All use cases

Convert

Guided selling concierge

"I need tennis equipment" is not a search query — it's the start of a sale, if someone is there to have the conversation.

Next best thingBehavioral triggerChatIn-appWhatsAppE-commerce & RetailHealth & TelehealthTravel

What it does

A shopper arrives with a broad need — equipment for a new sport, a routine for a goal, a week of dinners — and the site answers with a search box and a wall of filters. In a physical shop, an assistant would ask three questions and walk them to the right shelf; online, most of that intent dies in browsing. The agent runs that discovery dialogue at every visitor's elbow: it asks what the need actually is, reasons over the full catalog, and assembles the right product or a complete kit within budget.

How it works

  1. 1
    Trigger. the visitor states a broad need in chat, or browsing behavior shows them circling a category without converging.
  2. 2
    Decision. the agent runs a short needs-discovery dialogue — use case, constraints, experience level, budget — and combines the answers with persistent memory of past purchases and preferences for known customers.
  3. 3
    Action. it reasons over the catalog and presents a concrete recommendation or a complete kit in the chat, each item with the one-line reason it belongs, and hands off to the cart.
  4. 4
    Follow-through. the stated preferences are written to customer memory for next time; if the shopper hesitates on an item, the agent swaps it within constraints rather than restarting; outcomes are attributed per conversation.
Example
CustomerI want to start playing tennis but I have no idea what to buy.
AgentHappy to help. A few quick questions: have you played racket sports before, will you mostly play on hard courts, and do you have a budget in mind?
CustomerComplete beginner, hard courts at my local club, and I'd like to keep it affordable.
AgentPerfect — I've got everything I need. Here's your starter kit: a complete beginner setup for hard courts, focused on comfort, within your budget — each item with the reason it belongs. Want me to add the whole kit to your cart, or walk through it item by item?

Configuration

How the agent is wired for this use case.

Triggerthe visitor states a broad need in chat, or category-browsing behavior shows non-converging exploration — from the messaging channel or the web event stream.
Tools & actions
  • Messaging channel · run the needs-discovery dialogue (use case, constraints, experience level, budget) in chat.
  • Knowledge base · reason over the full catalog with structured attributes, live stock, and price.
  • CRM · read and write customer memory (past orders, stated preferences, sizes).
  • Cart API · add the recommended item or the complete kit in one tap, swapping items within constraints on hesitation.
Autonomythe discovery dialogue and kit assembly run unattended; every add-to-cart is the customer's choice, and the stated budget is a hard constraint the agent never crosses.
Channelschat · in-app · whatsapp
Escalationa need the agent can't satisfy from the catalog within constraints hands off to a human with the dialogue context.

What you need

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

Signals

chat messages with stated needs; category browsing patterns that show non-converging exploration.

Data

full product catalog with structured attributes, stock and price; customer memory (past orders, stated preferences, sizes).

Guardrails

recommendations constrained to in-stock, policy-allowed items; the budget the customer states is respected — the agent never silently recommends past it; in regulated verticals, product claims stay within approved language.

Metrics it moves

  • conversion-rateup: broad intent gets converted in one conversation instead of leaking through filters.
  • aovup: complete kits replace single hesitant purchases.
  • revenue-per-conversationup: every concierge dialogue is tied to an attributable basket.

Related use cases

Convert

In-session hesitation rescue

when the need is implicit in behavior, not stated

Convert

Conversational storefront with secure checkout hand-off

the full shop-in-a-conversation around this dialogue

Upsell & Cross-sell

Complete-the-kit cross-sell

the same kit logic after the first purchase

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