Use case · Pre-sale questions
Answer the question that comes before the order.
The AI recommends from your live Shopify catalogue, with price and link, and answers size, material and shipping questions from your own documents. Bigger or custom purchases go to your sales team, and into the pipeline.
Illustration: on Instagram, a customer looks for a running shoe under 90 euros as a gift. The AI searches the live catalogue and suggests two pairs in stock, with prices and links. For his wide feet, it adds a note from the brand's fit guide. When the customer asks about twelve pairs for a running club, a situation the brand reserved for a person, the AI hands the conversation to the sales team and a teammate adds it to the pipeline.
Illustrative demo · synthetic data
The problem
A shopper with a question is a shopper about to leave.
Product pages can't answer everything. The questions come in by message, often in the evening, and wait until someone is free.
Answers that arrive late
A question about fit or fabric that waits until the next morning gives the shopper time to buy somewhere else.
Recommendations from memory
Your team knows the range, but not always what's in stock in that size right now.
Big orders in the same queue
A request for twelve pairs, or a custom piece, sits in line with “what's your shipping cost?” and gets the same attention.
How it works in Kanesh
Your catalogue and your guides, answering.
No flow needed. Connect the store, give the AI your documents, and decide which conversations belong to your sales team.
Set up
Step 1: Connect your Shopify catalogue
The AI recommends products from your live catalogue, with the price and a link to the product page.
Set up
Step 2: Upload size guides, materials and care
Upload your guides, FAQs and policies. The AI answers from them, not from general knowledge about products like yours.
Set upBeta
Step 3: Add pages from your website
Index the pages you choose, such as a size guide or a shipping page, into the knowledge base.
AI agent
Step 4: Recommend and answer, with links
The AI suggests products that match the question and answers the follow-ups. If the answer isn't in your catalogue or documents, it hands off instead of guessing.
Handoff
Step 5: Send big purchases to sales
Describe the situations for a person, such as bulk or team orders, custom requests or high-value purchases. The AI hands those conversations to your sales team with the reason.
Your team
Step 6: Track them in the pipeline
Your team moves the conversation through a sales pipeline tied to it. If the shopper goes quiet, follow-ups (up to three) respect your quiet hours.
Illustration of a knowledge base: two uploaded documents, two pages from the brand's website (one indexed, one still to review) and a brand rule.
Illustrative demo · synthetic data
What you need
What the AI needs to answer well.
The AI can only be as good as the catalogue and documents you give it. That's the whole setup.
Connections and setup
Shopify
Native connection. The catalogue the AI recommends from, with prices, stock and links.
Your documents
Size guides, materials, care, shipping and returns, uploaded to the knowledge base.
Pages from your websiteBeta
The pages you choose, indexed into the knowledge base.
A sales team and pipeline
A team for handoffs and pipeline stages that match how you sell.
Good to know
- Website indexing covers the pages you add, not your whole site automatically. Check which pages are indexed.
- The AI answers from your catalogue and documents. When the answer isn't there, it hands off rather than guessing.
- The AI doesn't create discount codes or take payment in the chat. The purchase happens in your store.
- The AI replies in Spanish and English.
- On Instagram and Messenger, replies follow Meta's 24-hour messaging window.
What to measure
See where questions turn into orders.
Dashboard names, not promised numbers. You see your own figures; we don't publish benchmarks.
Sales
How conversations and the orders that follow them move, day by day.
Funnel
Where conversations stop on the way from first message to order.
AI quality
What the AI answered, what it handed off and why.
Follow-ups
Which follow-up step gets the replies.
Related
The pieces behind this setup.
Product
- AI agentCatalogue recommendations, your knowledge base and handoffs.
- SalesHow sales teams use Kanesh to turn questions into orders.
- Inbox & helpdeskThe shared inbox and the pipeline tied to each conversation.
Other use cases
Pre-sale questions
Where do the product recommendations come from?
Can it answer size, material or care questions?
Does it read my whole website?
What happens with a bulk or custom order?
Can the AI offer a discount?
Free diagnosis
See what your conversations are leaving on the table.
We review up to 180 days of your WhatsApp history and send you a written diagnosis within 48 hours. Free, and nothing is connected until we've agreed in writing how your data is handled.