Standard Work
Standard work documents the currently best, agreed way of performing a task — so precisely that deviations can be seen, training becomes consistent, and improvement has a fixed starting point.
Those who do the work write the standard with you — not just the engineer
There is agreement that there SHOULD be one shared method (have that conversation first)
The process is reasonably stable — chaotic processes are stabilised before standardisation
Watch the task performed by 2-3 different people and note the differences — order, grip, checks, time spent. The differences are not errors; they are the raw material. At each difference ask: why do you do it that way?
For each sub-element, the team chooses the method that is best on quality, safety and time — in that order. Disagreements are settled by trying both and measuring, not by seniority. The result is the currently best known method — not the perfect one.
One page per task where possible: numbered main steps, key points (what is critical and WHY) and photos rather than text. The template’s three columns: What — How — Why. The Why column is the most important: it carries understanding, not just instruction.
A colleague who didn’t take part in creating it performs the task alone using the standard. Every place where they hesitate or ask is an error in the standard — not in the colleague. Adjust and repeat until it holds on its own.
The standard hangs where the task is done — not in a folder system. Everyone who performs the task is trained on it (including the experienced), and the training is signed off on a competence matrix.
A fixed rule: if someone finds a better method, it is tested, and the standard is updated with a new version and date. A standard unrevised for a year is audited: is it still the best known method — or has practice quietly slipped elsewhere?
How it goes wrong in practice — and this is where most implementations part ways.
Then it describes how the task ought to be, not how it succeeds in reality — and the floor files it politely away.
A 12-page instruction is never read again after day one. Main steps + key points + why. Details that aren’t critical don’t belong in the standard.
If no routine (audit, gemba, board) detects that practice deviates from the standard, they drift apart within weeks — and the standard becomes a historical source.
The method only works when it becomes daily routines and leadership behaviour. The operations check shows in 4 minutes where your operations stand — and what should come first.