There is no recipe for building a unicorn. Luck and timing do more work than most of us like to admit. But I do have a few ways of thinking about the work that have been useful so far.

The startup is a nested experiment

We often treat startups as a linear path – first you have an idea, then you build, then you scale.

But in reality, a startup is one giant hypothesis.

Your investors (or your savings account) are betting on the macro-experiment: Does this business deserve to exist? But inside that macro-experiment, your daily work consists of hundreds of micro-experiments.

You don’t need to "innovate" on how to file taxes or set up a basic HR function. Just do what works. But for the things that actually determine your survival – finding product-market fit, pricing, or building a distribution channel – there usually isn't a playbook that works.

When the playbook fails, or simply doesn't exist, you have to design a valid experiment: What is the hypothesis? How do we test it cheaply? How do we validate the results?

This mindset is critical. If you treat every feature launch or marketing push as "the thing that will save us" rather than an experiment to learn from,  you set yourself – and your team – up for burnout and false hope.

The founder’s trap: preference vs priority

As founders, we lack objectivity. We all have an internal strategy for how we operate in the world, and we inevitably project that onto our companies. The techies want to keep their heads down and build features. The extroverts want to go to conferences and network.

You have to be brutally self-aware about this, because this bias creates a double-edged sword that can kill your company.

Side A: preference is not competence

Just because you prefer doing something doesn’t mean you are good at it. You might love sales, but if you’re terrible at closing, your preference is hurting the business. You might love coding, but if you create technical debt, you are a liability.

It is painful to admit you aren’t the best person for the job you love, but you have to know it.

Side B: competence is not priority

This is the more dangerous trap. Even if you are great at something, that doesn’t mean the business needs it right now.

You might be an amazing engineer capable of rewriting your MVP in carefully written Rust. But if you don't have users yet, that code quality makes zero difference to the business. You are optimising a non-priority.

Conversely, if you are a charismatic networker, you might spend all day at events. But if you don't have a strong product to sell, all that networking won’t help you. You are just being busy, not effective.

You have to constantly ask: Am I doing this because it moves the needle, or because it makes me feel comfortable?

Culture is ownership (not beanbags)

Because a startup is a series of experiments, you cannot micromanage. You need people who can run their own experiments.

Culture isn't about office perks or "vibes". It’s about how work gets done when you aren't looking.

At my company, QA Sphere, we're remote-first, so I couldn't look over anyone's shoulder even if I wanted to – and I don't. I'd rather give people clear problems, trust them to run with it, and judge the outcome. Different founders have different management styles, but I've rarely seen a startup micromanaged into success.

You must hire people who act like owners. These are people who identify a problem, design a solution, and execute it without waiting for permission.

This leads to the hardest part of the job: Hiring slowly and firing fast.

Your first few hires shape your DNA. If you hire someone who needs a playbook when you need an experimenter, it won’t work. Interviews rarely reveal how someone acts under pressure. If you realise you’ve hired the wrong person-someone who drains energy rather than generating it – don't drag your feet. Letting people go is hard, but keeping the wrong person is lethal.

Funding: pick the model for your reality

You can’t run experiments without a runway.

Bootstrapping is romanticised, but it is incredibly difficult to build an ambitious product if you are constantly terrified of next month's server bills. Most startups need external fuel to get off the ground.

However, funding is just a tool to buy you time to run your experiments. At the early stage, investors aren't looking for a perfect P&L; they are looking for proof that you are solving a real problem for real people. Passion helps, but a working prototype and data from your "experiments" earn trust.

Final thoughts

The funding environment comes in waves. Sometimes capital is abundant, sometimes it's tight, but the constant is that shallow stories don't hold up for long.

What does last are teams that solve real problems and founders who are honest with themselves about what actually matters versus what just feels good in the moment.

Treat your startup like a lab. Watch your own biases. Keep running experiments, and keep adjusting based on what you learn.

Good luck.

For more startup news, check out the other articles on the website, and subscribe to the magazine for free. Listen to The Cereal Entrepreneur podcast for more interviews with entrepreneurs and big-hitters in the startup ecosystem.