sales · AI shopping assistant

an AI salesperson that knows your catalog better than anyone.

heysteff is an AI platform for customer support and sales across WhatsApp, Instagram, Messenger, Gmail and Shopify; Steff is the AI agent that runs it. As a shopping assistant, it doesn't improvise: it recommends with your store's real stock, understands a product photo or a payment receipt, and follows a one-minute voice note without asking the customer to type it out instead.

◆ the problem it solves

A human seller can't memorize a full catalog with stock changing by the minute. Steff can — because it doesn't memorize anything, it checks your live system.

Recommends with live stock data, never invents

When a customer asks if something is available in their size or color, Steff doesn't guess: it checks your catalog and inventory through tools connected by API, at the moment it answers. If a product sold out five minutes ago, the answer already reflects it — no promising something you can't ship.

This runs on the same native integrations as the rest of the platform: Shopify, WooCommerce, TiendaNube, or any in-house system connected by webhooks. The agent is also trained with a knowledge base carrying your brand's tone, policies and rules, so recommendations sound like your business, not a generic bot.

Upsell, cross-sell and bundles that close

A good salesperson doesn't just confirm the product a customer asked for: they suggest the accessory that pairs with it, the alternative size if the one they wanted is out, or the bundle that already makes sense for that cart. Steff does the same, inside the same conversation, using rules you define for your catalog — without sounding like a forced script, because the suggestion comes from the same thread, not a separate message.

Since it's the same agent handling support, the upsell lands at the right moment: after the customer's real question is resolved, not before. Nothing gets pitched before the original need is actually addressed.

Reads photos and payment receipts

The agent's vision lets it look at a photo a customer sends — a product they want to match, a label with a code, a piece of furniture they're trying to coordinate with — and answer using that image as context, not just text. For manual payments, the same capability reads transfer receipts and helps confirm payment without anyone on the team reviewing screenshots one by one, even when the receipt arrives as a blurry phone photo instead of a clean scan.

Understands long voice notes

Many customers would rather send a voice note than type, especially when the question is long or specific. Steff transcribes the voice note and answers based on what the customer actually said, so nobody wastes time asking them to repeat it as text and the customer never has to.

Between live-stock catalog answers, vision and transcription, the assistant covers the three formats a customer actually buys through in chat: text, photo and voice — answering 24 hours a day without the business needing to be awake to respond.

FAQ

Does it make things up?

No. Steff answers with real data from your catalog and stock through tools connected by API, not with information it invents. If it doesn't have access to a piece of data, it says so and escalates to a person instead of guessing.

How does it learn my catalog?

It connects to your store (Shopify, WooCommerce, TiendaNube) or to your system via API/webhooks, and checks live catalog and stock data every time it answers. You can also train it with your own knowledge base of rules, tone and policies.

What happens if it doesn't know something?

It escalates the conversation to a human based on the rules and keywords you define, instead of forcing an answer. The team can take over the chat in one click with copilot mode.

Also relevant

◆ next step

Connect your catalog and let Steff sell with real data, not scripts.

Get a demo See pricing