I wanted to see how feasible vibe coding was... and built this in a week. I don't expect this ProductHunt post to blow up, I just want to share my learnings.
I used Cursor (with Claude 4) to build this app. I wanted to copy https://www.calculator.net/, as I noticed when doing market research, they get ~55M monthly visits to their site.
If you assume at a minimum, a general site (this product) gets $0.20 of CPM for ads... and Calculator.net is bringing in 55M monthly visits, they're bringing in $11k/mo. in revenue off their site. Crazy.
I decided I wanted to make a copycat of their site, expect I wanted to make it a better product. I have better graphics, reactive forms, and great Lighthouse performance numbers (95+ for all categories).
So I used Cursor to vibe code this app pretty quick. Honestly, the most tedious part of the process was QA'ing each calculator and verifying results by cross-checking against other calculators. I decided to not forgo creating unit tests (obviously a red-flag), as the unit tests Cursor was producing was overly restrictive. For example, for some finance calculators, unit tests would fail due to a rounding error ($0.01) and it was becoming more friction than getting a product out.
The cost to build this product was just the domain name (~$40/year). Besides that, the cost to develop this was literally $0, just time. It's a static site, and I'm using all free tools to run my site (Vite for deploying, PostHog for analytics) - no DB needed, no API calls.
alexlarch•7mo ago
I used Cursor (with Claude 4) to build this app. I wanted to copy https://www.calculator.net/, as I noticed when doing market research, they get ~55M monthly visits to their site.
If you assume at a minimum, a general site (this product) gets $0.20 of CPM for ads... and Calculator.net is bringing in 55M monthly visits, they're bringing in $11k/mo. in revenue off their site. Crazy.
I decided I wanted to make a copycat of their site, expect I wanted to make it a better product. I have better graphics, reactive forms, and great Lighthouse performance numbers (95+ for all categories).
So I used Cursor to vibe code this app pretty quick. Honestly, the most tedious part of the process was QA'ing each calculator and verifying results by cross-checking against other calculators. I decided to not forgo creating unit tests (obviously a red-flag), as the unit tests Cursor was producing was overly restrictive. For example, for some finance calculators, unit tests would fail due to a rounding error ($0.01) and it was becoming more friction than getting a product out.
The cost to build this product was just the domain name (~$40/year). Besides that, the cost to develop this was literally $0, just time. It's a static site, and I'm using all free tools to run my site (Vite for deploying, PostHog for analytics) - no DB needed, no API calls.