Analyzing 13 Years of My Investing History: Part One

I spent last weekend doing something I’ve wanted to do for years but never quite had the data or time for. Now that it’s been 13 years (sigh!), I finally sat down and analyzed my entire deal log from the last decade-plus.

Not just the winners, but the passes, the misses, and the long, messy updates. AI helped a bit too. 😉

Here’s a snapshot of a few early learnings and more to come…

Early visibility is incredibly hard in this business.

Take Company X. They’re now one of my top performers – a multi-billion-dollar category leader. In their seed deck, they projected meaningful revenue by Year 2. The actual result? Less than 10% of the plan.

That pattern showed up again and again. Most of my successful investments missed their plan badly in the first two years. In Company X’s case, those years were spent navigating regulatory complexity, working through early founder issues, and building a product that turned out to be far harder to ship than anyone expected. By Year 5, they were growing into the industry standard.

Catching the tailwind takes time.

We invested in Company Y – a backend software company just before a massive sector-wide crash. For nearly two years, volume was negligible. But when the market turned, they were well positioned. They went from a tough business to catching the wind and ultimately became a unicorn with real platform value. A lot of this business is simply about hanging around long enough to weather the storm.

Valuation vs. traction was another interesting lesson.

I found a clean “A/B test” in the data: two investments from the same vintage, at the same entry price (~$25M post money valuation, a bit later than our usual investments).

Company Z had single-digit millions in revenue and customers who genuinely loved the product, but it wasn’t obvious how it would scale. It was consumer. It was hard goods. Today, it’s a massive category winner.

Company Q had a repeat founding team, a slick deck, and was very much part of the zeitgeist. We spent four years searching for product-market fit and never quite found it.

It was a reminder that paying for progress, even early signs of product-market fit are super valuable. That may seem obvious, but in the echo chamber in which I operate often the curve of hope is valued more than that of hard numbers.

This is a business of small numbers, limited data, and very long feedback loops. That makes it dangerous to over-index on anecdotes, case studies, or early signals.

Still, having more than a decade of actual data makes it easier to at least try to make sense of what’s really going on.

Other VC nerds out there learning from their own data?