What $100K on my last project taught me, and what I built because of it
My last project, TRUPERS, made it to $100K in profit. The real driver wasn't a single good idea, it was refusing to ship any change without real data behind it. I was obsessed with onboarding: where visitors actually came from, exactly where they dropped off, whether a page was genuinely confusing or just slow and losing people. Worked, double down. Didn't work, change it. Borderline, keep tuning. Simple rule, but it's the one thing that actually compounded.
That habit followed me into TrukingsI. Most of us obsess over traffic and funnels, but almost nobody has a fast, honest read on what users actually think of the product itself once they're in it. Easy to use, good support, worth the price, does it deliver. That's the blind spot.
TrukingsI is my attempt to close it: listing your SaaS takes a couple minutes, voting is just as simple, and you can share your own link with your own users to build a real ranking off real votes. What you get back is data you can actually act on, the kind of thing that moves revenue instead of guesswork. Free: https://trukingsi.web.app
Happy to answer anything about the TRUPERS side too if anyone's curious.
Clear and practical, thanks. Did anything surprise you along the way?
Given your data-driven approach with TRUPERS, what behavior would validate TrukingsI—founders changing a product decision from the feedback, or users repeatedly returning to track improvements?
Really relatable. How much time do you put into this each week?
There is a tension between the habit that made TRUPERS work and the mechanic in the new one. The discipline you described was refusing to act on anything except real behaviour from real users. A public ranking built on votes collects opinions from people who mostly are not users of the thing they are voting on, which is exactly the signal your old rule would have discarded. The part that survives that test is the share your own link path, where votes come from actual customers. Those two data sets do not belong in one ranking, because mixing them makes the good one unreadable.