Blog » Lean

Standardized Work: How to Capture the Best Worker's Method for the Entire Shift

Lean · 4 min read · 05.10.2026

The same part, three shifts, three results. The morning shift assembles it in 14 minutes, inspection finds nothing to fault. The night shift holds at 16 but battles rework twice a week. The afternoon shift needs 19, and every few weeks a customer complaint arrives. The people? All three are capable and hard-working. The problem lies elsewhere: nobody captured how the best one works and passed it on. That's what standardized work is for. Amid turnover and labor shortages, it's the cheapest quality insurance — one A4 sheet that outlives every resignation.

At conferences, people discuss kanban and value stream mapping. At Foreast, on projects in manufacturing companies, we see something less showy but more effective: the best worker's method, approved by the team, hung at the workstation. No software. No logo.

What standardized work is — and what it isn't

By definition, standardized work is the best known, safest and least wasteful method for an operation. Not the best possible — the best known. A standard doesn't claim perfection; it says: today, this is how we know to make the part. Whoever proves a better way with numbers, we rewrite the standard.

So what isn't it? Not another ISO document an auditor pulls out once a year while it gathers dust on a shelf. Nor a cage for thinking — rather the place where thinking begins. Without a firm standard, we can't say what counts as a deviation. Without deviations, we don't see where we're wasting. Every improvement then ends up in a meeting room, not on the shop floor.

A good standard carries three components. Takt — the pace the operation must run at for the customer to get the part on time. You can work it out on a napkin: available time per shift divided by daily demand — derived from demand, not from the fastest person's record. Next, the sequence of steps, including the key points on which quality or safety stands or falls. Finally, standard in-process stock: the minimum material at the workstation, so the flow doesn't pile up as work in progress. It doesn't pin down every movement to the millimeter — only what decides quality, safety and takt. The rest is left to the operator.

How to write a standard in a single morning

Standardization needn't be a six-month project — with a team, you handle one operation in a morning. We've tested this process in dozens of workshops:

  1. Pick one operation with variation. Not the whole line — one operation. Measure times across all shifts for at least five days. Where the gap between fastest and slowest is largest, the standard is missing most.
  2. Observe the best worker. Record three to five cycles. Break the work into steps and note a key point at each one: hand angle, tightening torque, tool position. Next to each, write the reason — why it upholds quality or safety. Without a reason, people don't remember the points.
  3. Agree on the method with the team. The best performer demonstrates, the others try it. A standard must be repeatable by an average person on an ordinary day — if only the champion manages it, adjust it.
  4. Document it at the workstation. A4 format. Photos instead of paragraphs, times, key points with reasons. Shop-floor language, not auditor language. Print, laminate and hang it where the work happens — a standard in the office is dead.

Half an hour at the board with a camera replaces hours of meetings. On one assembly operation we captured a method that cut changeover from 45 to 12 minutes; the other shifts mastered it within three days.

A standard is the only shop-floor procedure that survives the departure of the best worker.

A baseline for improvement, not a cage

The most common objection: a standard ties your hands, kills initiative, turns people into robots. In reality we see the opposite — improvement needs a firm baseline. When every shift works differently, improvement doesn't exist — only opinions about who's better. A standard turns opinion into a number, and a number can be improved.

Two mechanisms keep the standard from gathering dust. First: a five-minute audit by the supervisor once a week. If the work follows it, he looks for improvements; if not, he looks not for a culprit but an answer — is the standard bad, the person untrained, or did the material arrive late? A deviation is information, not an offense. Second: the one-path rule. The standard changes only through a verified improvement. Whoever brings an idea tests it, measures the result and, if better, rewrites the standard with a date and name. The team then sees the standard belongs to them, not the documentation office.

One side effect today outgrows the main benefit: onboarding. Instead of three weeks of "watch and do," a new worker gets a step-by-step procedure with photos and works under supervision from day one. In one client's press shop, we cut training a new operator from three weeks to four days. Scrap fell, because the newcomer didn't rediscover known mistakes. When the best welder hands in his notice, his method stays with the company.

You don't need special software — one morning, a team and a camera. The payback comes in days.

Key takeaways

  • Measure one operation across all shifts for at least five days; the greatest variation shows where the standard is missing most.
  • Capture the best worker's method on video and next to each key point write the reason — without it, nobody remembers.
  • The standard belongs on a laminated A4 with photos, hung at the workstation. Nobody reads a document sitting in the office.
  • The supervisor checks compliance five minutes a week; changes go only through verified improvements, with a date and the author's name.


« Previous article  |  Next article »

« All articles