Failed experiments are good experiments
Failure is just data.
The Lonely Blue Pixel wouldn’t be complete without a space fact in each edition. From now on, I’m starting each post with one.
Random Space Fact 🪐: When a massive star collapses and fails to sustain its nuclear fusion, it explodes into a supernova, scattering elements across the universe that eventually form new stars, planets, and even life itself. Every atom in your body heavier than hydrogen was once forged in the heart of a dying star. We are literally made of stardust.
Sometimes the most spectacular transformations come from collapse.
Failure is just data
I never talk about failures publicly. There are way too many to count, and honestly, it’s often embarrassing.
But looking back, those so-called failures shaped me more than any success ever did.
At one point, I got into the seafood export trade. Unbelievable when I think about it now. How does someone who builds software for a living start their career trading fish across borders?
At 20 years old, I was negotiating with fish farmers and cargo agents and getting quotes for refrigerated shipments. I was knee-deep in red tape, customs regulations, and supply chain logistics.
Well, as you can imagine, I stopped after a year.
Profit margins were thin, problems were endless. I was competing against people who’d been in the field for 30 years, had deep networks and battle-tested intuitions.
At the time, I thought it was my biggest failure. A meaningless chapter in my life. A complete waste of time. I felt like a disappointment to my family who wanted me to scale up the corporate ladder.
But in hindsight, that failed venture gave me an education I couldn’t have paid for. School never taught me how to negotiate with freight forwarders or structure exclusive deals with suppliers. I learned more about international trade, the impact of currency fluctuations, and supply chains from that one year than from every lecture in Economics 101 combined.
Right after that came my other failures. I failed to scale an EdTech product I tried to launch in Asia. I failed to close investment deals even though I had a successful prototype with major VC interest, just because I didn’t have a technical co-founder and didn’t know how to find a good one. At the time, I didn’t know how to code, and AI coding tools didn’t exist for me to vibe code my way into faking it. I later failed at many other things, the list of which could turn into a book.
The thing about failure is that society teaches us to see it as something terrible to be ashamed of. But when I came across the concept of treating failure like a scientist in one of James Clear’s articles, my entire mindset shifted.
When scientists run experiments, they form multiple hypotheses and rule out the failing ones. In fact, failing experiments are a good thing. When you rule out the impossible, you’re closer to what’s possible. So why not do the same for our careers or for figuring out what we’re good at?
Seth Godin puts it perfectly: “Failure is simply a cost you have to pay on the way to being right.”
Here are some actionable ways to develop this mindset:
1. Keep a “lessons learned” log After each failure or setback, do a brain dump: What happened? What did I learn? What will I try differently? This shifts your brain from dwelling on the failure to extracting value from it. I mean, we are human after all. We’re allowed to be sad if something doesn’t go our way, but we shouldn’t let that define us for too long. My failed EdTech venture taught me I needed a technical co-founder and trustworthy partners who are willing to learn on the go.
2. Set experiment goals Instead of “I will launch a successful product,” try “I will test 3 different product ideas with real users, reach out to them for feedback, post about it on socials and see the response.”
I’m actually in the middle of such an experiment now with one of my event-focused apps called Disco! Building is so easy now, thanks to the explosion of AI tools. It has never been easier to just build a prototype and seek feedback. In fact, I’ve had lots of people reach out to me on various platforms like LinkedIn and Reddit to be a beta tester and early user for their products, and I’m doing the same for some of my own.
3. Study failure autopsies I’ve made a habit of reading stories about failed startups and pivots from the original idea to learn the lessons behind them. Was it cash flow problems? Operational? Marketing? Seeing how others extract lessons normalizes the process. I understand that reading failed startup stories is miserable and kinda scary, but look at it through a lesson-learning lens.
4. Time-box your experiments Give each venture a predetermined timeline. “I want to achieve X by Y months or years, whatever makes sense.” This creates a natural checkpoint to evaluate what’s working and what’s not, preventing you from endlessly chasing something that isn’t viable while still giving it enough time to prove itself.
The scorecard isn’t wins vs. losses. It’s experiments run vs. experiments avoided. Keep your lessons log and set your timelines. Run the tests and when something collapses that is alright. Now you know, and you’re closer to what works.
Experiments I am currently running
Disco: An app that connects people at in-person events based on their goals for attending. Users create profiles sharing what they’re looking for, whether that’s finding co-founders, landing clients, or learning from experts, so you can meet the right people at the right time. Early validations and feedback are strong, and I’m going to mess around and find out if this thing will actually work. I already have a list of beta testers, and hopefully in 2026 we can do a beta launch.
I don’t know if Disco will work out. I don’t know what failure #67 will look like. But I do know I’d rather have a long list of experiments behind me than a short list of things I was too afraid to try. What experiments are you running? What are you learning from them?