All use cases

Convert

Conversational storefront with secure checkout hand-off

The whole shopping trip in one conversation on your own site — discovery to assembled basket to your secure checkout, under your brand.

Next best thingBehavioral triggerChatIn-appE-commerce & Retail

What it does

Shoppers are learning to buy by asking, not by browsing — and if that conversation happens on someone else's assistant, the merchant loses the brand moment, the customer data, and the relationship. The conversational storefront brings that experience onto the merchant's own site: the shopper describes what they need, the agent reasons over the catalog, shows product carousels, and assembles the basket inside the conversation. When the shopper is ready to pay, the agent hands off to the merchant's own secure hosted checkout — it orchestrates the entire journey up to payment, and never touches payment itself.

How it works

  1. 1
    Trigger. a visitor opens the storefront conversation and states a need, from a single item to "put a full setup together for me."
  2. 2
    Decision. the agent runs a discovery dialogue, reasons over the full catalog with live pricing and stock, and applies persistent memory of the shopper's past preferences and sizes.
  3. 3
    Action. product carousels and a basket assembled turn by turn in chat; at "I'll take it," the agent passes the assembled basket to the merchant's secure hosted checkout — payment stays entirely on the merchant's PCI-scoped page.
  4. 4
    Follow-through. the order confirmation lands back in the conversation; the dialogue context persists for the next visit; every order that started in the conversation is attributed to it, and the merchant keeps the brand, the data, and the relationship.
Example
AgentTell me what you're putting together and I'll build the basket — budget and constraints included.
CustomerA home espresso corner, mid-range budget, small kitchen.
AgentHere's a compact setup that fits: grinder, machine, and the accessories most owners add. It's all in your basket — when you're ready, checkout continues on the store's secure payment page and the basket carries over.

Configuration

How the agent is wired for this use case.

Triggera visitor opens the storefront conversation and states a need, from a single item to a full setup — from the messaging channel.
Tools & actions
  • Messaging channel · run the discovery dialogue and show product carousels in chat.
  • Knowledge base · reason over the full catalog with live pricing and stock.
  • CRM · read and write customer memory (past preferences, sizes, consent state).
  • Cart API · assemble and update the basket turn by turn (add_to_basket).
  • Payment provider · hand the assembled basket to the merchant's hosted, PCI-scoped checkout; the agent never touches payment.
  • Order-confirmation webhook · return the confirmation into the conversation.
Autonomydiscovery and basket assembly run unattended; payment is out of scope by design — it stays entirely on the merchant's PCI-scoped checkout page, the money-moving boundary the agent never crosses.
Channelschat · in-app
Escalationa need the agent can't satisfy from the live catalog hands off to a human with the conversation context.

What you need

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

Signals

session and browsing context, in-conversation add_to_basket and basket-update events

Data

full product catalog with attributes, live pricing and stock, customer memory and consent state

Guardrails

payment is never handled by the agent — the hand-off goes to the merchant's hosted, PCI-scoped checkout; prices and stock quoted only from the live feed; brand-voice controls; the merchant owns the channel and the customer data throughout

Metrics it moves

  • conversion-rateup: discovery, objections, and basket assembly resolve in one session instead of leaking across browse pages
  • aovup: the agent assembles complete baskets, not single items
  • revenue-per-conversationthe native unit of this storefront: every order is attributed to the conversation that produced it

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