Acting Before You're Ready
Why the hardest product decisions show up before the understanding does
I write this newsletter to share my learnings about founder-led growth and bootstrapping, gained from my past 10 years of building AI startups and my current journey with MadScript. AI is pushing the cost of building products and launching startups down toward zero. What's left is authentic POV, based on lived experiences, shared compellingly.
Why Seeing the Signal Still Isn’t Enough
I’m still sitting with the discomfort from last time. Making a call before I have full information feels early. The moment I say we’ll build something, the cost becomes real. A timeline appears. Team staffed. Resource allocated. Tradeoffs I’ve been deferring come due. Some days I wish I know everything ahead of a decision. In practice, it often shows up after I’ve made it.
Early on at MadScript, I’d get on a call, hear a clear pain point, and decide in the moment to solve it. That was only possible because I knew exactly what we could deliver and how the team worked. The commitment felt credible. It also meant the second the call ended, I had to turn that promise into scope and quality and a plan that would work. That part still makes my stomach tighten, even when it works, because the responsibility is immediate and specific. Building turns a vague idea into an obligation with a clock on it.
Why the hesitation lingers even after the signal is obvious
Founders don’t hesitate because they’re short on information. Most of the time, we have more than enough signal. The hesitation comes from knowing that acting turns fuzzy feedback into something you’ll be measured against. Once I say yes, I’m accountable for showing a working thing to a real person on a real timeline. Even if they’re supportive, they’ll expect it to do what we agreed it would do. From that moment, the room for interpretation narrows. It’s no longer research. It’s delivery.
Delay feels safer because it preserves options. If I wait for the next conversation or one more set of notes, I can keep the decision in a reversible state. No one is chasing a date. No UI exists to critique. No alerts will pop up on Monday morning if I change my mind. What makes this tricky is that it doesn’t feel like fear. It feels like being responsible. But the real trade is between reversible ambiguity and irreversible commitment. I’ve felt that exact friction on sales calls. I’ve also felt the pull to stay in the comfort of gathering input rather than accept the clock that starts when I make a promise I intend to keep.
The failure mode of “one more data point”
A common trap I’ve fallen into is the slow drift into more calls, more notes, more synthesis, and somehow no decision. It’s seductive because every additional conversation feels productive. You get new phrasing of the same problem. You get a better understanding of the edge cases. You tidy your doc. The belief underneath is that if you accumulate enough input, certainty will follow. In reality, it rarely does. Certainty doesn’t grow like a straight line. It plateaus, then jumps after you commit. A concrete direction forces details to surface that were invisible when everything was still hypothetical.
I’ve seen this play out firsthand when we committed early and let shipping expose what the product actually needed to become.
What changed in how I decide now
I haven’t solved the discomfort. I have changed my threshold for making decisions. If the same issue shows up across different conversations, I treat that as enough to act. Not a full mandate, not a spec with every edge case handled, just enough alignment to say we’re going to build in that direction. Because I stay close to the product and to customers, I can make that call without outsourcing the consequences.
I also treat early commitments as directionally correct, not final. That makes it possible to start without pretending we know everything. The follow-through is where the decision gets real. We’ve had features where the first pass felt wrong, but the rework was the tuition for moving early. When we decided to lean into collaborative editing, our first version of previews was basic. It did the job but didn’t reduce the back-and-forth as much as we hoped. That wasn’t a sign that the decision was flawed. It was a sign that we were still translating the insight into a usable flow. The iteration is where judgment gets exercised, and that only happens after the commitment is made.
Another change is how I weigh customer promises. If I can credibly commit, I’ll do it on the call. If I can’t, I’ll say so directly. It’s the same reason founder-led sales works in the first place. The person who understands the product deeply and feels the customer’s pain can bridge them immediately. I try to keep that bridge intact for as long as possible.
The real cost of waiting
It’s easy to focus on the risk of building the wrong thing. That risk is real. What’s harder to notice is the cost of waiting. Momentum decays. Early customers who were excited to lean in lose interest when they don’t see movement. Learning cycles slow down because there’s nothing concrete to react to. Internally, the team drifts into abstract debate because there’s no artifact to orient around. It shows up as missed trust and a calendar that looks full but produces less value than it should.
Embrace the discomfort from not knowing everything
I don’t expect the discomfort to disappear. It’s part of the job. Seeing the signal is not the hard part. Acting on it is hard because it creates responsibility and second-order consequences you can’t cleanly reverse. That’s not a flaw in the process. That’s the work. My shift has been to accept that understanding will lag, that some rework is the cost of moving early, and that staying close enough to decide and deliver is what keeps us moving fast.
As AI drastically reduce the time and cost to build product, it is more important than ever to build quickly and leverage real user feedback to guide your direction instead of just guessing.