Example data. Fieldnote is a fictional company used to fill every board, so the examples tell one consistent story.
Fieldnote, mapping the last ten shipped items from idea to production, with Karim, Tom, Ana and Julien. Real dates pulled from the tracker.
What the team took away
The team was about to hire to go faster. Twelve percent working time said the problem was the 11 days before refinement and the single reviewer, neither of which a new hire would fix.
How to run it
Preparation
Pick one product and one type of work, end to end. Mapping 'everything the team does' maps nothing.
Get the real dates from the last ten items: created, started, first review, released. Memory always underestimates waiting.
Invite someone from every step, including the ones outside the team. The queues you cannot see are owned by people you did not invite.
Say out loud at the start that the map describes the system, not the people in it.
Agenda
30 min
Draw the steps List every step from idea to production, including approvals and handoffs. Do not judge anything yet, just get the boxes on the wall.
40 min
Add the two times For each step, the time spent working on it, and the time spent waiting before it. The second number is the one nobody has ever measured.
20 min
Compute the ratio Add up both columns. The share of working time in total lead time is usually under fifteen percent, and saying that number out loud changes the conversation.
30 min
Find the biggest queue Point at the longest wait, not the slowest step. Ask who owns it, and what it is waiting for.
30 min
One experiment Pick one queue and design one change to shrink it. One, measurable, with a date to re-measure the lead time.
Pitfalls
Measuring only the work and not the waiting. You will conclude the team is slow, when the system is where the time went.
Mapping the process as it is documented. Map what happened to the last ten items, which is usually a different process.
Fixing five queues at once. You will never know which change worked, and the team will be exhausted before the next measure.
Afterwards
Re-measure the lead time after one cycle with the change in place. If the biggest queue is a cross-team dependency, the dependency map is the next session.