The Training That Happens On The Floor

There is no training programme. There is a first shift with whoever was on, and after two weeks people either have it or they do not.

This works better than formal programmes do in most venues — which is why it survives. But it produces wildly different results depending on who the new person happened to be paired with, and nobody is looking at that.

What actually gets transferred

Three things, and only the first is what anyone thinks training is.

The mechanics. Where things are, how to ring something, the order of operations. Easy, and it is the only part a document helps with.

The judgement. When to intervene at a table, when to call a manager, how much to push. This is the part that decides whether someone is good, and it transfers only by watching someone do it.

The standard. What "acceptable" looks like here. ⚠️ This transfers whether or not anyone intends it — a new person paired with someone cutting corners learns that the corners are cuttable, permanently.

The invisible variable

Who they shadowed.

Nobody records it, so nobody can see that the people trained by one person stay and the people trained by another leave. That correlation exists in most venues and no one has ever looked at it, because the pairing was decided by whoever was on the rota.

The fix is small: record who trained whom. One field. Then in six months you can see something you currently cannot.

The people doing it are invisible

Someone is training every new starter. It is slowing them down, it is not in their job description, and it does not appear anywhere.

⚠️ They are frequently the same people carrying the guest relationships — which means when they leave you lose the relationships and the training capacity in the same week, and nobody connected the two.

Recognising it costs nothing and is the highest-return thing on this list. A person who knows the venue values their training does more of it.

The three changes worth making

One — choose the pairing deliberately. Not "whoever is on". This is the entire intervention, and it is free.

Two — one sentence at the end of the shift. Not a form. "What surprised you tonight?" is enough to surface the thing that will otherwise become a habit.

Three — a short list of what must be covered by week two. Not a curriculum — a checklist of the six things that cause problems when missed. ⚠️ Longer than about ten items and it will not be used, which is worse than not having one.

What documents cannot do

The mechanics can be written down. The judgement cannot, and attempting it produces a document nobody reads and a false sense that training exists.

The realistic split: write down the mechanics so the shadowing time is not spent on where the glasses are, and spend the shadowing time on judgement.

That is the whole design.

The test at two weeks

Not "do they know the system".

"Would I put them on a Saturday section alone?"

⚠️ If the answer is no and nobody has said so, the person is about to be put on a Saturday section alone anyway, because the rota needs filling and nobody was asked.

Ask it explicitly, at two weeks, about every new starter. It takes thirty seconds and it is the only checkpoint most venues need.

Three to hold

Who trained whom. The field nobody records and the one that explains the most.

Ninety-day retention by trainer. Available only if you record the first.

Share of new starters with an explicit two-week check. The cheapest quality gate available.

Where the record has to sit

If shifts are in a rota and nobody notes who was paired with whom, the biggest variable in your training is unrecorded — and the people carrying it stay invisible until they leave.

tasteck keeps shifts against people, so who worked alongside whom on a first shift is a record rather than a thing someone might remember.

You do have a training programme. It is just being assembled weekly by whoever writes the rota, without anyone intending it.