So but how do we build software that can be anything without falling into the Zombocom trap? I believe that Bezos interview shows us exactly how.

You start from a systems-level insight ( the internet is growing rapidly )

Then work backwards to a first best customer who has a burning need ( can’t find book ) that is related to your superpower ( internet allows unlimited shelf space ).

This customer acts as the edge-of-the-wedge for expanding into adjacent use cases ( next best customer is music )

Bezo’s approach is sublime. It holds two seemingly irreconcilable insights in tension and transcends either lesser alternative:

A single-purpose product knows its customer, but tends to paint itself into a corner without systems-level insight. Instead of landing and expanding, it ends up going deeper and deeper into its vertical.

A systemic platform has massive growth potential, but tends to flop without a first best customer. It’s sometimes possible to overcome this with a massive push distribution channel, but the truth is, if you build it, they won’t always come.

Starting from that systems-level insight and working back to the first best customer requires two different skillsets, two different ways of seeing.

From the outside, it may look like you’re diving deep into a vertical and building a single-purpose product. Systems thinkers don’t like that. They want to generalize the system.

From the inside, it may feel like you’re spending a lot of time on theory. Product thinkers don’t like that. They want to think about the customer.

But holding both of these contradictory perspectives in your head at once is a superpower. Doing the hard work of systems-thinking up front allows you to uncover exponential opportunities. Working back to a first best customer sharpens your edge-of-the-wedge.

It can be anything, but first it has to be something specific.

Gordon Brander • The Zombocom Problem