All use cases

Retain / Save

Lean-mass plateau window save

Months in, the scale has genuinely slowed and the patient has started worrying they are losing muscle as well as fat — two anxieties at once, in the window where many decide to stop.

Get aheadBehavioral triggerTime-based triggerRisk triggerPushChatWhatsAppIn-appGLP-1 & PeptidesHealth & Telehealth

What it does

There is a later window in a GLP-1 course, around months five to seven, where two things collide: a real weight-loss plateau, and the dawning worry that some of the loss has been muscle rather than fat. Either alone is enough to make a patient question the program; together they are one of the harder stretches to hold. Programs that treat this as "just another plateau" miss the lean-mass anxiety underneath it, and patients quietly conclude the medication is now costing them more than it returns. The agent addresses both at once — it reframes the plateau as expected at this stage, surfaces the protein, resistance-training and nutrition education that actually protects lean mass, logs the non-scale wins that the scale is hiding, and flags any clinically-relevant muscle-loss worry to the care team. It coaches behavior and escalates; it never adjusts a dose.

How it works

  1. 1
    Trigger. the patient is in the month-five-to-seven window and a real plateau shows in the logged trend, or they voice a lean-mass / muscle-loss worry — a combination of time, behavioral and risk signals rather than the calendar alone.
  2. 2
    Decision. the agent confirms the plateau against the patient's own trend and memory (titration stage, training and protein history, what has already been tried), separates a genuine multi-week plateau from normal noise, and decides between a behavioral coaching touch and, where the lean-mass worry is clinically relevant, a care-team surface. A judge reviews every unprompted message.
  3. 3
    Action. a supportive touch lands on a program-owned channel — the plateau-is-normal reframe paired with a concrete lean-mass lever (a protein target, a resistance-training nudge, a nutrition pointer) and a non-scale win the patient may not have noticed — written to re-open the coach conversation, not to lecture.
  4. 4
    Follow-through. replies continue in the coach chat; a clinically-relevant lean-mass concern is surfaced to the care team with the trend attached; non-scale wins and outcomes write back to memory for the next touch; queued touches cancel on self-re-engagement, and the lift is measured against a holdout.
Example
AgentYour weight's held about flat for a few weeks now, and I know at this stage some people start worrying they're losing muscle along with the fat. A plateau here is normal, and the best thing for protecting muscle right now is protein plus a little resistance training. One thing worth noticing too — you logged easier stairs this week, which is exactly the kind of win the scale won't show you.
PatientThat's been my fear honestly, that I'm just getting weaker. I was thinking of stopping.
AgentCompletely fair worry, and you're far from alone in it at this point. Let's set a daily protein target and add two short strength sessions this week — and because the muscle question matters, I'll flag it to your care team so they can take a proper look at how you're responding.

Configuration

How the agent is wired for this use case.

Triggerthe patient is in the month-five-to-seven window with a sustained flat in the logged weight trend (weight_stall_detected) or a self-reported lean-mass / muscle-loss concern, sourced from the program and weight-tracking systems.
Tools & actions
  • Weight-tracking / app backend · detect a sustained flat stretch against the patient's own trend and distinguish it from normal fluctuation
  • Patient record / CRM · read titration stage, training and protein history, and what has already been tried
  • Memory store · log non-scale wins; read coaching history; write the plateau touch and outcome back for the next touch
  • Messaging channel · send the judged reframe-plus-lever touch on a program-owned channel and re-open the coach chat
  • Clinical escalation path · surface a clinically-relevant lean-mass concern to the care team with the trend attached
Autonomywindow-and-stall detection and the behavioral coaching touch run unattended, judge-gated on every unprompted message, using approved coaching content only under a never-new-medical-advice clinical policy; any dose, response or muscle-loss medical question is the clinician's, and sends are suppressed on self-re-engagement.
Channelspush · chat · whatsapp · in-app
Escalationa clinically-relevant lean-mass worry, a persistent stall, or any dose/response concern is surfaced to the care team with the weight trend attached; clinical questions route to a human.

What you need

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

Signals

logged weight entries and trend, sustained-flat detection, self-reported lean-mass / muscle-loss concern, program-month position, patient replies.

Data

weight and body-composition history, titration stage, training and protein history, coaching history, persistent memory, channel consent state.

Guardrails

the agent coaches behavior and escalates only — never adjusts or advises a dose; judge review on every unprompted message; clinical and muscle-loss medical questions always route to a human; approved coaching content only; frequency caps; suppression on self-re-engagement; holdout measurement of save-rate.

Metrics it moves

  • save-rateup, by addressing the plateau and the muscle-loss worry together in the window where both peak, instead of treating it as a generic stall.
  • churndown, by defending a later, named drop-off window with a touch tuned to its specific anxiety.
  • adherenceup, by replacing the "it's costing me muscle" conclusion with concrete lean-mass levers that keep doses on schedule.

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