This week I released the 120th episode and here are four lessons that really stood out and changed the way I think about DevTools startups:
1. Be an expert on the problem, not the solution (via Adam Frankl)
One instinct you probably have is to write code.
Another is to talk about the code you wrote (your solution).
But, if you're like me or most of my friends, you are probably underinvesting in deeply understanding the problems your users face.
And also, if you are scratching your own itch, it doesn't mean you get a free pass.
Adam Frankl suggests setting up a "Technical Advisory Board" (TAB) to stay close to your users: basically a set of people in your target audience who you talk to regularly about their problems. Learn more about it in the episode we did.
2. Experiment! Your startup is not the same as your competitors, so stop copying them (via Ant Wilson from Supabase)
Try every channel and strategy to see what sticks. Then double down.
No expert would have advised Supabase to focus on memes, but that's what worked!
So, stop looking for answers in textbooks and start throwing stuff at the wall.
3. Invest disproportionately in packaging (via Zeno Rocha from Resend)
Building core logic is one thing, but how you present it matters just as much.
Don’t just push to GitHub and say job done.
Obsess over your README, your tagline, and how you explain your value.
Take Zeno’s 2015 project, Clipboard.js, for example—it skyrocketed to 10,000 stars in just a couple of months.
Why? Because although he spent only a few days writing the core logic, he spent weeks refining the documentation, crafting an engaging landing page, and ensuring the README was clear, concise, and compelling.
4. Different is better than better (via Gonto) .
It's hard to be better. And even if you are better, if you're just a bit better no one will care anyway. You'd need to be like 10x better to really stand out.
Therefore, it's probably better to focus on what makes you unique and lean into it.
+ do the uncomfortable thing
And one more thing… this is just my lesson from my own experiences: do the uncomfortable stuff.
Whether it’s emailing people, posting on Reddit, or writing blog posts—it’s awkward but worth it. Don't lie to yourself and spend all your time building one more feature.