01 / Section
Repetition is a starting point, not a conclusion
A familiar sequence can mean several different things. It may reflect a genuine workflow, a one-off exception, a shared template, or the way one captured batch happened to unfold. Treating the first repeated sequence as a pattern creates false confidence about how work is done.
Start with the outcome the team was trying to reach. Then ask whether the same kinds of steps appear in more than one distinct attempt to reach it. A pattern is stronger when the repeated steps are linked to comparable outcomes, not merely when the same apps or sites appear in a record.
- Name the business outcome before naming the workflow.
- Use distinct attempts as the unit of comparison.
- Record exceptions instead of smoothing them away.
02 / Section
Verify the evidence across attempts
Review the captured batches side by side. Mark which actions, handoffs, decisions, and outputs recur. Note what changes between attempts, including different inputs, routes, approvals, or external conditions. The goal is not to force every attempt into one standard path; it is to see which parts remain recognizable when the circumstances differ.
Keep recorded evidence separate from interpretation. A captured sequence can show that an event or transition was recorded. It does not, by itself, prove why the step happened, whether it was necessary, or whether the same step would be appropriate next time.
- For each repeated step, identify the supporting capture and its place in the attempt.
- Label gaps, ambiguous transitions, and conflicting evidence.
- Avoid treating a single citation as proof that the whole workflow is established.
03 / Section
Use app and site sequences as discovery signals
Applications and websites can help you find where to look, especially when the same sequence appears in several attempts. They are useful clues about possible handoffs, research stages, or record updates. They are not a complete description of the work or its purpose.
Keep these sequences separate from the evidence map for the business outcome. The same site may support different tasks, and the same task may move across different tools. Follow the surrounding captured evidence before assigning meaning to a sequence.
- Ask what evidence connects a tool transition to the intended outcome.
- Do not infer intent from tool use alone.
- Treat citations as provenance: they show where a claim came from, not that the claim is certain.
04 / Section
Turn a verified pattern into a careful next step
Once recurring elements are supported across attempts, describe the pattern narrowly. Say what was observed, where it appeared, and what remains uncertain. For example, a team may observe that comparable requests repeatedly move through the same review and record-update stages. That is more useful than claiming the entire process is fixed or universal.
Planning assumptions should be labeled as assumptions. If the team wants to test a clearer handoff, define the proposed change, the evidence it is meant to address, and the conditions under which it should be reviewed. Keep the plan open to revision when new attempts show a different route.
- Write separate fields for recorded evidence, interpretation, and planning assumption.
- Choose a small, reviewable change rather than declaring a process complete.
- Recheck the pattern when the outcome, inputs, or constraints change.