A Shopify owner answers 80% of customer messages with free AI tools
In plain words
A store owner built a simple assistant with ChatGPT and Zapier's free tier that answers order, shipping and returns questions on the site and Facebook Messenger.
What happened: About 80% of repetitive messages handled; 10-12 hours a week saved; after-hours leads caught (their own figures)
These numbers come from the builder and haven't been checked independently.
Source: reddit.com ↗
How to do it yourself
Our suggested steps. The company didn't publish theirs.
- 1
List the 10 questions you answer most.
- 2
Write one clear answer for each (shipping times, returns, countries).
- 3
Connect your chat and Messenger to ChatGPT with Zapier and paste the prompt.
- 4
Test with 20 real past messages.
- 5
Go live and read the replies daily for a week.
The instructions to give the AI
Paste this into ChatGPT (the free plan is fine), swap the [brackets] for your details, and test it on a few real examples.
This is one part of the setup. It won't connect to your phone, inbox or other apps on its own; the steps above cover that.
Agent instructions (prompt)
You answer customer messages for [store]. Use only these facts: [shipping times, countries, return policy]. If they give an order number, share the tracking link from [source]. If they're angry or it's not covered, say [owner] will reply within [time]. Under 60 words.
[Highlighted] = replace with your own details before you run it.
What you should get back: "Hi! Your order #1043 shipped Tuesday — here's tracking: [link]. Most UK orders arrive in 3-5 days."
Written by Quickly and Good, inspired by this case.
More detail: tools, prices and what you'll need
The company used: ChatGPT + Zapier
You could use: ChatGPT + Zapier (free tier) + Shopify + Messenger
- ChatGPT
General AI assistant: writes, summarises and answers from files you give it.
Free plan; paid plans available · Official site ↗
- Zapier
No-code automation between 9,000+ apps, with AI steps.
Free plan; Professional from $19.99/mo · Official site ↗
Prices from each tool's own pricing page, checked Oct 2026.
What you'll need
- ☐An automation tool (n8n, Make or Zapier)
- ☐Connections to each app involved (email, CRM, sheets, calendar)
- ☐An AI model account/API key
- ☐The agent instructions (prompt)
- ☐Error handling: what happens when a step fails
- ☐Testing with real data for a week
- ☐A human review step before anything goes out to customers
Can I do this myself?
Yes if you have built a Zapier/Make/n8n workflow before. If those words are new, hire a builder.
Similar cases
CMA CGM's AI voice agents resolve 46% of cargo release calls
More than 46% of cargo release enquiries are resolved by the AI agents, and about a third of 4,000 customers called by the sales agent moved on to active discussions.
ElevenLabs' own AI support agent resolves 65% of its tickets
65% of tickets resolved end to end, first reply in about 60 seconds instead of hours, and 70% positive ratings.
Telstra lets an AI agent sort refunds while customers wait for repairs
Refunds are worked out and paid automatically once service is back, which takes admin off 1,000+ staff.