Describe what happens before proposing a fix

A useful frame names the place or task, an observable pattern, who or what is affected, and what evidence would clarify it.

Use this sentence pattern: During [task or situation], we observed [specific event or count]. This affects [work, handoff, or group]. We need to learn [open question] by checking [observable evidence].

Keep the description neutral. Separate what you saw from what you think caused it, and avoid including personal or sensitive information that is not needed to understand the work.

Make the problem observable

Fictional Northstar Supply example: a shift handoff sometimes leaves the next team unsure which open item needs attention.

Too broad to test

“Shift handoffs are bad and people need to communicate better.”

  • Does not identify where or when the issue occurs.
  • Blames people without describing an observable event.
  • Does not say what evidence could clarify the problem.

Specific enough to explore

“In the evening-to-morning handoff, the next owner is not recorded for some open requests. We want to count how often the next owner and open item are both clear at the start of the following shift.”

  • Names the handoff and the missing information.
  • Describes a checkable observation.
  • Leaves the cause and best change open for discussion.

Gather enough context to frame it

Use observations that are relevant to the work question, not a large data collection exercise.

  1. 1

    Pin down the moment

    Name the task, location or handoff, and the situation in which the issue appears.

  2. 2

    Record a small set of observations

    For the sample handoff question, note whether both the next owner and open item are recorded. Keep the unit and observation period clear.

  3. 3

    Check whether others recognize the pattern

    Ask people who perform or receive the work what they observe. Note differences by shift or site rather than assuming one account applies everywhere.

  4. 4

    Name boundaries and dependencies

    Identify the work owner and any approved procedure, system, or other team that constrains possible changes. Do not use an informal experiment to replace a controlled process.