Subscription

Subscription details live entirely in the heads of account managers and scattered email threads. A CSM says 'I think Acme is on the annual plan, maybe $50K?' but there is no written record of plan tier, billing cycle, or renewal date. When finance asks 'what subscriptions renew next month?' the answer requires calling three people and hoping they remember.

None — AI cannot perform any subscription analysis, renewal forecasting, or billing validation because no subscription records exist in any system.

Subscriptions are tracked in a spreadsheet or basic CRM with customer name, plan name, and a rough dollar amount. But the details vary: one AE logs 'Enterprise Annual $120K,' another logs 'Ent - 10K/mo.' Billing frequency, seat counts, discount terms, and auto-renewal clauses are captured inconsistently or not at all. Finding out whether a customer's contract includes a price lock requires digging through the original email chain.

AI could parse the subscription list for basic revenue totals, but cannot reliably calculate MRR, forecast renewals, or detect billing mismatches because subscription records lack standardized fields for pricing structure, term length, and contractual commitments.

Subscriptions are entered into a billing platform or CRM with consistent fields — customer, plan, seat count, unit price, billing frequency, start date, and renewal date. Finance can run a report showing 'all annual subscriptions renewing in Q3.' But subscription records don't link to usage telemetry, support ticket history, or expansion pipeline. Answering 'is this customer likely to renew?' requires the CSM's gut feel, not system intelligence.

AI can generate renewal calendars, calculate MRR/ARR by segment, and flag subscriptions with unusual pricing. Cannot predict churn risk or identify expansion opportunities because subscription records are isolated from usage and engagement signals.

Subscriptions are managed in a billing platform with enforced fields linked to customer health scores, usage analytics, and support history. A revenue operations manager can query 'show me all enterprise subscriptions renewing in 90 days where product usage declined more than 20% last quarter and open support tickets exceed five' and get an accurate, current result. Subscription amendments are versioned with change history.

AI can score renewal likelihood, recommend pricing adjustments based on usage patterns, and flag at-risk subscriptions for proactive outreach. Cannot yet autonomously adjust subscription terms because pricing rules and approval thresholds are not machine-readable.

Subscriptions are schema-driven entities with formal relationships to pricing rule engines, customer contracts, usage entitlements, revenue recognition schedules, and expansion playbooks. Every subscription state transition — creation, upgrade, downgrade, renewal, cancellation — is captured with timestamps and actors. An AI agent can evaluate 'what is the optimal renewal offer for this customer given their usage trajectory, contract history, and competitive risk?' and produce a quantified recommendation.

AI can autonomously manage routine subscription operations — processing renewals within approved parameters, recommending tier changes based on usage, and generating expansion proposals. Full autonomous subscription lifecycle management is possible for standard scenarios.

Subscriptions are living event streams — every usage change, billing cycle, support interaction, and customer engagement signal generates a real-time event. The subscription is not a static billing record but a continuously updating digital twin of the customer relationship. Self-service upgrades, API-triggered plan changes, and automated renewals all flow into the same structured stream automatically.

Can autonomously manage the full subscription lifecycle in real-time — renewals, expansions, downgrades, and retention interventions execute based on continuous customer signals without human intervention for routine transactions.

Ceiling of the CMC framework for this dimension.