I Built an AI Customer Support Assistant for eCommerce businesses

JC — Sat, 29 Aug 2026 05:11:43 +0000

I’ve spent a lot of time working with eCommerce and online businesses, and I kept noticing the same problem:

Businesses spend way too much time answering the same customer questions.

“Where is my order?”

“What’s your return policy?”

“Do you ship to Singapore?”

“How much does this cost?”

“Does this product work for X?”

The questions are simple, but when you have dozens or hundreds of them every day, they become a significant support workload.

And when customers have to wait hours for an answer, there’s also a good chance they simply leave.

So I decided to build something to solve it.

Meet ClerkChat

ClerkChat is an AI customer support assistant that turns a company’s existing knowledge into a 24/7 chatbot.

Instead of having to manually create hundreds of responses, businesses can give ClerkChat their:

Website
FAQs
Product pages
Help centre
Documentation
Policies
Other business information

ClerkChat uses that information to answer customer questions automatically.

The idea is pretty simple:

Your business already has the answers. ClerkChat makes them instantly accessible to your customers.

Why I built it

I didn’t want to build another generic AI chatbot that just gives customers vague answers.

The goal was to make something that is actually useful for businesses.

For example, imagine you run a Shopify store selling skincare products.

A customer asks:

“How long does shipping take to Malaysia?”

Instead of a support agent having to look up the shipping policy and reply, ClerkChat can find the relevant information from the business’s knowledge and answer immediately.

The same applies to questions about returns, products, pricing, sizing, availability and other FAQs.

What businesses get from it

The main benefits are:

⚡ Faster customer support

Customers get answers immediately instead of waiting for a support agent.

🤖 Less repetitive work

Support teams don’t have to spend their day answering the same basic questions.

🌙 24/7 support

The chatbot can answer questions even when the business isn’t online.

📈 More opportunities to convert customers

Website visitors can get answers while they’re actively considering a purchase instead of leaving because they couldn’t find the information they needed.

👥 Scale without immediately scaling headcount

As a business gets more customers, support volume doesn’t necessarily have to increase at the same rate.

I’m also interested in where AI customer support goes next

Right now, I’m particularly interested in the intersection between AI, customer support, sales and business messaging.

There are so many places where businesses communicate with customers today — websites, live chat, WhatsApp, Instagram, Facebook Messenger and other channels.

I think AI can do much more than simply answer FAQs.

It can potentially help businesses qualify leads, recommend products, answer pre-sales questions and handle much of the repetitive communication that currently requires a human.

That’s the direction I’m exploring with ClerkChat.

Building it has been interesting

I’m coming from more of a marketing and business background, so building a SaaS product has been a very different experience from simply coming up with an idea and running ads.

I’ve been learning a lot about:

Building with AI
SaaS product development
AI knowledge bases
Customer support workflows
Product UX
Getting real users to test the product
Turning a real business problem into software

There’s still a lot I want to improve.

I’m sharing ClerkChat here because I’d love to get feedback from other developers, founders and people building with AI.

If you’re running a business and dealing with repetitive customer questions, I’d also be interested in hearing how you currently handle them.

You can check out ClerkChat here:
https://clerkchat.co

Would love to hear what you think. 🚀

DEV Community: JC

I Built an AI Customer Support Assistant for eCommerce businesses