Skip to content
Kanesh

Health & wellness

Product and order questions answered. Health questions go to a person.

Kanesh's AI answers from the product information you upload, your policies and live Shopify orders. It's set up not to give medical advice: under the rules you set, questions about symptoms, medication or health conditions are handed to your team.

Illustration: on WhatsApp, a customer asks a wellness store whether its plant protein is vegan and how many servings a bag has. Inside Kanesh, the AI reads the product in the Shopify catalogue, checks the draft and replies with the published facts and a product link. When the customer asks whether she can take it with her thyroid medication, the AI doesn't answer the question: following the store's rule, it hands the conversation to a person on the team with the reason.

Illustrative demo · synthetic data

The conversations you get every day

Two kinds of questions, and they need different answers.

Questions about the product can be answered from your label. Questions about the person's health can't, and the AI shouldn't try.

  • Product information

    Is the plant protein vegan? How many servings are in a bag?

    What the answer needs
    The information you've published for that product.
    In Kanesh
    AI answer from your product information
  • Directions

    How do I take it?

    What the answer needs
    The directions on your label, quoted as written.
    In Kanesh
    AI answer from your label information
  • Health and medication

    Can I take this with my blood pressure medication?

    What the answer needs
    A person. The AI doesn't answer this.
    In Kanesh
    Handoff by your rules, with the reason
  • Pregnancy and conditions

    Is it OK to take during pregnancy?

    What the answer needs
    A person, every time.
    In Kanesh
    Handoff by your rules, with the reason
  • Order status

    When will my order arrive?

    What the answer needs
    Live status and tracking, shared only after checking who's asking.
    In Kanesh
    Order status after verifying the customer
  • Flavours and formats

    Which flavours does it come in, and is there a bigger size?

    What the answer needs
    Flavours, sizes and stock from your catalogue.
    In Kanesh
    Suggestions from your catalogue

How Kanesh handles them

Answers about your products. People for anything about health.

The AI stays inside the information you publish. The line between product questions and health questions is yours to draw, and the AI follows it.

  • Answers from your label and policies

    Upload product information, directions of use and your policies. The AI is instructed to answer only from them, and each reply is checked against them before it's sent.

    Source: product information you upload

  • Health questions handed to a person

    You define the situations: symptoms, medication, pregnancy, health conditions. When one is detected, the AI hands the conversation to your team with the reason instead of answering.

    Source: your handoff rules

  • Replies checked before they're sent

    Every reply is checked against your data and rules before it's sent. If a statement can't be backed up by what you've published, a person answers instead.

    Source: your data and rules

  • Suggestions by flavour, format and size

    The AI can suggest products from your live Shopify catalogue by flavour, format or size, with the price and a link. Choosing a product for a health need goes to a person.

    Source: your Shopify catalogue

  • Order status after verification

    “When will my order arrive?” gets live Shopify data, including tracking, once the AI has checked the customer is on the order.

    Source: live Shopify orders

  • Campaigns only to people who opted in

    Send WhatsApp campaigns to opted-in customers. Every opt-in is stored with the exact text the person accepted and how they gave it.

    Source: your opted-in audience

What stays with your team

The AI is set up to talk about your products and orders, not about someone's health. Under your rules, these go to a person, with the reason and the context:

  • Symptoms, medication, pregnancy or any health condition
  • Choosing a product for a health need
  • Changing, cancelling or refunding an order
  • Anything your published information doesn't cover
  • Anything your own rules send to a person
  • Your product claims: you approve them under EU rules such as Reg. (EC) 1924/2006, and the AI answers only from the content you approved

The idea for health and wellness

The AI answers about products. People answer about health.

In this category, the line matters more than the speed. You decide which topics always go to a person, and every reply is checked against your data and rules before it's sent.

  • Health, medication and pregnancy questions routed to a person by your rules
  • Answers based on the information you've published
  • Replies checked against your data and rules before sending
  • Opt-ins stored with the exact text the person accepted

Questions from health and wellness brands

Does the AI give health or medical advice?
No. It's set up to answer only from the product information you publish. Questions about symptoms, medication, pregnancy or health conditions are handed to a person on your team, following rules you set. Kanesh doesn't give health advice.
How do we decide what goes to a person?
You define the situations that always go to your team, for example any mention of medication or a health condition. When one comes up, the AI hands the conversation over with the reason instead of answering.
Does Kanesh make our product claims compliant?
No. Each brand is responsible for its own product claims: for foods and food supplements, Regulation (EC) No 1924/2006 on nutrition and health claims, and for cosmetics or medical devices, the rules that apply to them. The AI answers only from the content you approve and upload, and is set up not to add claims of its own; if a statement can't be backed up by your data, a person answers. Review your content before you upload it.
Can we send campaigns on WhatsApp?
Yes, to people who opted in. Every opt-in is stored with the exact text the person accepted and how they gave it. Messages outside the 24-hour window use templates approved by Meta.
Which languages does the AI reply in?
Spanish and English. The interface is also available in both languages.

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.