How a Solo SaaS Founder Automated Customer Support

**Lucas** (0:01)
So there's this bootstrap SaaS founder I came across, runs a tool for freelance designers, does about $12,000 in monthly recurring revenue, and for the longest time he was handling all customer support himself.

**Luna** (0:14)
Just him, no part-time help?

**Lucas** (0:16)
Just him. And it was eating up three to four hours a day.
He got to a point where he knew he needed to either hire someone or find a different way. But hiring a support person, even part-time, would have cost him around 3,000 a month, which is a huge chunk of his MRR.

**Luna** (0:33)
Right. That's a quarter of his revenue gone just to answer emails.

**Lucas** (0:38)
Exactly. So instead, he built a custom AI chatbot.
He used the OpenAI API, hooked it up to a private knowledge base he built from his own help docs and past support tickets, and connected it all through Zapier to his help desk. Total cost, $87 a month.

**Luna** (0:56)
$87 vs $3,000. That's a no-brainer on paper. But how well did it actually work?

**Lucas** (1:04)
Surprisingly well.
His first metric was response time.
It went from an average of 12 hours to under 2 minutes, and his customer satisfaction score stayed above 95 percent. The chatbot was handling about 70 percent of incoming questions end-to-end.

**Luna** (1:22)
What about the other 30 percent? The tricky ones?

**Lucas** (1:26)
Those got escalated to him.
But because the chatbot was filtering out the routine stuff, he only had to deal with maybe four or five complex tickets a day. That took his support time down to about 30 minutes.

**Luna** (1:40)
So he effectively bought himself three hours back every day for less than $100. That's the dream.

**Lucas** (1:47)
It is. And there's another interesting side effect. He noticed that the chatbot started answering pre-purchase questions on the pricing page. Things like, does this integrate with Figma?
Or can I cancel anytime? And his refund requests dropped by 40%.

**Luna** (2:06)
Because people knew what they were getting before they bought. That's actually really smart, using support automation as a conversion tool.

**Lucas** (2:14)
Yeah, exactly. The chatbot became a kind of friction reducer.
He told me that before, customers would email, wait half a day, get an answer, maybe buy, or maybe not. Now they get an answer instantly and often purchase right after.

**Luna** (2:30)
I love that. But I have to push back a little. What about customers who hate talking to bots?
Some people just want a human.

**Lucas** (2:38)
That's a fair point. He handled it by making the chatbot feel human, it used a friendly tone, it always offered to connect to a human if the customer asked, and it never pretended to be a person.
It introduced itself as the support bot right away.

**Luna** (2:53)
Transparency is key. If someone feels tricked, they'll be twice as irritated.

**Lucas** (2:59)
Absolutely.
And he tracked sentiment. The number of customers who explicitly asked for a human was under 5 percent. So the vast majority were fine with the bot.

**Luna** (3:11)
That's a really useful data point. For a micro-SAAS with under 500 customers, this seems like a playbook worth copying.

**Lucas** (3:20)
I think so too.
And honestly, this kind of thing is part of why we keep these shows going. A couple of dollars a month from listeners who find them useful goes a long way to keeping them ad-free and independent. If you've gotten something out of episodes like this, buymeacoffee.com/fexingo is where that support lands.

**Luna** (3:39)
Yeah, it genuinely makes a difference. Even a few bucks helps cover hosting and keeps us focused on the content.

**Lucas** (3:46)
So back to the chatbot, the other thing he did was train it on his refund and cancellation policies.
He said that before the bot, a lot of cancellation requests came in because people didn't understand how to use a feature. The bot would walk them through it and a bunch of those customers ended up staying.

**Luna** (4:04)
So it's not just answering questions, it's proactively retaining customers.

**Lucas** (4:10)
Exactly. He shared a specific example.
A user wanted to cancel because they couldn't figure out how to export their designs. The chatbot asked a few clarifying questions, then sent a step-by-step guide with screenshots. The user replied, Oh, thanks. I didn't see that button. And they stayed on for another six months.

**Luna** (4:32)
That's a really clean example of automation done right. Not replacing the human touch, but extending it.

**Lucas** (4:40)
And the cost savings are real. Let's do the math.
$3,000 a month for a human versus $87 for the bot. That's about $35,000 a year saved. For a bootstrapped founder, that's the difference between profitability and breaking even.