On this page
Key takeaways
- The most useful training material in a bar is what happened on its own floor last week.
- Real events teach better because they are already remembered, carry the bar's actual constraints and show real consequences.
- It rarely happens because of retrieval: the practice needs a shift's detail within about two days, and by the time most numbers arrive the details have faded.
- Pick one instructive event, establish the facts, ask the people who were there, and end with one change checked next week.
- Keep the session about the event rather than the individual, or people stop volunteering information.
Bars buy training material. Courses, videos, laminated guides, occasionally a consultant. All of it is general by construction, because it was written for every bar, and the gap between a general example and your Friday is where the attention goes.
Meanwhile every venue generates a continuous stream of specific, relevant, memorable teaching material, and discards it.
Why do real events teach better?
They are already remembered. A large group that went badly on Saturday needs no setup. Everyone who was there can picture it, which means the conversation starts at the analysis rather than at the scenario.
They carry the actual constraints. A general example assumes a general bar. Your bar has a specific layout, a specific bottleneck at the service well, and a specific pattern of who arrives when. Training that ignores those teaches a version of the job nobody does.
They demonstrate consequence. The comp total from that Saturday is a real number. It is more persuasive than any hypothetical, and it is not an accusation, because it is a fact about a night rather than about a person.
Why does it rarely happen?
Not disagreement. Retrieval.
By the time anyone has the numbers for Saturday, it is the following week and the details have faded. What survives is an impression, weighted toward whatever went visibly wrong, which is usually not the most instructive thing that happened.
So the venue keeps buying general material, because general material is available and specific material is theoretically available but practically not.
A workable practice
Pick one event, not a review of the week
One thing that happened, chosen because it is instructive rather than because it was the worst. A week reviewed comprehensively is a meeting. One event examined properly is training.
Start with what happened, not with what should have
Establish the facts before the lesson. This matters because the assumed story is frequently wrong: the comps were not generosity, they were a service failure at eight that nobody reported.
Ask the people who were there
They know. They have rarely been asked, because the number arrives long after anyone could connect it to a night. Asked within a couple of days, you generally get the answer in a sentence.
End with one change, checked next week
Not a list of improvements. One thing, specific enough that next week it is obviously either happening or not.
Keep it about the work
The failure mode is obvious and worth naming. If examining real events becomes a way of identifying who was at fault, people will stop volunteering information, and volunteered information is the whole mechanism.
The framing that holds is the event, not the individual. This happened, here is what it cost, here is what we do differently. If a specific person is genuinely the cause, that will be evident without a session having been built to reveal it.
What this needs from your tooling
Very little, and one thing that is hard to get. You need to be able to see what happened on a specific shift, at the level of products, comps, sections and timing, within about two days.
That is the whole requirement. It is also precisely what monthly reporting cannot provide, which is why the practice is rare rather than because operators think it is a bad idea.
CoreTAP reads the point of sale you already run and makes the shift legible while it is still remembered.
Frequently asked questions
Why use our own shifts for training instead of a course?
Bought material is general by construction, while your own events are already remembered, reflect your layout and bottlenecks, and carry real consequences. The conversation can start at the analysis rather than at the scenario.
How do I run a training session on a real event?
Pick one event because it is instructive, not because it was the worst. Establish what happened before the lesson, ask the people who were there, and end with one change that is checked the following week.
How do I stop it turning into a blame session?
Frame it around the event, not the individual: this happened, here is what it cost, here is what we do differently. If examining events becomes a way of finding fault, people stop volunteering information, and that information is the whole mechanism.
What information do I need to do this?
You need to see what happened on a specific shift, at the level of products, comps, sections and timing, within about two days. Monthly reporting cannot provide that, which is why the practice is rare.