Convert
Intake-quiz abandonment recovery
A prospect stops halfway through your eligibility quiz; the agent brings them back to the exact question they left, before the intent goes cold.
What it does
Someone wanted your program enough to start the eligibility quiz, then stalled on a question about medications, payment, or photos and never came back. In telehealth intake flows, a large share of started quizzes are abandoned mid-flow, and each one is a prospect who already raised their hand. The agent treats an abandoned intake as an open conversation: it reaches out shortly after the drop, answers whatever caused the hesitation, and resumes the quiz from the precise question where the prospect stopped, not from the start.
How it works
- 1Trigger. a
quiz_step_completedstream goes quiet mid-flow; the agent registers an abandonment event with the exact drop-point question attached. - 2Decision. the agent checks consent and contact preferences, recalls anything it already knows about this prospect, and weighs the likely blocker at that step (eligibility doubt, price question, effort) against outreach policy. Every unprompted message passes a quality judge before send.
- 3Action. an SMS goes out within 15 minutes of the stall, conversational and specific to the drop point; if there is no response, a follow-up email lands at 24 hours. Replies open a two-way conversation that can answer questions and complete the remaining steps in chat.
- 4Follow-through. if the prospect finishes the quiz on their own, every queued message is cancelled before it sends. Completed intakes are written back to the client's systems, and the sends go out through channels the client already owns.
Configuration
How the agent is wired for this use case.
quiz_step_completed stream goes quiet mid-flow; the agent registers an abandonment event with the exact drop-point question attached, plus a stall timeout.- Intake/quiz platform · read the drop-point question and quiz answers so far, resume the quiz from that precise step
- Customer memory store · recall anything already known about the prospect and weigh the likely blocker at that step
- Messaging channel · send the SMS within the stall window, the 24-hour follow-up email, and run the two-way conversation in chat or WhatsApp
- CRM / patient-lead store · write completed intakes back and cancel queued messages on self-completion
What you need
The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.
Signals
quiz step events (quiz_started, quiz_step_completed, quiz_submitted) with the drop-point question, plus a stall timeout.
Data
prospect contact details, consent state, quiz answers so far, eligibility rules for the program.
Guardrails
explicit contact consent before any outreach, frequency caps, judge review of every message, no clinical advice in-channel, stale-outreach cancellation when the prospect self-completes.
Metrics it moves
- intake-completionup, by resuming abandoners from the exact drop point instead of asking them to restart.
- conversion-rateup, because completed intakes are the gate to consultations and first orders.
- recovered-revenueup, by converting already-interested prospects who would otherwise be lost at the top of the funnel.
Related use cases
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