Definition

The share of a signup cohort reaching a defined activation event, computed as a single fraction the team optimizes against.

What the rate computes

The numerator is signups who reached the activation metric; the denominator is signups in the cohort. Teams adopt the rate as the metric their improvement work is answerable to — they design interventions to lift it, read performance against it, and take movement on the rate as evidence of progress on activation.

The structure carries two assumptions: activation is a property of the cohort as a unit, and lifting the rate is the team's primary improvement work.

Two failures inherited from the metric

The activation rate inherits structural problems from how the activation metric is typically chosen and used.

• Cohort-blended: users arrive with different intents, so a singular activation metric collapses multiple intent-cohorts into one signal, and the rate computed against it composites that collapse into a single fraction.
• Not scale-verified: once the metric is adopted, teams optimize the rate without continuously testing whether the metric still correlates with the commercial outcome it was chosen to predict.

The two failures compound. A metric that conflates intents and drifts in its outcome-correlation produces a rate whose movement may have no relationship to whether more users are reaching value.

Four failure modes that follow

1. Threshold-relaxation without value-amplification: interventions move the rate without producing more sustaining customers. Looser gating, friction reduction, removed qualification — these can lift the rate while leaving the cohort's value-realization profile untouched. The activated population grows; the sustaining-customer count stays flat.

2. Signal erosion through homogenization: interventions that smooth differences across the cohort (universal onboarding, single-path activation flow) reduce the variance the analysis depends on to identify which intents and behaviors are converting to sustaining customers. The team optimizes the rate for a cycle or two, then loses the signal that would have shown them what to do next.

3. Forensic-first inversion: teams default to interventional ("what should we change") before forensic ("what's already producing economic gain, and does the metric still track it"). Targeting the rate supports this inversion because the rate is what the team is answerable to; the forensic work feels upstream of "the work."

4. Composition substitution: when the metric won't budge through within-cohort improvement, teams shift to changing who's in the cohort via acquisition-side targeting. The rate moves; the apparatus's selection criteria don't.

Portfolio of milestones replaces the rate

The structural answer is a portfolio of milestones — multiple per-intent value milestones, each validated against a sustaining commercial outcome, replacing the singular activation metric and the singular rate computed against it. Accountability shifts from industry benchmarks and cohort averages to the portfolio's own anchored outcome.

The work shifts from "lift the rate" to building and maintaining per-intent milestones validated against sustainability. Each milestone candidate is verified through ongoing scale verification — not chosen once and assumed to hold. Per-intent paths improve by amplifying the behaviors that lead to milestone completion in each intent-cohort. Multiplier analysis is the operational tool for candidate validation and ongoing verification.

Example

A team lifts a "created 3 projects" activation rate from 22% to 28% by simplifying onboarding and dropping a qualification step. The rate moves; the month-2 customer count is unchanged. They've grown the activated population without amplifying value-realization, and they may have broken whatever correlation the threshold had with sustaining customers — but absent scale verification, they don't know.

See also

• Multi-intent model
• Concentrated leverage