Transcripted pricing: $0 conversions at $9/mo

Zero paid conversions in the first week at $9/month.

Not a single one. I shipped the paywall, added the upgrade prompt, watched the metrics. Crickets.

The problem: we built too good a free tier

Transcripted’s free tier gives you local transcription and speaker diarization — fully on-device, no cloud, no cost to us. It’s genuinely better than what Granola charges $18/month for. Granola doesn’t even do speaker diarization on Desktop.

So people try Transcripted. They love it. They keep using it. And they never upgrade — because why would they? The thing they came for is already free.

The paywall isn’t compelling enough

The paid feature is AI summaries. That sounds good in theory — who doesn’t want a meeting summary? But in practice, the summaries we’re generating aren’t meaningfully better than what users could get by copying the transcript into ChatGPT themselves.

The value isn’t obvious enough to be worth $9/month. Especially when the free tier is this good.

What I’m thinking

The monetization wedge has to be something that’s only possible because of the data we accumulate over time — not something you could replicate with a one-off API call.

The candidate: speaker identity persistence. After five meetings with your team, Transcripted knows everyone’s voice. “Speaker 1” becomes “Nathan.” “Speaker 2” becomes “Abdulbaqi.” It learns automatically, it gets better over time, and you can’t get that from a ChatGPT prompt.

That’s the paid tier. Not summaries — identity. Your meeting history, organized by who actually said what.

Still figuring out the right price point. But first, ship the feature. Then test the upgrade rate.

Red bar. The learning is worth more than the revenue we didn’t make.