Source of many bad hiring and staffing decisions: failing to recognize the difference between innovation risk and execution risk. Innovation risk = identifying new opportunities and figuring out how to solve them Execution risk = delivering against known opportunities and

A question I get often from analytics, bizops, and strategy people: how do I stay relevant as AI takes off? My answer: There are three parts of the job: 1. Working with data at high scale 2. Synthesizing insights and recommendations 3. Figuring out how to turn recommendations

Often people hit peak productivity and impact at a startup after 2-3 years. Why is this? There are two competing forces for startup employees: 1. Building context about the customer, the business, and what works. It often takes 1-2 years to truly know what you're talking about.

A common experimentation mistake: discarding a newly launched experience (like an onboarding flow) because it underperforms the old version. The old version has been optimized over the course of many smaller experiments to a local maximum. The new version is raw and unoptimized

Redditor uses chatgpt as his personal doctor/nutritionist/trainer. Reversed diabetes in 12 months. Thread is full of people doing the same

the best networking "hack" in tech: 1. join a startup with high talent density 2. make a lot of friends 3. in 10 years, realize they are all now doing incredible things

A failure mode that I and many others have fallen into early in their startup careers: The “all quant, no qual” trap. Here is how it goes: In the short term, it's easier to get quick answers by digging into the data. So you just analyze the funnel, dice the data on a bunch of

The most important A/B test on the planet right now

Pascal's Wager for AI: Three possible scenarios over next 10 years: 1) AI capabilities hit a wall and don't add as much value as people expect 2) AI absolutely supercharges what most humans can do, but stops there 3) We reach AGI; game over for humans contributing

On joining an early stage startup: When you come to America, they tell you the streets are paved with gold. But when you get there, you learn three things: 1. The streets aren't paved with gold 2. The streets aren't paved at all 3. You're going to be the one doing the paving

Companies themselves are a network effect product. People want to join because of the other people who work there. People want to stay because of the learning and relationships and support they get from the other people who work there. Important implications of this that many

"Speed is the new moat" would have absolutely crushed on Twitter in 1999. Execution speed is very important for startups. It's probably the MOST important thing. But speed itself is not a moat. It is a means of making sure you dig a moat before anyone else does. If you're

By far the most common mistake people make when choosing a startup: failing to realize that basically the only thing that matters is whether the startup wins They get excited about a specific role, or being on a certain team, or a better cash comp But virtually everything they

For anyone who hasn't been following the discourse over on LinkedIn, here is a handy list of things that are dead SEO SaaS Product Management Project Management Data Science Software Engineering Software Excel CRM Social Media The Internet FAANG Google Airbnb LinkedIn ChatGPT