Investor requests are not like precious gems. Many founders think you only get a certain number of "chips" to cash in. This is wrong — investors WANT to help. By not asking, you're denying them that joy.

"A good strategy is, in the end, a hypothesis about what will work." — Michael Porter

There's more code in Stripe's API to handle edge cases than in the actual main flow. Most people wouldn't do that — but it's one of the things users tell me delights them most about the product.

Planning annually is fiction. We now plan every 6 months for a rolling 12 — high confidence in the near half, low in the far. It reduces thrash and lets you actually move when strategy shifts.

As a rule of thumb: try to get through your Series A with less than 40% investor dilution. Add 10–15% for ESOP and total dilution hits 50–55%. Less is better, but getting the right capital matters more.

Don't assume something working in one product will work in yours. Every system has its own consumer expectations. You have to learn from what's working elsewhere and then rebuild it for your users specifically.

If the marginal cost of writing code goes to zero, how many of the abstractions we've built in software development were designed around human programmer productivity — and should we throw all of them out?

Leaving a full-time job feels shameful — but it shouldn't. Full-time employment for all of adulthood is an aberration in the history of the world. We've normalized something that's historically very unusual.

Some clients need to see a bad direction before they can confirm the right one. Rather than fighting them, I give them a menu where everything mixes and matches. Walking down a wrong path is sometimes how you know the right one.

I made a big mistake early in my career: I thought my job was delivering the go-to-market strategy. It's not. Your job is to land it with sales together. Delivering without adoption means you failed.

This PM salary data is based on actual comp from over 23,000 product managers globally across 7,500+ top companies.

You can evaluate PM communication through all the other interview questions — no special question needed. Just look for clarity, conciseness, and convincingness throughout.

Best YC interview question: 'What have you done since you applied?' The application was 1-2 months ago. The answer should be a long list of things accomplished. If it's not, that tells you everything.

Have every strategy working group member watch 1-3 user interviews. Not to generate insights — but to build empathy and keep the strategy process human-centric rather than abstract.

Reference checks reveal what interviews can't. An amazing H-level PM runs great meetings — but a reference will tell you: 'Actually, the designer solves the hard problems, and the CEO picks what to work on.'

A major strategy consulting firm ran their own internal case study on Superhuman and found it saved their partners 3.3 hours per person per week. The only other tool that matched it was ChatGPT.

Most product owners are innovators or early adopters. You literally cannot understand an early majority user unless you watch them. If you build for yourself, you're building the wrong product.

Platforms didn't win the creator economy by accident. Once you aggregate supply or demand, network effects make it nearly impossible to break in. The VC rush into creator startups was always fighting against this gravity.

Even if you're a senior PM joining a new team, act like you're starting out. You don't know the context, you haven't earned the trust. Ship things, be useful, prove yourself before shaping strategy.

You can't compare. The market is huge. There are hundreds of thousands of companies out there. We serve a particular set of segments. We would do better learning from them to expand to others than watching what competition is doing.

The biggest profit-making scheme of our time is telling people to love themselves and charging them money for it. The best way to actually love yourself is to keep evolving.

In AI, there are major new releases every week, not every year. In mature SaaS you studied Salesforce's annual releases. That's now the equivalent — but you have to keep up weekly, and you have to experience it, not just read the TechCrunch writeup.

When you build with the right abstraction in Notion's codebase, the system works for you. Build the wrong way and the system works against you. That's how you know when you've violated your core architecture.

The Highlight: Start each morning by asking, "What do I want the highlight of my day to be?" Pick one thing — 60–90 min of focused attention. Write it on a sticky note. Everything else in your day gets built around protecting that.

Large companies take 12-24 months to respond to competitive threats. By the time a PM posts about your app, a study gets done, a framing deck gets made, and development starts — you've already won.

Product management is two things: clarity and conviction. Clarity means sifting out everything polluting the core problem. Conviction is your feeling of how the world should be — not certainty, not perfection, but a strong enough belief that you'll fight for it.

If you have perfect strategy but poor execution, you learn nothing. You don't know if the strategy was wrong or the execution was wrong. Great execution with good-enough strategy tells you exactly what to fix next.

More time spent on a product does not yield a better product. More time creates more questions, more complications, and more invented assumptions about what users need. Ship fast, get the feedback that says 'this is premature' — that's priceless.

Journalists in smaller newsrooms have quietly been doing product management, engineering, and coding hacks for years — because there was nobody else to do it. They just didn't have a name for it.

Cursor built 14 different models just to understand how a real developer works. That's not a 'thin wrapper on GPT.' The same phrase was used in the '80s — 'thin wrapper around an RDBMS' — for what became Salesforce.