← All articles

Workflow evidence

Describe the Workflow You Can See, Not the One You Assume

When collection is incomplete, describe the captured attempt and its limits. Treat gaps as questions to investigate, not proof that a step did or did not happen.

Text-free Shadow evidence-map composition with linked evidence cards and open gaps, in ink black, warm cream, mustard gold, and muted sage.

A gap narrows the claim

A workflow description is only as strong as the evidence behind it. If a captured batch shows a request being prepared and a later record being updated, it can support a description of those observed steps. If the middle of the attempt is missing, it cannot establish exactly how the work moved between them.

Write what the batch supports, then name what it does not show. A missing record is a limit on the account, not proof that a step was skipped, delayed, or completed elsewhere.

Keep three kinds of statements apart

Separate recorded evidence from interpretation and planning assumptions. Recorded evidence is what a cited item shows. Interpretation is a cautious reading of how the items may fit together. A planning assumption is a condition to test—for example, that a handoff usually happens outside the captured tools.

A citation establishes where a statement came from; it does not make the statement certain. Keep the wording calibrated: “the batch shows,” “this may indicate,” and “confirm whether” tell readers what is observed, inferred, and still open.

Compare attempts before describing a pattern

One captured attempt can illustrate a possible path. It is not enough to establish the usual path. Compare distinct attempts and note which evidence-linked steps recur, which vary, and where collection is incomplete.

If a step appears in some attempts but not others, describe that variation rather than smoothing it into a single standard sequence. Record the scope of the comparison so readers know which attempts informed the account.

Keep discovery signals in their place

An app or site sequence can suggest where to investigate, but it is a discovery signal, not a confirmed business workflow. Keep it separate from evidence-linked attempts until the relevant actions and outcomes are supported by captured evidence.

For an operations review, end with a practical next question: what evidence would resolve the gap, and which part of the description should remain provisional until it is available? That makes uncertainty useful without turning it into a conclusion.