All use cases

Convert

Website context-aware product offer

A visitor is flipping between two medication pages trying to work out which one is for them; meet them there with help choosing, not a generic "how can I help?".

Next best thingBehavioral triggerIn-appChatGLP-1 & PeptidesSubscriptionHealth & Telehealth

What it does

A visitor on a weight-medication site opens several product pages and tabs to compare options, then gets stuck deciding which one fits — and a generic chat opener does nothing to move them. That comparison moment is exactly where intent is highest and a small nudge converts. The web-embedded agent reads which product pages and open tabs the visitor is actually looking at and opens with help tuned to that context — offering to walk them through choosing the right medication or surfacing a relevant offer for what they are comparing — instead of a one-size opener. Conversation starters tuned to the page the visitor is on lift engagement where a blank prompt is ignored.

How it works

  1. 1
    Trigger. the visitor views or opens multiple product pages or tabs; the embedded agent registers the page context (which products, which comparison) in session.
  2. 2
    Decision. the agent reads the page and open-tab context, weighs what the visitor is comparing, and decides whether a context-tuned opener is the next best thing — gated by a quality judge so a weak or mistimed prompt is held back rather than shown.
  3. 3
    Action. it opens with a starter tuned to that exact context — offering to help choose between the products on screen, or surfacing a relevant offer — and, if the visitor engages, carries them toward the right option and into the start flow.
  4. 4
    Follow-through. an engaged visitor is handed into the conversational intake; clinical "which is right for me medically" questions are framed and routed, not answered as advice; the agent stays in-session and does not chase the visitor off-site.
Example
AgentLooks like you're comparing two of the weekly medications — happy to help you tell them apart. Want a quick side-by-side on how they differ, or shall I ask a couple of questions and point you to the better fit?
CustomerA side-by-side would help. I can't tell which one's right for me.
AgentHere's the short version of how they differ. Which one is medically right is your clinician's call, and I can take you into a quick eligibility chat that hands them everything — want to start there?

Configuration

How the agent is wired for this use case.

Triggeran in-session signal that the visitor is viewing multiple product pages or open tabs, with the page/product context attached.
Tools & actions
  • App backend · read the visitor's open product pages and tab context in session.
  • Knowledge base · retrieve the context-tuned conversation starter and the comparison content for the products on screen.
  • Messaging channel · open the tuned starter and run the in-session conversation.
  • CRM · record the session context and the offer shown for measurement against a holdout.
Autonomycontext detection and the tuned opener run unattended, with every unprompted starter gated by a quality judge before it shows; the agent hands clinical product-choice questions to the right route and never advises which medication to take. Clinical = surface and route, never diagnose or advise a dose.
Channelsin-app · chat
Escalationa clinical "which is medically right for me" question or a request beyond product information routes into the eligibility intake or to a clinician with the session context attached.

What you need

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

Signals

in-session page-view and open-tab context (which products, multiple tabs), surfaced to the embedded agent.

Data

the product catalogue and comparison content, the context-tuned starter library, consent state, anything already known about a returning visitor.

Guardrails

every unprompted starter passes a quality judge; offers stay on the product-information side and clinical choices route out; frequency caps so a comparing visitor is not interrupted repeatedly; uplift measured against a holdout.

Metrics it moves

  • conversion-rateup, by turning a high-intent comparison moment into a guided choice and a started flow.
  • intake-completionup, by carrying an engaged visitor straight into the eligibility intake.
  • click-through-rateup, because starters tuned to the page the visitor is on are engaged where a generic opener is ignored.

Related use cases

Convert

Conversational intake replacing the long form

the intake an engaged visitor is handed into

Convert

Guided selling concierge

the cross-industry parent of context-aware in-session selling

Engage / Lifecycle

Personalized conversation starters

the memory-driven opener this tunes to live page context

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