View plans

How it works · a worked example

One product problem.
Follow every decision.

What happens after you bring an idea? Follow an onboarding-improvement mission from the first question to work you can inspect.

Walk through the mission Write your own brief
From evidence to a judgement
From evidence to a judgementEvidence to Option A; Evidence to Option B; Option A to Challenge; Option B to Challenge; Challenge to Decide. Dissent stays in the record. High-impact decisions escalate to you.EvidenceResearch SynthesisOption AProduct recommendationOption BAlternative approachChallengeReviewers + veto rulesDecidePanel / your call
  1. EvidenceResearch SynthesisMission input
  2. Option AProduct recommendationFrom Evidence
  3. Option BAlternative approachFrom Evidence
  4. ChallengeReviewers + veto rulesFrom Option A + Option B
  5. DecidePanel / your callFrom Challenge
Dissent stays in the record. High-impact decisions escalate to you. A visual explanation of the process—not a screenshot of a live mission.

The example

“Why do people leave during setup?”

Start with a question. Bring what you know. Keep the assumptions visible.

Example mission · Improve onboarding

Who contributes. Who challenges. Who decides.

Follow the direction of a judgement through the team. This is an illustrative workflow, not a running customer mission.

Research Planning → Web Research → Evidence Verification

A source register with citations and gaps

Research Synthesis sends the evidence brief to Product Discovery.

Missing evidence stays visible; synthetic interviews do not count as customer evidence.
Evidence becomes product memory
Evidence becomes product memorySources to Evidence; Brief to Evidence; Evidence to Decision; Decision to Artefacts. Source references and decision history travel with the work.SourcesYou + ResearchBriefProduct ownerEvidenceResearch SynthesisDecisionPanel + youArtefactsDesign + Engineering
  1. SourcesYou + ResearchMission input
  2. BriefProduct ownerMission input
  3. EvidenceResearch SynthesisFrom Sources + Brief
  4. DecisionPanel + youFrom Evidence
  5. ArtefactsDesign + EngineeringFrom Decision
Source references and decision history travel with the work.
Illustrative mission · no real customer results

Step 1 of 6

Start with a problem worth understanding.

You run a small software product. People reach setup, but you don't yet know why some leave. A useful mission starts with a question, not a predetermined redesign.

You bring

Investigate where new users struggle in onboarding. Compare options and draft an experiment. Do not change the live product.

  • Clarify the audience, desired outcome and non-goals.
  • Identify which notes or product data are available.
  • Propose bounded work and a credit ceiling.
You receiveA mission brief and a plan to review.

Your call: Does this scope answer the question you actually have?

Frame
Specialists contribute. Reviewers challenge.
Specialists contribute. Reviewers challenge.Objective to Research; Objective to Product; Research to Challenge; Product to Challenge; Challenge to Your call. This is a sample team structure; enabled capabilities vary by mission.ObjectiveYouResearchEvidence briefProductOptions + trade-offsChallengeSecurity + EngineeringYour callHuman approval
  1. ObjectiveYouMission input
  2. ResearchEvidence briefFrom Objective
  3. ProductOptions + trade-offsFrom Objective
  4. ChallengeSecurity + EngineeringFrom Research + Product
  5. Your callHuman approvalFrom Challenge
This is a sample team structure; enabled capabilities vary by mission.

A team with a purpose

The right disciplines.
A shared brief.

Research examines the sources. Product compares the options. Design makes the journey tangible. Reviewers challenge the output. Each contribution has a job and a place in the record.

The 150-role catalogue describes the wider workforce. Missions use enabled capabilities; it does not mean 150 agents run on every request.

ResearchProductDesignEngineeringReview

Your part in the process

Delegate the work.
Keep the judgement.

A useful AI team makes it easier to decide. It should surface uncertainty, pause for consequential calls and let you send work back.

  • Set the objectiveName the audience, outcome and constraints.
  • Inspect the evidenceAsk what is supported, inferred or still unknown.
  • Approve the directionAccept, change or reject the recommendation.
Approval controls the next handoff
Approval controls the next handoffProposal to Review; Review to You; You to Next task. The moving trace illustrates a handoff; it never approves real work.ProposalSpecialist teamReviewEvidence + risksYouApprove or returnNext taskApproved scope only
  1. ProposalSpecialist teamMission input
  2. ReviewEvidence + risksFrom Proposal
  3. YouApprove or returnFrom Review
  4. Next taskApproved scope onlyFrom You
The moving trace illustrates a handoff; it never approves real work.
From approved scope to reviewable output
From approved scope to reviewable outputPRD to UX; PRD to Architecture; UX to Prototype; Architecture to Prototype; Prototype to Inspect. A reviewable prototype is distinct from runtime-tested, deployed software.PRDProductUXDesignArchitectureEngineeringPrototypeBuild + QA reviewInspectYou + project export
  1. PRDProductMission input
  2. UXDesignFrom PRD
  3. ArchitectureEngineeringFrom PRD
  4. PrototypeBuild + QA reviewFrom UX + Architecture
  5. InspectYou + project exportFrom Prototype
A reviewable prototype is distinct from runtime-tested, deployed software.

Know what you are receiving

A prototype is a prototype.
A test needs evidence.

Current programmes produce documents, decision records, a lightweight HTML prototype and an export. Generation mode is disclosed. Static checks do not mean the application ran, and a release plan does not mean it was deployed.

Get the review templates

The next loop

The result should inform
the next question.

Connected outcome review is the next PSI capability. For now, use the experiment brief to record the expected result, then add what you actually observe. Learning begins with an honest comparison.

Download the experiment brief
Observation informs the next question
Observation informs the next questionDecision to Build; Build to Observe; Observe to Next question; Next question to Decision. Connected measurement and automatic learning remain planned.DecisionExpected resultBuildInspectable artefactObserveYou supply resultsNext questionProposed mission
  1. DecisionExpected resultFrom Next question
  2. BuildInspectable artefactFrom Decision
  3. ObserveYou supply resultsFrom Build
  4. Next questionProposed missionFrom Observe
Connected measurement and automatic learning remain planned.

Your turn

Bring the question
your product needs answered.

Start with your idea