What it means
A distribution report says orders are leaving late, so a manager visits packing during a normal shift and sees labels printing across the building, an observation that may reveal a delay hidden in the summary numbers. Choose a purpose first, such as how orders flow, why rework occurs or where a handoff stalls, because wandering without a question becomes a tour rather than learning.
The Lean Enterprise Institute describes a gemba walk as going to the actual workplace to see and understand the process, and its guidance pairs observation with asking why and showing respect. Tell staff what the visit is for, since people may assume a senior manager is auditing them personally, and state that the aim is to understand the work and improve conditions.
Observe the normal process by following a product, customer request or document from start to finish, including handoffs, because a staged demonstration may hide common interruptions and delays often occur between teams rather than at one station. Visit at different times too, as a morning walk may miss evening staffing, peak demand or weekend constraints, and set the cadence to match process risk and improvement work rather than treating a fixed weekly rule as dogma.
Ask open questions such as what happens when this input is missing, where do you wait and what do you do when the system fails, and avoid leading questions that imply the answer is already known. Listen to the people doing the work, because they know details a dashboard may omit, but do not promise every suggestion will be adopted; explain how decisions will be made.
Avoid blame as well: if workers use a workaround, ask what constraint made it necessary, because punishing the workaround without fixing the system can drive it out of sight. Separate observation from inference, since seeing a queue at 10 in the morning does not prove it exists all day, so check logs and other shifts before declaring a root cause.
An illustrative observed wait share is waiting minutes divided by total observed process minutes, so if a sample order takes 50 minutes and waits 30, the share is 60%, though one observation is not the whole operation. Compare planned with actual work too, because written procedures may describe an ideal sequence that real tools cannot support, and that gap is a useful finding, not necessarily employee noncompliance.
Follow safety rules, as visitors should use protective equipment, approved routes and induction where required, and a leader's presence does not exempt them from site controls. Respect customer and worker privacy by not recording faces, screens or private conversations casually, and take notes about process facts with appropriate permission.
Keep the group small, because a crowd can disrupt work and change behaviour, and build trust over time, since if leaders only visit after a failure people may brace for blame. Do not solve everything on the spot: an immediate safety hazard needs action, but a complex process change needs evidence and input, and a quick order can create a worse bottleneck downstream.
Capture facts and questions, distinguishing observation from interpretation, and assign owners and a time to report back, because notes sitting in a notebook do not improve work; check proposed improvements with staff, since a new label printer near packing might help but space, network and maintenance requirements matter. Review outcomes, because a faster step may push work to another team, and share what was learned so that future questions do not feel extractive; for owners, a gemba walk grounds decisions in how work actually happens, and its strength is disciplined observation followed by respectful action.
In practice
Real-world examples.
Example
A manager follows an order from picking through packing to dispatch, noting every wait and handoff. She sees that labels are printed at a remote printer, so packers leave their station repeatedly. She records the observation and then checks the system timings before drawing a conclusion.
Example
A leader asks why a team re-enters data already held in another system. The team explains that the two systems do not share a customer code, so staff retype it to avoid mismatches. The leader records the answer as an observation and asks IT to test a shared code instead of telling the team to stop.
Example
A walk across shifts reveals a handoff problem unseen in daytime reports. The evening team receives incomplete pick lists because the day shift closes its paperwork late. The manager compares logs from both shifts before naming this as a cause.
Formula
Calculation
Observed wait share = waiting time / total observed process time. For one sampled order, 30 minutes of waiting in a 50-minute process gives 30 / 50 = 60%, which describes that order, not the whole operation.
For comparison, suppose a trial printer closer to the packing benches cuts the waiting to 12 minutes in a 32-minute process for a later sampled order. The share becomes 12 / 32 = 37.5%, a fall of 22.5 percentage points. Several orders across different shifts would be needed before treating that improvement as reliable.Case study
Seen in the real world.
This entirely fictional example follows Amber Logistics. A manager saw packing staff walk repeatedly to a remote printer. The team tested a closer printer and tracked errors and dispatch time afterward. The case illustrates observation and testing, not proof that equipment placement is always the cause of delay.
Before moving anything, the manager recorded how long orders waited for labels across two shifts and asked packers what they would change. The trial ran for two weeks, and dispatch time and label errors were compared with the previous fortnight rather than judged by impression. The trial reduced walking, but it also showed that some label errors came from a mislabelled shelf, which the team corrected separately. The manager reported back to staff on what had changed and what had not, so that the next walk began with trust rather than suspicion.
Watch out
Common mistakes.
- Using the walk as a surprise individual performance inspection.
- Generalizing from one observation without checking other times and data.
- Collecting ideas and never reporting what happened to them.
Questions
People also ask.
What is a gemba walk?
A respectful observation of work where it actually happens to inform improvement.
What does gemba mean?
It means the actual place where value-creating work occurs.
How often should they happen?
Use a cadence that matches the work and improvement goal; no universal weekly rule applies.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%