Gemba walks: looking at the work without turning it into an inspection

Being shown the plant is not the same as seeing it

A guided tour follows the route the guide chose, arrives where the guide is confident, and lingers where the area photographs well. Seeing the work means standing at one process long enough to watch several cycles, following a single job through consecutive steps, and noticing what the people there do about the things that go slightly wrong. It is slower and covers less ground. That trade is deliberate, because depth at one place beats a circuit of the site, and the interesting information lives in the small recoveries nobody mentions afterwards.

The urge to fix it on the spot is the main failure mode

A senior manager who spots a problem and issues an instruction has just done three damaging things: bypassed the supervisor, taught the area that raising an issue produces an unfunded order, and removed the chance of finding out why the situation persisted. The alternative is uncomfortable. Ask what the standard is, whether it is achievable, what gets in the way, what has already been tried, and leave with a commitment about who will look at it rather than a solution. Anything unsafe is the exception and should be stopped immediately, without discussion.

A route, a rhythm and a question you came to answer

Wandering produces anecdotes. A walk with a subject produces comparison: today it is material availability at the assembly stations, next time it is what happens at handover, later it is why the tagged defects on one line are not clearing. Fixed timing matters as much as the route, because a walk that happens only when the diary allows will stop happening in the first busy month, and the floor reads that accurately. Different levels should walk at different rhythms and depths, with the supervisor's version frequent and short and the plant manager's less frequent and more deliberate.

What happens afterwards decides whether people talk to you next time

Every issue raised needs a visible destination: fixed now, assigned with a name and a review point, or explicitly declined with a reason. The third option is legitimate and badly underused, because people accept a considered refusal far better than silence. Without that ledger the walk generates a list that decays quietly, and the floor's rational response is to raise nothing requiring effort. Follow-up is also where the walk repays its cost, since the pattern of what keeps being raised across several areas is usually a management-system problem no individual supervisor could have solved.

Signs that the floor is managing your visit

Preparation is a compliment and a warning. Boards freshly updated on the morning of a scheduled walk, a route that always avoids one corner, an operator who answers while looking at the supervisor, an area whose problems are invariably described as solved. Each suggests the walk has become a performance. The countermeasures are simple enough: vary the timing, walk occasionally without the area manager, ask about the last thing that went wrong rather than about performance, and react well the first time somebody tells you something genuinely awkward. That single reaction sets the tone for months.

Frequently asked questions

How often should senior managers walk the floor?

Frequently enough to be unremarkable and predictably enough to survive a busy period. Supervisors are effectively walking continuously; a plant manager doing it several times a week with a subject in mind is common, while a monthly circuit is closer to a visit than a routine. The failure worth avoiding is a rhythm so demanding that it is abandoned within a quarter, because starting and stopping teaches the floor more than a modest cadence held for a long time.

Should we take notes and photographs while walking?

Notes yes, sparingly, and tell people what you are writing down. Photographs are useful for equipment condition and layout problems, and are read very differently by the person standing in the frame, so ask first and explain where the image will go. The larger risk with heavy documentation is that the walk becomes an audit in everyone's perception, and once that happens the answers you receive change. Recording commitments matters more than recording observations.

How is this different from a safety tour or an audit?

An audit tests conformance against a standard and produces findings; a safety tour looks for hazards and carries legal weight. Both are necessary, and both are inspections, with the behaviour inspections produce. Walking the work is an attempt to understand why a process behaves as it does and what obstructs the people running it. Mixing the purposes is tempting because you are standing in the same place, and it costs you the candour, so keep them separate activities with separate names.