Done-for-you offer · Fractional CMO with AI Swarm · save-flow propensity scoring
Done-for-you save-flow propensity scoring for DTC, multi-location franchise, multi-unit service brand, multi-location retail, multi-channel retail, and PE-sponsored portfolio subscription operators — a 4-skill Cancel-Intent-Capture + Save- Propensity-Scoring + Save-Offer-Routing + Attestation bundle on the subscription-agent that scores propensity before the offer routing decision.
The descriptive industry pattern under generic save-flow operation: every cancel intent triggers the same blanket discount or pause offer regardless of save-propensity; high-LTV subscribers receive the same retention offer as low-LTV churn- bound subscribers; save-flow attribution gets logged at the offer level rather than the intent-to-outcome cycle; FTC Negative Option Rule + per-state auto-renew-disclosure + per-state cooling-off-period compliance is approximated rather than per- intent gated. Stripe Billing, Recurly, Chargebee, Zuora, Chargify, Maxio, Paddle, Lemon Squeezy, Recharge, Bold Subscriptions, Ordergroove, Stay AI, Loop Subscriptions, and Subbly ship excellent subscription billing + retention primitives. The 4-skill Cancel-Intent-Capture + Save-Propensity- Scoring + Save-Offer-Routing + Attestation pipeline that scores save-propensity before the offer routing decision and re-emits outcome feedback to the propensity model is operator-side architecture. Completions builds and operates the 4-skill bundle on the subscription-agent. Operator owns every artifact and can in-house at any time.
Published September 24, 2026
Frequently asked
What does the done-for-you save-flow propensity scoring engagement deliver?
Completions builds and operates a 4-skill bundle on the subscription-agent: Cancel-Intent-Capture, Save-Propensity-Scoring, Save-Offer-Routing, Attestation. Skill 1 (cancel-intent capture) captures cancel intents across 10 sources: in-app cancel-flow start, email cancel-link click, SMS cancel reply, outbound-call cancel request, chargeback initiated, payment-method removed, pause requested, downgrade requested, negative feedback flag, NPS-detractor signal. Skill 2 (save-propensity scoring) scores save-propensity before the offer routing decision with save-success probability, save-offer elasticity, per-segment historical success, per-LTV-tier priority, and per-cohort signal as features. Skill 3 (save-offer routing) routes to one of 7 decisions: offer-discount, offer-pause, offer-downgrade, offer-loyalty-credit, offer-concierge-call, offer-no-intervention, escalate-to-human; with a closed-loop edge that re-emits outcome (saved vs. churned, observed elasticity, post-save behavior) back to the propensity model for A/B-cohort calibration and rollback triggers. Skill 4 (attestation) emits an attestation record with attestor-identity, timestamp, WORM-storage commit, chain-of-custody, and per-vertical compliance overlay reference. The per-vertical compliance overlay covers FTC Negative Option Rule + CCPA + GDPR + state-AG policy + CASL + TCPA + per-state auto-renew-disclosure + per-state cooling-off-period. Operator owns every artifact: 4 registries in operator data infrastructure; 4-skill model code aligned with operator retention-engineering team; per-billing-platform credentials (Stripe, Recurly, Chargebee, et al.) under operator billing; per-vertical compliance overlay rule library; propensity-model library; LLM prompts; audit trail. Completions owns the swarm-orchestration knowledge.
Why is propensity-scored save-flow typically operator-side rather than billing-vendor-shipped?
The subscription billing + retention vendors (Stripe Billing, Recurly, Chargebee, Zuora, Chargify, Maxio, Paddle, Lemon Squeezy, Recharge, Bold Subscriptions, Ordergroove, Stay AI, Loop Subscriptions, Subbly) ship excellent billing primitives — pause flows, discount offers, downgrade flows, cancel-reason capture, billing-state webhooks. They do not ship the propensity-scored save-flow layer because the cancel-intent taxonomy reflects operator-specific UX surfaces (in-app cancel flow + email cancel link + SMS cancel reply + outbound-call cancel + chargeback + pause/downgrade flows + feedback flags + NPS detractor signal); the propensity model features integrate operator LTV cohort + segment-historical-success data; the offer routing decisions integrate operator merchandising + retention + customer-service workflows; the per-vertical compliance overlay (FTC Negative Option Rule + per-state auto-renew-disclosure + per-state cooling-off-period) tracks operator legal team approved frameworks; the audit trail integrates with operator data infrastructure. The work that remains operator-side spans four engineering surfaces: intent-engineering for 10-source cancel-intent capture; ML-engineering for propensity scoring with 5+ feature classes and per-segment + per-LTV-tier + per-cohort calibration; routing-engineering for 7-decision-type routing with closed-loop outcome feedback; audit-engineering for WORM-storage attestation + chain-of-custody + 8-jurisdiction compliance overlay. Completions absorbs all four surfaces under one Tier 3 Fractional CMO with AI Swarm engagement and hands the artifacts back at engagement end.
