Ayni — Managing Bias

Let me paint you a scenario. Imagine this: when Pat was new to investing in private companies, back in 2008, the first investment closed on February 29th. It happened to be a Friday. After a few years, and in spite of the financial crisis, that investment returned 100x. Congratulations, Pat! Outstanding performance for a new investor with limited experience. In subsequent years, Pat's investments haven't fared so well -- there have been 2x and 3x returns on occasion, but most of the companies have failed. What does Pat learn from these experiences? Is the takeaway to wait until 2036 to close the next investment, when February 29th falls on a Friday again?

That may be a contrived scenario, but it illustrates a larger point, and I'd bet you can find examples in your own life. We all establish biases, conscious and unconscious, over the course of our careers, based on our experiences and, more importantly, our interpretation of those experiences -- the causal relationship we assign to them. These experiences build "pattern recognition": our sense of what makes, or breaks, a good investment opportunity. In short, biases underpin our intuition about investing.

"Following our intuitions is more natural, and somewhat more pleasant, than acting against them." -- Daniel Kahneman, Thinking, Fast and Slow

At early stage, that pattern recognition isn't a shortcut around rigor -- it's often the only real tool available. A seed-stage company has no P&L to benchmark, no repeatable sales motion to diligence, no multi-year track record to check a thesis against. Much of the actual work is reading the team: their grasp of the problem, their capacity to execute, their commitment when the easy version of the plan stops working. That is pattern matching, not a workaround for the absence of it.

The point was never to eliminate that instinct. It's to hone it.

Pattern matching that's never checked against outcomes doesn't improve -- it just repeats itself, confidently, whether or not it was ever right. Pattern matching that gets checked against real results, deal after deal, sharpens. The research on what actually separates the best forecasters in any field found the same thing: not raw expertise or unwavering confidence, but calibration -- knowing how much a given signal should move your certainty, and updating hard when an outcome says you were wrong. That only happens if the call gets written down before the outcome is known, and revisited honestly once it is.

Calibration is one half of honing it. The other is simpler: seeing enough real deals, across enough founders and enough outcomes, that more patterns are actually available to draw on in the first place. An investor who has seen a hundred seed-stage teams recognizes something in the hundred-and-first that an investor who has seen ten cannot -- not because the first investor is smarter, but because there is more pattern on file to be matched against. Exposure builds the library; calibration keeps the library honest.

Given how much weight intuition carries in investment decisions, it's important to explicitly declare your biases -- that's part of why we built our platform. Stating the criteria your decisions are actually made on lets you see which biases help you, which are irrelevant, and which quietly work against you.

We believe applying consistent criteria to the investment decision process is itself a way of managing bias -- of reducing the influence of the biases that don't deserve a vote. The act of weighing criteria and rating an opportunity against them makes explicit the parts of a deal that actually matter to you. Combine that with the ability to look back at outcomes -- both the deals you did and the ones you passed on -- and we believe your investment outcomes improve.