L
Listen
raw signal, both tracks
Open the aperture. Capture signal from users, markets, internal telemetry, and support — deliberately broader than the question in play.
HumanInterviews, diary studies, leadership 1:1s, field observation.
AITranscription, tagging, clustering, behavioral digest, weak-signal surfacing.
I
Learn
pattern, not summary
Convert signal into structured pattern. Humans judge meaning and novelty; AI handles scale, recall, and reproducibility.
HumanSensemaking workshops, affinity, judgment on framing and weight.
AITheme frequency, segment lift, cross-corpus consistency checks.
V
Lead
decision, with evidence
Name the bet. Humans own the commit; AI tests it against prior patterns, counter-examples, and second-order effects.
HumanDecision memo, owner, guardrail, go/hold/kill.
AICounter-evidence scan, prior comparison, risk simulation.
E
Live
in-market, always on
Ship it, and keep the loop open. Outcome data feeds the next Listen — learning becomes a state, not a moment.
HumanPost-ship review, narrative, credible updates to strategy.
AITelemetry, drift detection, re-cluster against live signal.
Figure 6b — LIVE Loop vs. Double Diamondwhy the replacement matters
Double Diamond
LIVE Loop
Learning cadence
Discrete phases — diverge, converge, repeat.
Continuous. Learning is the operating state, not an event.
Role of AI
Not defined; layered on as a tool.
Architected in as a parallel track, not an afterthought.
Decision ownership
Implicit; handed off at convergence.
Explicit. Every phase names the human commit and AI contribution.
Post-ship learning
Typically out of scope.
In scope. Live-market signal feeds the next Listen directly.