Discover. Test. Fix. Ship.
A short, repeatable cycle that produces something you can see every week — and a written readout at every decision point.
Six phases, color-coded, in the order they always run.
Each phase has one output and one decision point. Phase lengths scale with the engagement; the order does not change.
Discover
Week 1Stakeholder interviews, system walkthrough, complaint and analytics data.
Test
Weeks 1–3Moderated usability sessions; requirements-to-test traceability review.
Prioritize
Week 3Fix list ranked by user impact and effort, agreed in a readout.
Fix
2-week sprintsUser stories with acceptance criteria; tests on every change; sprint demos.
Ship & adopt
ReleaseRelease gates, rollout comms, help content and adoption metrics.
Operate & improve
OngoingMonitoring, evaluation runs for AI features, monthly improvements.
Four weeks, start to readout
Kick-off & discovery
Goals, access, participants recruited, test plan agreed.
Sessions & review
Usability sessions run; requirements-to-test traceability built.
Analysis
Metrics, patterns, ranked fix list, draft stories for the top items.
Readout
Written report, recordings, recommendation and a scoped proposal for what comes next — if anything.
The rules we work by
- Watch before you buildNo roadmap without observed user behavior behind it.
- Everything has acceptance criteriaIf we can't say how we'll know it's done, it isn't scoped yet.
- Tests are part of the fixA change without a test is a change that will come back.
- AI earns its placeWe recommend AI only where it beats the simpler fix — and we prove it with an evaluation harness.
- Say it plainlyReadouts in plain language your leadership can act on.
Tell us what is broken.
Book a 30-minute system review. We listen, look at the system with you, and tell you plainly what we would fix first — no obligation.