Thread by @astralwave

Having a persistent model of a network of beliefs > isolated hypotheses.

Network of beliefs helps you do this:
1. When A is true, B will be true.

2. When B is true, C is true 75% of the time.

3. A is true, so C is most likely true.

When things change, such as when A is now false, you are able to update your beliefs and probably no longer want to bet on C anymore.

But hypothesis testing can only tell you C is true.

And it can only tell you C is true at one specific point in time, unless you keep persistent cohorts and do longitudinal tracking.

That’s rarely sustainable nor practical (I’m referring to product development here).

This explains the paradox of why “benevolent dictators” product leaders who lead by intuition tend to create beloved products.

Even though that defies all logic. How can a leader with LESS data/research/rigor make better decisions?

It must follow from that, that Popperian science (aka hypothesis-driven thinking), is actually less rigorous than your garden-variety OODA loop and abductive thinking.

Because our brains are more than capable of tracking a network of beliefs and then updating our prior beliefs when new information surfaces.

It’s simply called “learning”, and we do it naturally.

Now here’s the rub: when teams try to make decisions this way, it completely breaks down.

Shooting from the hip with a dictatorial leader works. Shooting from the hip while coordinating with others: doesn’t work.

It’s because we are extremely bad at vocalizing/formalizing our tacit beliefs so that we can “join up” with other’s.

Shared understanding is hard!

Tools like @RoamResearch or @CausalHQ try to make tools that let us vocalize and share our belief system with others.

But tools alone are not enough.

We need ways of thinking and ways of working to go along with it.

I sense something is on the cusp… something that will absolve certain people from their distrust of A/B testing or traditional user research.