All use cases

Engage / Lifecycle

Wearable-driven rest and recovery nudge

The watch already knows you slept five hours and your strain is high — the agent turns that reading into one timely, caring message instead of a dashboard you have to go and check.

Get aheadBehavioral triggerSystem triggerPushIn-appChatHealth & TelehealthSubscriptionWellness (B2C apps)

What it does

A wearable generates a stream of signals about sleep, strain, and recovery, but most apps leave the member to open a screen, read a score, and work out what to do with it — so the insight that could change today's behavior goes unused. The agent watches for the readings that matter and turns one of them into a single, supportive nudge tied to the actual data: a short night, an elevated resting heart rate, a low recovery score, a streak of high-strain days. It is not a coaching program and not a medical opinion; it is the right small message at the right moment ("you slept under six hours — take it easy and keep water close today") that keeps the member in the habit loop and the app part of their daily life.

How it works

  1. 1
    Trigger. a wearable or health-data signal crosses a meaningful threshold — sleep below a target, recovery low, resting heart rate up, sustained high strain — synced from the device or the health platform.
  2. 2
    Decision. the agent reads the reading against this member's own baseline and recent pattern (not a population average), checks it has not already nudged them today, and decides whether a rest-and-recovery message genuinely helps right now or whether silence is the better call.
  3. 3
    Action. it sends one concise, encouraging nudge on the member's preferred channel, naming what the data showed and offering a light, non-clinical suggestion (ease off, hydrate, an earlier wind-down tonight); every unprompted message passes a quality and tone judge before it is sent.
  4. 4
    Follow-through. the reading and the nudge are written to memory so the next message builds on this one rather than repeating it; if a reply raises a health concern beyond rest and lifestyle, the agent stops giving guidance and routes to the program's clinical pathway; stale nudges are cancelled if the member's data recovers before the message would land.
Example
AgentMorning — your watch logged about 5 hours of sleep last night and your recovery is on the low side. Worth taking today a bit easier: lighter movement, keep water close, and aim for an earlier wind-down tonight. I'll keep an eye on it with you.
Customeryeah I've been wired late all week
AgentThat tracks with what I'm seeing. Want me to nudge you around 10pm tonight to start winding down? Small, but it tends to move the next morning's number.

Configuration

How the agent is wired for this use case.

Triggera wearable/health-data threshold event (sleep_below_target, low_recovery, elevated_resting_hr, sustained-strain flag) synced from the device integration or health platform.
Tools & actions
  • Wearable / health-data integration · ingest sleep, recovery, resting heart rate, and strain readings and detect threshold crossings.
  • Member memory · read the member's personal baseline, recent trend, and goals to judge whether the reading is meaningful for them.
  • Knowledge base · retrieve the program's approved, non-clinical rest-and-recovery guidance for the message.
  • Messaging surface · deliver one judge-gated nudge on the member's preferred channel and schedule an optional follow-up.
  • Clinical-escalation pathway · hand off to the program's clinical route if a reply surfaces a concern beyond lifestyle and rest.
Autonomythreshold detection, baseline comparison, and the single approved rest nudge run unattended under judge gating and frequency caps. The agent gives general rest, recovery, and lifestyle encouragement only — no diagnosis, medication, or clinical instruction from a biometric reading; anything that reads as a health concern is a mandatory hand-off to the clinical pathway under a clinical-safety policy class.
Channelspush · in-app · chat
Escalationa reply indicating a medical symptom, a reading pattern outside safe lifestyle-guidance bounds, or any request for clinical interpretation routes to the program's clinical pathway rather than being answered by the agent.

What you need

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

Signals

wearable and health-platform events (sleep, recovery, resting heart rate, strain) with thresholds, plus member replies.

Data

the member's personal baselines and trends, goals, channel and notification preferences, consent for health-data use, persistent memory, and the program's approved rest-and-recovery guidance.

Guardrails

the agent gives lifestyle and rest encouragement only — no clinical or medication advice from a biometric reading, with a stated boundary and mandatory clinical hand-off on any health concern; judge gating on every nudge; strict frequency caps and quiet hours so it supports rather than nags; explicit consent and compliant handling of sensitive health data; stale-nudge cancellation when the data recovers; visible opt-out.

Metrics it moves

  • dau-mauup: a timely, data-tied nudge gives the member a daily reason to engage with the app instead of ignoring a passive dashboard.
  • adherenceup: gentle, well-timed rest prompts keep members inside the habit loop the program depends on.
  • opt-out-rateheld low: judge gating, tight frequency caps, and genuine relevance keep the nudges welcome rather than spammy.

Related use cases

Engage / Lifecycle

Proactive health-coach companion

the always-on coach this single rest nudge lives inside

Engage / Lifecycle

Proactive coaching companion for fitness and wellness

the broader habit-and-streak engagement for non-clinical wellness

Engage / Lifecycle

Rich-UI self-serve data exploration in thread

when the member wants to explore the sleep data behind the nudge in-thread

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