Use case · Returns & exchanges
Answer returns questions. Hand the returns to your team.
The AI explains your returns policy and reads return and refund status from Shopify after verifying the customer. It doesn't create returns, refunds or exchanges: requests go to your team with the order, the reason and the conversation.
Illustration: on Instagram, a customer says the trainers from order 3318 are too small and asks to return them. The AI verifies her, reads the order and its return and refund status from Shopify, quotes the brand's returns policy and hands the request to customer care with the order and the reason. A person from the team replies and takes it from there. Kanesh doesn't create the return or the refund.
Illustrative demo · synthetic data
The problem
Half the returns inbox is questions, not returns.
“Can I still return this?”, “Where's my refund?”, “Do you have it in a 40?” Your team answers the questions and still has to do the returns.
Policy questions on repeat
Windows, conditions, who pays the shipping. The answer is in your policy, but someone has to find it and type it again.
“Has my refund gone through?”
Checking the status means finding the order in Shopify and reading its refunds, one message at a time.
Requests without context
When a return does need doing, the person handling it often has to ask again for the order number and the reason.
How it works in Kanesh
The AI answers; your team decides and does.
No flow needed. You give the AI your policy and your Shopify data, and you tell it when a person has to take over.
Set up
Step 1: Upload your returns and exchanges policy
Windows, conditions, costs and how to send items back. The AI answers from it and quotes it; it doesn't add terms of its own.
Set up
Step 2: Connect Shopify
The native connection lets the AI read orders, fulfilments, returns and refunds. It reads them; it doesn't change them.
AI agent
Step 3: Answer status questions after verifying
“Has my refund gone through?” The AI checks that the person matches the customer on the order, then answers from the return and refund status in Shopify.
AI agent
Step 4: Check the other size for an exchange
For an exchange, the AI can look in your catalogue to see whether the size or colour they want is available.
Handoff
Step 5: Hand every request to your team
Add “wants to return or exchange an item” to the situations that need a person. The AI hands the conversation over with the reason, the order and a summary.
Your team
Step 6: Your team processes it where it always does
The return, refund or exchange is made in Shopify or your returns tool. In Kanesh, assign it to a team, add a tag and reply with a saved reply in the same thread.
Illustration: in a shared inbox, the AI answers a WhatsApp customer with her live order status, then hands a size exchange to the support team with the reason and an internal note, and the agent gets a suggested reply to approve.
Illustrative demo · synthetic data
Kanesh doesn't create returns, refunds or exchanges today, and the AI never issues a refund. It reads the status and explains your policy; a person makes the change.
What you need
What to have in place, and where the line is.
Your policy, your Shopify store and a team that owns returns. The line between the AI and your team is written into the setup.
Connections and setup
Shopify
Native connection with read access to orders, fulfilments, returns and refunds.
Your returns policy
Uploaded to the knowledge base. If it changes, update the document and the answers change with it.
A returns team
A team the AI hands requests to, with saved replies and tags for the usual cases.
Your channels
WhatsApp, Instagram, Messenger, email and web chat in one shared inbox.
Good to know
- Kanesh doesn't create returns, refunds or exchanges. Requests go to your team, who process them in your usual tools.
- The AI sees returns and refunds recorded in Shopify. If your returns tool keeps its data somewhere else, the AI doesn't see it.
- Status is shared only after the AI confirms the person matches the customer on the order.
- Exceptions your policy doesn't cover go to a person. The AI doesn't make them up.
- On Instagram and Messenger, replies follow Meta's 24-hour messaging window.
What to measure
See what the AI answered and what your team took.
Dashboard names, not promised numbers. You see your own figures; we don't publish benchmarks.
AI quality
What the AI answered, what it handed off and why.
Support & operations
Open and resolved conversations and how fast your team replies.
Custom dashboards
A view with the metrics your returns team follows.
Related
The pieces behind this setup.
Product
- AI agentHow the AI answers from your policy and Shopify, and hands off.
- Inbox & helpdeskAssignment, tags, notes and saved replies for the team that does the returns.
- How our AI worksWhat the AI never does, including refunds.
Other use cases
Returns and exchanges questions
Can the AI process a return, a refund or an exchange?
What can the AI answer on its own?
What does my team see when the AI hands off?
Does it work with my returns app?
What if the customer asks for an exception?
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.