All use cases

Win back

Dormant-buyer reactivation

Thousands of people bought once in the November sale and never came back — and the repeat purchases they would have made are the cheapest revenue the business will never see.

Get aheadTime-based triggerBehavioral triggerWhatsAppEmailIn-appPushE-commerce & RetailSubscription

What it does

Every retail and subscription base carries the same dormant cohorts: promo buyers who purchased a plan, a course, or a tracker on Black Friday and never engaged; repeat purchasers whose buying rhythm quietly stopped; and a warm loyalty base nobody talks to between campaigns. Repeat purchases from these people carry no acquisition cost and sell at full price, which makes even a 15% reactivation uplift direct profit. The agent finds each dormant buyer, works out from their history what would actually bring them back, and re-engages them on every reachable surface — WhatsApp, email, push, or a message waiting in the product itself.

How it works

  1. 1
    Trigger. a dormancy condition fires — a purchase with no engagement after a defined window, a repeat-purchase rhythm broken, a loyalty member with no activity since the last promo cycle — and the agent picks per-customer timing instead of batching everyone into one campaign day.
  2. 2
    Decision. the agent reads the customer's memory — what they bought, what they used, what they ignored, channel consent and preference — and chooses the angle that fits the cohort: activate the thing they bought, restart the rhythm they had, or surface what's new for the loyal-but-quiet. A judge gates every drafted message for relevance, tone, and policy.
  3. 3
    Action. a specific, personal message lands on the customer's best reachable surface and opens a conversation, not a one-way blast — the customer can ask, object, or buy in the same thread.
  4. 4
    Follow-through. replies are handled through to a purchase, an activation, or a clear no; the outcome writes back to memory; outreach is cancelled if the customer comes back on their own first; and reactivation revenue per cohort is measured against a holdout so the program proves its own ROI.
Example
AgentHi Priya — you picked up the annual plan in the November sale but never set up your first program. Your year is still running, and setup happens right here in this chat. Want me to build your first week around mornings or evenings?
CustomerMornings, I guess. Honestly I forgot I had this.
AgentMost people who bought that week did — that's why I'm pinging you. Short morning sessions, starting tomorrow; I'll check in after your first week, and if mornings don't stick we'll move them.

Configuration

How the agent is wired for this use case.

Triggera dormancy condition (a purchase with no engagement after a defined window, a broken repeat-purchase rhythm, or a loyalty member inactive since the last promo cycle), with per-customer timing rather than one batched campaign day
Tools & actions
  • Commerce / subscription platform · read order and subscription events, complete a repeat purchase or restart a plan
  • Loyalty / CRM system · read memory (what was bought, used, ignored; preferences) and loyalty status, write the outcome back
  • Engagement store · read product-engagement signals (or their absence) to pick the per-cohort angle and timing
  • Messaging channels (WhatsApp, email, push, in-app) · send the personalized message on the best reachable surface and run the two-way conversation
Autonomydrafting, channel selection, sending, and conversational handling through to a purchase, activation, or clear no run unattended once a judge gates every unprompted message for relevance, tone, and policy and frequency caps and suppression clear; no clinical or money-moving policy class applies here
Channelswhatsapp · email · in-app · push
Escalationout of scope for routine flow; the conversation runs to a purchase, an activation, or a clear no

What you need

The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.

Signals

order and subscription events, product-engagement events (or their absence), loyalty-program activity, promo-cohort membership

Data

purchase history, persistent customer memory (what was bought, used, ignored; stated preferences), channel consent and reachability per customer

Guardrails

judge gating on every unprompted message; frequency caps and do-not-contact suppression; stale-outreach cancellation on organic return; per-cohort holdout for honest measurement

Metrics it moves

  • reactivation-rateup, because each dormant cohort gets the message that fits why it went quiet, not one generic promo
  • recovered-revenuerepeat purchases from already-acquired customers, measured per cohort against a holdout
  • ltvup, as one-time promo buyers convert into repeat purchasers

Related use cases

Win back

Memory-personalized win-back

the one-to-one version of this volume play

Engage / Lifecycle

Memory-segmented lifecycle marketing

the always-on engine behind these cohorts

Convert

Back-in-stock ping with cross-sell

another high-intent reason to reopen a quiet buyer

See it on your own customer journey

Bring one drop-off, one churn cliff, or one silent segment. We will show you what a proactive agent with memory and judgment does with it.

Book a demo