What does the engagement look like across Tier 1 to Tier 2 to Tier 3?
Tier 1 AI Readiness Assessment (2-3 weeks, diagnostic): audits four axes. Tier 2 AI Swarm Setup Sprint (4-8 weeks): builds 4-skill bundle on subscription-agent — completing the 17th 4-skill bundle + Closed-loop-with-feedback family 8th variant + Save-Flow propensity-decision mechanic NEW + Pre-action decision mechanic taxonomy emerging architecture. Tier 3 Fractional CMO with AI Swarm (6-month minimum, 1-2 days/wk embedded): continues operating end-to-end + cross-agent swarm coordination.
Who owns the registries?
Operator owns 100% of every artifact: 4 registries (in operator data infrastructure), 4-skill bundle model code (operator-owned + operator-retention-engineering-team-aligned), per-billing-platform credentials (Stripe + Recurly + Chargebee under operator billing + operator credentials), per-vertical compliance overlay (rule library in operator repo with attorney-approved updates), propensity-model library, LLM prompts, audit trail. Completions owns the orchestration knowledge.
What measurement and reporting does Completions commit to on Tier 3?
Measured against the operator pre-engagement baseline across five workstreams, reported weekly. (1) Cancel-intent capture coverage trajectory — share of cancel intents under active capture across the 10 sources; per-source ingest health (in-app cancel flow + email cancel link + SMS cancel reply + outbound-call cancel + chargeback + payment-method removal + pause + downgrade + feedback flag + NPS detractor). (2) Propensity-score quality — back-tested precision against an operator-labeled validation set; per-cohort + per-LTV-tier calibration curves; confidence-interval coverage rate. Reported against pre-engagement labeling baseline rather than at an 80-percent promised precision target because precision depends on the labeled-validation regime operator runs and the cohort definitions operator approves. (3) Offer-routing distribution — share of intents routed to each of the 7 decisions; per-decision-type observed elasticity. (4) Save-rate trajectory and elasticity — share of routed offers resulting in retention vs. churn, segmented by intent source + LTV tier + offer type; reported against the operator pre-engagement generic-save-flow baseline rather than at a promised 30-percent uplift because save-rate depends on operator tier-benefit design + offer-bridge quality + customer segment composition + competitor activity + macro retention dynamics outside the routing layer. (5) Attestation completeness — per-intent WORM-storage commit rate; attestor-identity + chain-of-custody completeness; per-jurisdiction compliance-overlay rule-evaluation completeness (FTC Negative Option Rule + per-state auto-renew-disclosure + per-state cooling-off-period). Process commitments are firm: weekly intent-source health check; weekly propensity-model recalibration queue (with closed-loop outcome feedback ingestion); weekly per-decision routing rule sync; weekly per-vertical compliance overlay update; weekly audit-trail emission. Outcome targets are not promised because save-rate uplift + propensity precision depend on operator-side labeled-validation regime + tier-benefit design + offer-bridge quality + customer-segment composition + competitive + macro retention dynamics Completions does not control.
How does engagement end and what is the operator transition path?
Tier 3 engagements are 6-month minimum with 90-day notice. At engagement end, Completions transitions back to operator in-house in 30-60 days: operating-playbook hand-off + in-house staff training + 4 registries hand-off + 4-skill bundle model code hand-off + per-billing-platform credentials hand-off + per-vertical compliance overlay rule library hand-off + propensity-model library hand-off + LLM prompts hand-off + audit trail hand-off; Completions credentials revoke immediately.
Engage Completions
Start with the AI Readiness Assessment (Tier 1, 2-3 weeks). Hand off to Tier 2 (4-8 weeks). Continue under Tier 3 Fractional CMO with AI Swarm (6-month minimum, 1-2 days/wk embedded).
Or take the 3-question shape diagnostic first — no email required.