When invoices sit unpaid, the first explanation is often that Finance is too busy. The actual delay may begin earlier, when a work order arrives without a property code. An operational assessment follows the work closely enough to tell the difference. KAPCO’s published improvement project offers a concrete example of investigating the current process before committing to a change.
The Case: Observation Before Investment
NIST’s January 2026 account describes Ohio manufacturer KAPCO examining changeovers and production processes through time studies and data analysis. The work led into improvement planning, training, and visual management. KAPCO’s project provides a public example of using observations to shape an operational improvement. The worked invoice example below applies that method to a different task; it is not a claim about a My Legal Works engagement.
Apply the Lesson to an Operating Question
Worked example: start with one responsibility and one operating question. For a Los Angeles property manager, that might be why approved vendor invoices remain unpaid after receipt. Review a representative period, including ordinary items, urgent exceptions and returned submissions. Select the sample before interviewing people so memorable incidents do not become the entire evidence base. Record the limits of the sample. Twenty invoices can reveal a recurring handoff problem; they cannot establish every cause of delay across a company.
A useful finding states a condition and its effect: “Seven sampled invoices waited for a property code before entry.” The recommendation can then target that problem, such as collecting the code before submission. Do not jump from a delay to a software purchase. Compare a changed intake field, a clearer handoff and a system change. Estimate effort and dependencies for each option without presenting an untested saving as a result already achieved.
Keep the Evidence Connected to the Decision
Capture observations in a consistent format. Keep the underlying record or a secure reference available so another reviewer can trace the finding. Avoid copying unnecessary personal or financial information into a general assessment file.
Present a short list of observed problems, ranked by business impact and confidence in the evidence. Name the owner who can approve a trial. Set a narrow success measure, such as fewer returns for missing codes in the next sample. Preserve unresolved questions in the assessment so the trial can answer them. The owner should be able to choose a trial from the assessment and see which finding it will test.
| Item | What to Record |
|---|---|
| Observed Event | Invoice received Monday, returned Wednesday for a missing property code. |
| Elapsed Time | Record time waiting separately from time actively worked. |
| Evidence | Reference the relevant invoice, email or system event. |
| Likely Cause | Property code was not requested at intake; confirm with the receiving team. |
| Confidence | State whether the pattern appears once or in multiple sampled items. |