Informed Intuition over Deep Rigor

How To Build Zero-to-One Gen AI Products

I’ve recently been thinking a lot about building new AI products - going back to the drawing board and starting something from scratch. There's so much potential for fresh ideas, innovative features, and groundbreaking applications; the idea of exploring something new has me really excited. There is no shortage of new problems to solve and new things to build - the big question is, where to focus!

I was talking with one of our user researchers about how to think about zero-to-one ideas, especially when it comes to engaging with UXR. She mentioned a phrase that really resonated with me: informed intuition over deep rigor. I love this concept as it dovetails with several other things I've been feeling strongly about.

Founder Problem Fit

First, when you're working on zero-to-one ideas, there's a crucial thing to consider: founder problem fit. I say problem fit over product fit because I believe you should start by identifying a problem space that you're deeply passionate about and think has huge potential - this problem isn’t something that will change. The product, on the other hand, is something that you should expect to change and evolve. 

Once you've found that problem, you can figure out the right product to build. It'll go through iterations, refinements, and different attempts, but as long as you stay focused on solving the problem you feel strongly about, this should keep you motivated and driven to keep building your product. The key to finding the right problem to solve can benefit greatly from “informed intuition” - this can help you identify what problems are worth tackling, as well as the right time to address them.

Where Technology is Going

This also plays into my thoughts on technology-led product opportunities. In a world of generative AI, I’ve often encouraged folks to think about where the technology is going, not just where it is at today. When you think about it from this point of view, you open your mind up to the idea that the problems worth solving and the opportunities worth seizing don't necessarily exist today. Then it becomes fun to think about what those future problems and opportunities are – and once again, “informed intuition” can help you out here.

To make this concept more concrete, take the (almost certainly apocryphal, but still interesting) quote from Henry Ford:

“If I would have asked people what they wanted, they would have said faster horses.”

Someone who was able to see that the emergence of cars was about to come (or at least caught on pretty quickly after it started happening) could have better planned for future success by seizing all the new opportunities it entailed such as new products and services: gas stations, tire manufacturers, body shops - rather than continuing to build carriages…

The Downside to Deep Rigor

While thinking through how to build a new product, it can be beneficial to leverage things like user research. Standard approaches often involve talking to a group of people and asking them to pinpoint exactly what their problems and pain points are, then using that to come up with a solution. Another technique I’ve heard to improve products is the use of friction logs, which involves listing the elements that make a tool hard to use, and then directly improving those.

There's a time and a place for these techniques, such as when you're figuring out how to optimize a product. But, that's the exact opposite approach from how I think about building generative AI products from zero-to-one.

Informed Intuition

Informed intuition, on the other hand, is about educating yourself as much as possible, then trusting your gut. So how do I stay informed? Here’s a few of the ways:

• Follow content on social platforms: LinkedIn, Reddit, X, HackerNews

• Attend events

• Talk to coworkers

• Read research papers

• Listen to podcasts

• Use Generative AI in my everyday life, as much as possible

• Write PRDs, strategy docs, and “write to think” memos - then solicit feedback early and often

• Research any potential Markets

• Engage with UXR early and often (using the right techniques)

The goal is to stay as up-to-date as possible, follow trends, have a deep understanding of capabilities and user problems / opportunities… then apply this to a problem that’s worth solving!