A decision-first field method
Sifting for Better Decisions
Which fragments are noise and which are trying to tell you where a decision is stuck?
Repeated questions, quiet workarounds, delayed approvals, unexplained metrics, and uncertain ownership are easy to dismiss alone. Sifting gives them a path without pretending every fragment matters.
Before you collect more
Start with the choice, not the artifact.
Use Sifting when fragments may point to a recurring decision problem. Skip it when the decision is already explicit, owned, and supported by enough evidence to choose responsibly.
Name who owns the choice, what the selected option protects and trades away, and which event or review date should reopen it. Accepted uncertainty is part of the decision not permission to let it drift.
Effort boundary and stop condition: Stop active collection when newly arriving material produces redundant signal rather than new clarity. Reach a decision or define the smallest viable test step once marginal information ceases to change the outcome ffBdo not confuse data volume with decision confidence.
Follow a signal
Let one piece of friction change shape.
Choose one signal, then use Continue or Back to inspect five distinct stages. The final stage names an owner, a learning limit, and when to review the choice.
- 01Signal
- 02Meaning
- 03Decision
- 04Artifact
- 05Next step
Signal · 1 of 5
People keep asking the same setup question.
Record the question, where it appears, and who asks it. Do not decide what it means yet.
One question may be noise. Recurrence makes it worth pairing with another observation.
Review before you act: Confirm who owns the next step, what it protects, and which evidence or event should reopen the choice.
Change the lens
One signal. Four useful questions.
Keep the observation fixed, then change the question around it. Choose a lens below, or use the arrow keys while a lens has focus.
The evidence has not changed. The lens determines what you inspect next.
Orient the decision
Notice. Combine. Decide. Make.
Collect a few recurring fragments, test whether they form a pattern, name the decision they might support, and choose the smallest useful artifact.
Connect the evidence
Signals need somewhere to meet.
Compare the practice
Two moments in the same wider practice.
Spiral combines existing artifacts to see where opportunity may be forming. Sifting works earlier, while the material is still fragments, behaviors, and questions.
Act on the pattern
Bring three observations, not a predetermined answer.
Write what each observation may mean, what else could explain it, and which decision connects them. Stop when more collection no longer changes the choice.
Signal Pattern Atlas
The pattern is a proposition, not a diagnosis.
Choose a recurring shape to compare its complete signal-to-artifact chain. The examples are prompts for investigation, not diagnoses.
- Weak signalRepeated setup question
- Paired withDrop-off before the next step
- May point toExperience moves before orientation
- DecisionClarify, redesign, or guide?
- ArtifactJourney map with confusion points
- Weak signalManual workaround becomes normal
- Paired withDelay at the same handoff
- May point toReal workflow split from official flow
- DecisionFormalize, automate, or replace?
- ArtifactReal-versus-official workflow map
- Weak signalAgreement without movement
- Paired withOne proof point keeps returning
- May point toConfidence missing, not alignment
- DecisionValidate, pause, reframe, or reduce risk?
- ArtifactDecision brief with evidence
- Weak signalMetric shifts without explanation
- Paired withBehavior changes in one segment
- May point toNumber lacks decision context
- DecisionAct, wait, investigate, or segment?
- ArtifactKPI context card
- Weak signalEveryone participates; nobody commits
- Paired withThe topic returns every meeting
- May point toDecision rights are missing
- DecisionWho owns, contributes, approves, escalates?
- ArtifactOwnership map or decision tree
Interactive engagement formats
Choose the kind of participation the decision needs.
Compare six ways for people to assess, explore, reflect, size trade-offs, build shared context, or work together. Choose the format that best supports the people involved and the next useful decision.
When the same kind of decision comes up often
A checklist keeps the needed information, decision rules, person responsible, exceptions, and next step clear and consistent.
When the path is unclear
Show where people enter, pause, loop, wait, hand off, or lose confidence.
When confidence is missing
Name the decision, assumptions, evidence, uncertainty, options, and trade-offs.
When trade-offs need sizing
Compare cost, effort, timing, impact, and the price of waiting.
When shared understanding is the bottleneck
Bring signals, assumptions, options, and decision rights into one room.
When the group is not ready for more
Use the smallest intervention: the question that changes the next conversation.
Choose how far to go
Continue the thinking—or take one useful question with you.
Read the uninterrupted argument, explore the related decision questions, or contact Richard about a consequential choice that remains hard to frame. Each route stands on its own; none is required.
Methods and playbooks