The Head of Product Who Built a Sales-Closing Chatbot Solo

No developer. No vendor. Just a Head of Product who went in, built an AI chatbot himself, and watched it start converting.

Sanjay Robson runs product and AI transformation at Compare Broadband in Australia. The company needed an AI chatbot. Instead of writing a spec and handing it to a developer, he opened Voiceflow and built it himself.

“I built our AI chatbot myself using Voiceflow. I didn’t have a developer do it. I just went in and built it.”

No waiting for engineering capacity. No vendor evaluation that drags through two quarters. No scoping document that sits in someone else’s backlog. He learned the tool, designed the conversation flows, and shipped a working chatbot. And then something happened that separates this story from the hundreds of prototype stories I hear on the podcast.

The chatbot started making money.

It converts

Most AI projects I see from senior leaders land somewhere between “impressive demo” and “we’ll find a use for it eventually.” Sanjay’s chatbot skipped that phase entirely.

“The chatbot is now making sales for us. It’s not a demo anymore — it’s actually converting.”

That sentence is worth sitting with. A Head of Product, with no developer involved, built an AI product that generates revenue. Not engagement metrics. Not “efficiency gains.” Revenue. The chatbot talks to customers and closes sales. The ROI case writes itself because the chatbot is literally writing its own ROI.

This is what the AI player-coach payoff looks like when it’s concrete and measurable. Not credibility, not “I understand the constraints better now” (though Sanjay gets all of those too). Actual money coming in from something he built with his own hands. The kind of result that makes every “but I’m not technical” objection sound like what it is: an excuse to stay comfortable.

Why he builds first

Sanjay could delegate every build. He has a team. He has the seniority to never touch a tool again if he wanted. But he keeps building first, and his reason is one I hear from the strongest leaders on the podcast.

“When I’m building something myself, I understand the constraints in a way that changes how I spec things for my team.”

That feedback loop is the whole game. When you’ve wrestled with the edge cases yourself, your specs get sharper. Your estimates get more honest. Your conversations with engineering change because you’ve felt the friction they feel. You stop asking for things that sound reasonable in a PRD but fall apart the moment someone tries to implement them.

Sanjay described himself as the person who always goes first. “I am always the one who has to figure out how to do the new thing first before I can help my team do it.” That instinct to lead by doing, not by directing, is what makes the player-coach model work. The team trusts the direction because the person setting it has been in the weeds. They’ve hit the same walls. They know which corners cut clean and which ones collapse.

The revenue test

There are a lot of ways to measure whether getting hands-on with AI is worth a senior leader’s time. Sanjay’s answer is the simplest one: his chatbot makes sales. It’s hard to argue with a revenue line on a dashboard that traces directly back to something you built on your own.

Most leaders I talk to start with smaller wins. A prototype that changes a meeting. A dashboard that reframes a conversation. A workflow that saves the team five hours a week. Those matter, and they compound. But Sanjay’s story is the version where the payoff is impossible to dismiss, because it shows up in the numbers the board actually cares about.

You don’t need to build a sales-closing chatbot to be an AI player-coach. But it helps to know that someone did. No developer. No vendor. No permission slip. He went in and built it, and now it makes money while he sleeps.

The question for every senior leader watching from the sidelines: what’s your version of that?

Listen to the full conversation: Episode 48, The AI Product Leader

(For more, drop AI Career Boost a follow on YouTube!)

Polly Allen is the founder of AI Career Boost and host of The AI Product Leader podcast. She spent years leading AI at Amazon Alexa before building the AI Career Boost Blueprint, an 8-week program for Director+ product leaders becoming indispensable AI player-coaches. Subscribe to The AI Player-Coach newsletter →