ScreenCursor

ScreenCursor is a Chrome extension that records your screen and adds the zooms for you. Every click, drag and keystroke becomes a camera move, so you stop recording and the video is basically finished. $49 once, no subscription, no account, and nothing ever leaves your computer.

No free tier. No watermarked trial. You buy, then you install. Everyone told me that was a mistake, so here's how it went.

The numbers: launched on Product Hunt on 13 September, finished #4 Product of the Day with 198 upvotes. 12 paying customers and $377 so far. Not life-changing money. Enough to learn from.

Four things I learned

1. Product Hunt kept selling for days. I expected one spike and then silence. Most of my traceable sales came in over the following three days, and the page still sends people now. Launch day is the start, not the whole thing.

2. Most buyers decided in under three minutes. They arrive already interested, hit the button at the top of the page, and go. The careful copy further down is for a minority. The top of the page does nearly all the work.

3. "No free trial" costs you sales and buys you something back. People asked about a trial in the comments, and I'm sure some left. But the people who do pay have already decided it's worth $49, and they tell me what's wrong instead of drifting away silently. I'm still torn on it, and I'd genuinely like opinions.

4. A newsletter I'd never heard of sold a copy. I only know because that link happened to be tagged. Without tracking links, that sale would have shown up as "direct" and I'd have learned nothing from it. Tag every link you post anywhere, even the ones you don't expect anything from.

The strangest feedback: someone told me publicly, in the launch comments, that I'm charging too little. He runs a software company himself. I'm still thinking about it.

A question for you: if you sell a paid tool with no trial, what did you do to make buying feel safe? A money-back promise, a longer demo video, something else? That's the next thing I want to change.