Aditya Sonkar

Hey Product Hunt I'm Aditya.

Started my career as a consultant. Suits, CXO calls, global clients. Good money, zero soul. Three years of that and I knew it wasn't it.

So I quit and built RevMeUp from scratch. A social review platform. Spent 2 years getting it to 120K+ users, 50 person team, 1000+ reviews a day. Learned everything the hard way. Unit economics, hiring, product, distribution. All of it.

Then it ended. Joined Monster as a PM. Then Guidepoint where I got to work on some genuinely exciting AI stuff. Built ByteMe on the side because I had FOMO and honestly just needed to keep building something.

We've been talking to dozens of AI agent founders over the past few months and one thing keeps coming up, only few actually knows what it costs them to serve a single customer.

With traditional SaaS it's simple. Seat-based pricing, predictable infra costs, clean unit economics.

With AI agents it's a completely different beast. You've got LLM tokens, voice minutes, vector DB queries, tool calls, cloud compute. All variable, all billed separately, none of it tied to a specific customer or workflow.

You get a monthly invoice from 5 different vendors. You make a guess. And you find out months later you've been losing money on your best customers. Even OpenAI itself uses Metronome just to handle ChatGPT's billing complexity.

We were drowning in data. Page views. Session duration. Bounce rate. Time on site. New users. Returning users. Feature adoption. Support tickets. NPS scores.

None of it told us who was about to leave.

We had retention data. We had churn data. But it was backwards. You only knew someone churned after they cancelled. By then, it was too late.

So we looked for a leading indicator. One metric that predicted churn before it happened.