- In short
- Judging lifecycle phase-gating means determining which artifact must exist to gate a transition to the next lifecycle phase and judging whether that gate is currently satisfied. Every transition has a gating artifact -- a translation table gates design, a complete documentation set gates handoff, a governance table gates monitoring, an outcome document gates a decision to expand or iterate. Judging a gate means checking the artifact against its completeness criteria, not merely confirming the artifact exists.
The skill that ties the whole domain together
This is the highest-difficulty knowledge point in the domain and its synthesis capstone. The CCAR-P exam asks you to take a single deployment, place it correctly on the five-phase lifecycle, determine which artifact must exist to gate the next transition, and judge whether that gate is actually satisfied. Doing it well requires drawing on every earlier task statement at once, which is why it sits at the top of the difficulty scale.
The organising idea is that phase transitions are not automatic. Each one is gated by a specific artifact, and you may only advance when that artifact is complete. Knowing which artifact gates which transition is half the skill; judging whether the artifact meets its completeness criteria is the other half, and it is the half candidates most often get wrong.
- Judging lifecycle phase-gating
- Determining which artifact gates a given lifecycle transition and judging whether it is complete enough to permit advancing. Each transition has a gating artifact -- translation table gates design, complete documentation gates handoff, governance table gates monitoring, outcome document gates expand-or-iterate -- and the judgement is against completeness criteria, not the mere existence of the artifact.
The four gates
Each transition has its gate. A discovery translation table gates the move from discovery into design: without the constraints captured and traceable, design has nothing sound to build against. A complete documentation set gates handoff: without decision logs, a control register, and a document that passes the completeness test, the system cannot be safely transferred. A governance table gates the move into monitoring: without signals mapped to triggers, owners, and actions, monitoring has no decision layer. And an outcome document gates the decision to expand or iterate: without a before-and-after business metric and its control, there is no basis for that decision.
Existence is not completeness
The evaluate-level core is that judging a gate means checking the artifact against its completeness criteria, not confirming that an artifact of the right type exists. A governance table with blank owner fields exists but does not gate monitoring. An outcome document full of technical metrics but missing the business before-and-after and its control exists but does not gate expansion. A documentation set that is a detailed diagram with no rationale exists but fails the completeness test and does not gate handoff. In every case the artifact is present and the gate is still not satisfied, because presence and completeness are different questions. The whole skill lives in refusing to accept the first for the second.
Why it is a synthesis skill
Phase-gating cannot be done from a single earlier concept, which is what makes it synthesis. To judge the design gate you need discovery knowledge; to judge the handoff gate you need documentation knowledge; to judge the monitoring gate you need feedback-loop knowledge; to judge the expansion gate you need outcome-document knowledge, which itself rests on trade-off and control-register thinking. Placing one deployment correctly means holding all of these at once and applying the right completeness criteria to the right gate. This is why the exam reserves it for the highest difficulty: it is the point where discovery, trade-offs, feedback loops, documentation, and outcomes all have to be live in the same judgement.
What the exam trips candidates on
The first trap is treating the existence of any document as satisfying the gate, without checking it against the specific completeness criteria for that phase. The exam presents a deployment with an artifact of the right type that is quietly incomplete, and rewards judging the gate unsatisfied because the criteria are not met.
The second trap is advancing a regulated deployment past a gate because the technical work is done, while a required governance or documentation artifact is still incomplete. The exam sets up finished-looking engineering alongside a missing control-register row or governance rule, and credits recognising that technical completion does not substitute for the gating artifact's completeness.
Common misreadings to avoid
Misconception
If the artifact that gates the next phase exists, the deployment can advance.
What's actually true
Misconception
Once the technical work is done, a regulated deployment can move to the next phase.
What's actually true
How this shows up on the exam
A scenario places a deployment somewhere on the lifecycle, describes its artifacts, and asks whether it can advance and what gates the next transition. The reliable method is to name the gating artifact for that transition, then judge it against its completeness criteria rather than its existence, and to refuse advancement when technical work is done but the gating artifact is incomplete, especially for a regulated deployment.
This synthesis rests on the five-phase lifecycle and its activity mapping, and it applies the completeness standards of the translation table, the documentation completeness test, the governance table, and the outcome document all at once. It is the point where the whole domain comes together.
A regulated deployment has finished all engineering and has a governance table, but several rows list a signal and trigger with blank owner and action fields, and a required scheduled-audit row is missing. The team wants to move into the monitoring phase. Can it, and why?
People also ask
Which artifact gates each phase transition?
Why is the existence of a document not enough to pass a gate?
Why is phase-gating a synthesis skill?
Watch and learn
Official Anthropic Academy lessons first, then hand-picked walkthroughs. Videos load only when you press play.
No videos curated for this concept yet
We are still curating the best official and community videos for this topic.
Official prep for this domain
Anthropic's own free prep module for this part of the syllabus, on the official prep course. Free with an Anthropic Academy sign-in.
References & primary sources
Master this concept with Archie
Practice it inside an adaptive study session. Archie, your Socratic AI tutor, tracks your mastery with Bayesian Knowledge Tracing and schedules the perfect next review.