: run a finger over the screen after a shot. If a spent puck knocks out with loose grounds on top, they came from the screen.
The fix: short flush, wipe screen and gasket, then dose. The highest-return habit on this list.
Inconsistent distribution before tamping. Coffee arrives as a mound: dense under the chute, sparse at the walls, clumps throughout. A tamp compresses that arrangement without rearranging it, so the density difference survives into the shot and water finds the loose parts. This — not tamp pressure — causes most channelling.
How to tell: erratic or split streams, a shot that starts fast and runs pale, craters or tunnels in the spent puck, shot times that do not match the taste.
The fix: something that moves coffee sideways before the tamp — levelling, a needle stir, a settle-and-tap — done the same way every time. Consistency matters more than which method.
Obsessing over tamp pressure. A level tamp matters. Force, beyond enough to consolidate the bed, does not: past the point where coffee stops compressing, more pressure changes nothing except the barista's wrist — while the limiting variable, distribution, goes unattended.
How to tell: baristas comparing how hard they tamp; wrist complaints; a tilted spent puck surface, a levelness fault no force corrects.
The fix: tamp level, tamp the same, and move the attention to the step before.
Not dosing by weight. Scoops and eyeballing return the largest variable in the shot to chance. Ground coffee's bulk density changes with grind setting, bean density, roast level and how it settles, so the same scoop is a different dose from one bag to the next.
How to tell: the grinder is "correct" but shots wander with no other change; two baristas on one setting make different drinks.
The fix: a scale on the bar, dose to a target, checked through service rather than only at dial-in.
Where does dialling in go wrong?
Chasing time instead of taste. A stopwatch target is a diagnostic reading, not the objective. Treated as the objective it produces baristas who adjust grind until the timer lands where training said it should, then send out coffee they have not tasted. Time is a proxy for flow resistance, not for flavour: the same time can come from a correct shot or a badly channelled one.
How to tell: someone can state the shot time exactly and cannot say what the coffee tasted of. Or a bar whose shots all land inside the window and whose coffee is inconsistent anyway.
The fix: taste first, then read the clock to explain what you tasted. Debunking the ten-second espresso myth covers why timing folklore persists, and can you dial in grind size by taste alone covers how far the opposite approach gets you.
Re-dialling for the wrong reason. Shots change and the reflex is to move the grinder, but often the grinder is not what changed. Coffee ages and degasses, altering flow. Water changes when a filter cartridge exhausts. The burr path fills with fines, changing both particle distribution and how freely grounds leave. Adjusting grind for any of those hides the cause and moves you further from a repeatable setting.
How to tell: a grind setting that has drifted steadily one way over weeks; a shot that behaves oddly first thing and settles later; coffee sitting in the hopper.
The fix: before touching the grind, check three things — the coffee's age, when the grinder was last cleaned, and when the water treatment was last serviced. Keeping coffee fresh in the hopper and grinder maintenance and calibration schedules cover the two most common of the three.
What are the recurring milk mistakes?
Steaming in the wrong order. Air goes in at the start, while the milk is cold. Introduce it once the milk is hot and there is no time left to break the bubbles down and fold them into the liquid: stiff foam sits on hot milk rather than forming one textured mass, and swirling does not recover it.
How to tell: large visible bubbles, foam separating in the jug, latte art that will not pour because the texture is in two layers.
The fix: air first and briefly, then bury the tip and spin to temperature. The complete guide to steaming and texturing milk sets out the movement; when foaming milk, timing is everything covers why the window is short.
Overheating. Scalded milk is finished: it cannot be corrected, blended back with cold milk, or re-steamed, because the proteins that hold structure are already denatured. Every overheated jug is product paid for and thrown away, plus a remake.
How to tell: a cooked, faintly sulphurous smell; foam collapsing seconds after pouring; a jug too hot to hold; a bar pouring milk away routinely without counting it.
The fix: stop by feel on the jug rather than by watching, and pour immediately. If milk goes down the drain regularly it is a measurable cost line — milk waste in a busy café covers sizing it.
Not purging and wiping the wand immediately. A hygiene failure before a quality one: milk residue on a hot wand is a growth medium, and dried milk in the tip is not removed by a purge later. It becomes a performance problem too, because a partly blocked tip delivers uneven steam and the barista compensates by changing technique.
How to tell: crusted residue at the tip, steam that spits or pushes sideways, a wand scraped at close.
The fix: wipe with a dedicated damp cloth and purge every time, before the jug leaves your hand.
What gets lost between the group and the customer?
Letting the cup and the drink diverge. A cold cup pulls heat out of an espresso; a cup straight from the glasswasher adds water to it; the wrong-sized cup changes the coffee-to-milk ratio and the balance of the drink. All three are decided before the shot is pulled and none appears in the recipe.
How to tell: espresso that tastes flat and cools fast; drinks that vary with which shelf the cup came from.
The fix: one cup per drink, warm and dry. Should you preheat your cups covers what this is worth and where it stops mattering.
Serving a shot that sat. An espresso left while the milk is steamed is not the drink you made: crema collapses, the surface oxidises, temperature falls, bitterness comes forward. A milk drink masks some of that; a straight espresso masks none.
