The feature is continuity. The difficult cases define it.

What’s built, what’s planned

Honest about the engine and the skin

Capacity-N is a shared, built engine mode across the schedule family — named once here as the seat law under every occurrence, and not re-argued. The recurrence-expansion, series-enrollment, and roster-continuity surface that makes cohort.software cohort.software is planned, not yet live on this site.

Capacity is a component, not the whole product

The capacity-N seat law is identified as a shared built engine mode. Cohort's proposed use is to carry the series commitment across the occurrences generated by a rule. Those are different scopes. The existence of the capacity component does not prove that this branded site can publish a series, collect tuition, issue an enrollment or keep a roster through an edit. The build ledger keeps the narrower component and the unbuilt composition separate so a reader can judge what evidence is actually available.

For a fictional twelve-person workshop, a useful capacity question is whether the program intends one continuing group or different participants on different evenings. A count alone cannot answer that. Write down the intended membership rule before evaluating the feature. If the program needs a mixture of series members and individual drop-ins, describe it as an unresolved requirement rather than assuming the planned model already supports the combination. This marketing site does not settle a hybrid capacity policy or expose an operational control for it.

Recurrence needs an inspectable set of occurrences

The planned recurrence feature names an RFC-5545 rule and occurrence expansion, with moved weeks and holiday skips represented as exceptions. The product question is what an organizer and participant can understand from that result. For an eight-meeting arc, inspect the intended dates, the time zone and the finish after an exception. A compact recurrence label is useful only if it agrees with the commitment people think they are making. This page describes the proposed model; it is not a recurrence editor or a standards-conformance demonstration.

A common objection is that a calendar already repeats events. That may be enough when the problem is only drawing dates. Cohort's planned difference is the continuing enrollment attached to the series those dates express. The evaluation therefore needs to ask about both the dates and the membership that survives a change. If the dates are correct but the group must enroll again after a move, the central requirement has not been demonstrated. If no continuing membership is needed, the additional series layer may not solve your problem.

Series enrollment must explain the late join

One planned action enrolls a person into the arc rather than requiring a separate booking for every meeting. That sounds straightforward at intake; a week-three join tests what it really means. The facilitator needs to decide what remains available, whether the program still makes sense for the person and what commercial treatment is appropriate. The existing product plan mentions proration, but no working calculation or charge is offered on this site. A clear written policy is an input to evaluation, not a result that the marketing page can manufacture.

Use a fictional late joiner to compare two questions: does the existing group retain its place, and what exactly does the new person receive? The first is about continuity; the second is about the new commitment. Keeping them separate helps prevent a convenient interface from silently making program policy. A prospective buyer can reasonably refuse a pilot whose late-join behavior remains undefined. That refusal is useful evidence of a requirement, and it does not need an actual participant record or a paid transaction to be discussed.

Continuity is not a promise that every change is harmless

The planned roster-continuity layer keeps membership tied to the series through a moved week, a holiday skip and a mid-series join. Its purpose is to preserve the enrollment relationship while the occurrence set changes. It does not mean that a new evening suits every participant, that the program policy is automatically fair, or that completing the arc enrolls someone in the next one. The next intake is a separate opt-in step in the plan. These limits are part of the feature's meaning, not incidental fine print.

Ask for a walkthrough that ends with a before-and-after explanation: what changed, what stayed attached to the series and what still requires a human decision. If the answer is only that everything stays in sync, ask for the actual exception and its consequences. The current site can explain these evaluation questions and show display-only pricing. It cannot demonstrate a live Cohort enrollment flow. Related products and a shared engine family do not remove the need for evidence of this particular series behavior.