On this page
Bar work is often treated as a single skill that people are either good at or not. It is more usefully seen as several, with quite different learning curves, and the difference decides where training effort is worth spending.
The skills that plateau
Speed of execution plateaus. A bartender gets faster for a few months and then reaches roughly the limit of what the physical task allows. Beyond that point more repetition adds very little.
Recipe recall plateaus, quickly. Once someone knows the menu, knowing it harder is not a thing.
Till operation plateaus almost immediately.
None of this is a criticism. These are the skills that have to be in place before anything else works, and they are the right thing to teach first precisely because they finish. The mistake is treating them as the whole job, which is what happens when training stops after induction.
The skills that compound
Reading a room
Knowing that a group is about to become a problem, that a regular is being quiet for a reason, that the room is thirty minutes from a rush. This improves for years, because the pattern library keeps growing, and it is the difference between a bar that handles Saturday and one that survives it.
Diagnosing rather than reporting
A junior member of staff reports that the ice machine is slow. A senior one works out that it has been slow since the delivery schedule changed and says so. The second is a much rarer skill and it is learnable, mostly by being asked to explain rather than to notify.
Judgement about exceptions
When to comp, when to hold the line, when to break a rule because the situation warrants it. This compounds because every exception encountered adds to the reference set, and it is the skill that makes someone promotable.
Teaching
The most compounding skill of all, and the one bars invest in least. A bartender who can develop the person next to them changes the trajectory of the whole team, and the ability improves every time they do it.
What this means practically
Teach the plateau skills fast and stop. They are prerequisites, not a curriculum. Spending month three on speed is spending it on something that has largely finished improving.
Develop the compounding ones deliberately. They do not develop on their own with time served. Someone can work five years and get no better at diagnosis, because nobody ever asked them to diagnose anything.
Hire for the compounding ones. Speed and recall can be taught in weeks. Judgement and the instinct to explain rather than report are much harder to install, so they are worth weighting heavily at interview.
The connection to retention
People stay in jobs where they are visibly getting better. A role that consists entirely of plateau skills stops being interesting at about month six, which is roughly when hospitality turnover tends to bite.
That is an argument about mechanism rather than a measured claim, and it is offered as one. What we can point to is a single venue in our case studies that attributed a 73 percent reduction in turnover to LearnTAP, which is client reported and specific to that site.
The broader point stands independently of the number: a job with a visible ceiling is a job people leave, and most bar roles have a ceiling only because nobody built anything above it.