How to tell: the sequence itself. If shots are pulled before the jug is picked up, you are serving old espresso.
The fix: steam first or in parallel so the shot lands last. For a straight espresso, pull it when the customer is ready for it, not when you are.
Working out of sequence under pressure. A workflow mistake rather than a technique one, and the one that loses most quality per hour. Under pressure people revert to finishing one drink before starting the next, idling the group and the wand alternately — or start everything at once and lose track of which shot belongs to which cup.
How to tell: the queue stops while the barista does something serial; remakes; a bar fast when quiet and slow at peak.
The fix: a fixed order of operations for a round of drinks, drilled until it survives stress. Maximising barista workflow covers building one for a specific bar.
Not tasting your own coffee during service. A bar drifts a long way in a day: coffee warms, the grinder heats and its output shifts, the hopper refills from a different bag, the filter passes a threshold. Nobody notices, because nobody is drinking the coffee.
How to tell: if you cannot say what today's coffee tastes like, this applies to you.
The fix: taste at open, mid-service, and after any adjustment — the cheapest quality control there is.
What is the most expensive mistake of all?
Blaming the machine before checking the coffee, the grinder and the water. The machine is the most visible and expensive object on the bar, so it attracts blame. In a well-maintained bar it is also the least likely cause of a change in the cup: it does the same thing every day until something specific fails, and when it fails it shows up as a pressure, temperature or flow symptom rather than a taste one.
The order to check things in, cheapest and most likely first:
- The coffee. Same bag, same lot, same roast date as yesterday? How long roasted coffee actually stays good.
- The dose and the basket. Weighed? The basket the recipe was built on? Clean and undamaged?
- The grinder. Last cleaned and calibrated when? Burrs worn? Coffee sitting in the hopper or chute? Grinder maintenance and calibration schedules.
- The water. Treatment serviced? Visible scale? Water changes taste and machine behaviour, slowly enough that nobody connects the two. Why hard water damages your coffee and your machine.
- Cleaning. Group, screen, baskets, wand, milk lines. A residue problem imitates almost every other fault.
- Technique. Watch someone else make the drink and see whether the fault follows the person.
- The machine. Now check pressure, temperature stability and flow — and now a technician's time is worth paying for.
Working that order costs an hour. Working it backwards costs a callout, often a machine.
Symptom, likely cause, first thing to check
| Symptom | Likely cause | Check first |
|---|---|---|
| Bitter and sour in the same cup | Channelling from uneven distribution | The spent puck, and the distribution step |
| Shots wander with no change made | Dose by volume, or grinder retention | Whether the dose is weighed |
| Grind has drifted a long way over weeks | Coffee age, dirty grinder, water | Roast date, then grinder cleanliness |
| Espresso tastes flat, cools fast | Cold or wet cup, or the shot sat | The cup, then the drink-build order |
| Foam bubbly and separating | Air introduced too late | Where the air goes in |
| Foam collapses in seconds | Milk overheated | Jug temperature by hand |
| Steam spits or pushes sideways | Blocked or crusted tip | Wipe-and-purge routine |
| Stale, ashy first sip | Group not purged or wiped | Shower screen after a shot |
| Fast at quiet times, slow at peak | Serial workflow under pressure | The order of operations |
| Everything tastes worse this month | Water, or a cleaning routine nobody owns | Water treatment service date |
None of this needs a new machine
We build coffee machines, so this needs saying plainly: every mistake above is free to fix, and none requires buying anything.
A bar convinced it needs a better machine should work through this list first. Often the fault is on the list, the money is not needed, and the new machine would have arrived into the same problem and produced the same coffee.
The commonest real cause of poor coffee in an otherwise competent bar is neither the machine nor the barista. It is a maintenance and cleaning schedule nobody owns — not written down, not assigned to a named person, not signed off, so it happens when there is time and stops when there isn't. Quality degrades gradually enough that no one can name the day it got worse, and it gets diagnosed as an equipment problem because nothing else explains it.
We would rather say so than sell a machine into a cleaning problem. If the answer after working this list is genuinely capacity — a bar that cannot clear its peak, or a machine that leaves the counter for routine work — that is a real equipment conversation. Until the list is done, it is a guess.
Frequently asked questions
If I fix only one thing, which one? Purging and wiping the group between shots, then distributing consistently before the tamp. The first improves every drink; the second is behind most channelling.
Is tamp pressure really irrelevant? Not irrelevant — enough force to consolidate the bed is necessary. Beyond that more does nothing useful, and level matters far more than hard. Effort spent on force is effort not spent on distribution.
My shots are inconsistent but my times are always right. What now? Time and taste can disagree, because channelling changes flow. Watch the spent pucks and the streams rather than the timer, and taste every adjustment.
Can overheated milk be saved by adding cold milk? No. The structure is gone once the proteins are denatured, and diluting gives cooler milk with no texture. Pour it away, then work out why the jug went too far.
How do I stop standards drifting without standing over people? Write the specification down, assign the cleaning schedule to a named person, and taste the coffee at fixed points in the day. A standard that lives only in someone's memory drifts with their memory.
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.