Index
Searching only for the manager can hide what happens during shift changes, peaks and exceptions. In operational software, work observation needs to represent the contexts in which rules are actually executed.
How to evaluate this decision
Select participants by role and usage condition, not just availability. Watch for pass-through, interruptions, and parallel systems. Separate official procedures from adaptations created to be able to work. These adaptations may reveal a legitimate need or a risk; the report must explain the difference without individually stating who participated.
Criteria for comparing proposals
- Sample: include shifts, units and experience levels relevant to the problem investigated.
- Observation: monitor a completed task and record where waiting, rework and consultation with third parties arise.
- Synthesis: relate findings to product decisions, indicating what has not yet been observed.
A scenario to discuss with the supplier
Hypothetical example: during business hours, a registration question is resolved by the supervisor. At night, the same doubt blocks service. A search restricted to the first round would not identify the need for an exception rule.
What to validate upon delivery
Require a matrix of contexts covered and anonymized evidence of key findings. Ask what conclusions would change if another unit were observed. Avoid treating frequency in a qualitative sample as a percentage of the entire operation.
Prepare the conversation about the project
To plan research with Quantum9, inform shifts, profiles, access restrictions and the process that needs improvement. Combine operation participation and a way to observe without interrupting critical activities.
Prototyping and validation · Map the company's priority