# Positive Unknown-Unknowns

By Simon Hørup Eskildsen · 2018-03-10 · https://sirupsen.com/unk-unk

When we make decisions, it’s useful to be cognizant of unknown-unknowns. Almost
in every case, we think about unknown-unknowns in a negative sense. If we’re
venturing into unknown territory, we accept that it’s likely we’ll stumble upon
[Black
Swans](https://www.amazon.ca/Black-Swan-Improbable-Robustness-Fragility/dp/081297381X):
improbable events that throw a wrench into our plans. Typically, we’ll draw on
our experience to take the path we figure has the fewest negative
unknown-unknowns. We may choose to stretch something we already know instead of
adopting something new. Brooding on negative unknown-unknowns is extremely
useful, and fairly commonplace.

I think it’s equally useful to invert the traditional thinking about
unknown-unknowns and ask ourselves: How many *positive* unknown-unknowns might
we face with this option? Might we face more positive black swans, than
negative? In effect, what would give us the most positive optionality?

When making decisions, we weigh most strongly the first-order effects. We're not
taught to [systematically think through the second- and third-order
effects](https://www.fs.blog/2016/04/second-level-thinking/). As we get further
away from first-order effects, our ability to predict effects decreases
exponentially. There's a higher chance that we've missed second-order effects,
than first-order effects. These missed effects are what we call
unknown-unknowns. There are too many variables to keep track of and the
interactions between them, while governed by simple rules, become unmanageable
to the human brain. You can attempt to combat this with expertise, but you must
face that you won't catch them all.

![](/images/unk-unk.png)

An example might help. Consider the Internet, which had a fairly niche purpose
at first. Yet, it seemed to many that connecting the planet would be a good
idea. There’s no way that those connecting the globe could’ve anticipated the
amount of positive unknown-unknowns ramifications of the Internet. What they did
project, however, was that the space of unknown-unknowns positives for the
Internet was enormous.

Similarly, if we look at cryptocurrencies today, people are smitten with the
potential for the positive unknown-unknowns (and others by greed). What the Internet,
cryptocurrencies, and the printing press have in common is that they’re
foundational platforms with an enormous surface area for positive
unknown-unknowns. 

I’ve seen positive unknown-unknowns numerous times when people build platforms.
Someone builds something great and simultaneously takes the time to solve the
problem one layer deeper than they otherwise might have. They sense the
potential in increasing the probability of positive unknown-unknowns, by
supplying the vision of a platform. Internally, two years ago we had an
[employees-only single podcast](http://sirupsen.com/podcast). Today, we have
around ten ranging from training, interviews to learn more about how to build an
internal product or history lessons about the company from our executives. When
it was clear that there was an internal podcast *platform*, it exploded. The
first podcast went one level deeper to provide a platform, increasing the
surface area for positive unknown-unknowns.

We will have to remain humble to the fact that often we can’t predict all
effects, positive and negative. We can attempt to reason about their size, but
we won't know for sure. There’s an old Taoist fable that we can interpret as a
story unknown-unknown second and third-order effects:

> “When an old farmer’s stallion wins a prize at a country show, his neighbour
> calls round to congratulate him, but the old farmer says, “Who knows what is
> good and what is bad?”
> 
> The next day some thieves come and steal his valuable animal. His neighbour
> comes to commiserate with him, but the old man replies, “Who knows what is
> good and what is bad?”
> 
> A few days later the spirited stallion escapes from the thieves and joins a
> herd of wild mares, leading them back to the farm. The neighbour calls to
> share the farmer’s joy, but the farmer says, “Who knows what is good and what
> is bad?”
> 
> The following day, while trying to break in one of the mares, the farmer’s son
> is thrown and fractures his leg. The neighbour calls to share the farmer’s
> sorrow, but the old man’s attitude remains the same as before.
> 
> The following week the army passes by, forcibly conscripting soldiers for the
> war, but they do not take the farmer’s son because he cannot walk. The
> neighbour thinks to himself, “Who knows what is good and what is bad?” and
> realises that the old farmer must be a Taoist sage. ”

It is tempting to believe at any of the critical points in this story that you
know what will happen next with certainty. With the most prized stallion in the
land, riches await! Or, when stolen, that you'll never see it again.  While the
series of events in this story seem _highly_ unlikely, it teaches us that
effects will happen that we could never have imagined. The sum of the
probabilities of unknown-unknowns may outweigh the knowns.

You may be looking at two options for a decision that seem equally good. Have
you considered which one has larger optionality long-term? Third-order effects
that you could by no means predict? With a small modification, could you
increase the surface area for unknown-unknown positives? Can you expose even a
fraction of a platform?

Considering positive unknown-unknowns has changed my mind quite a few times in
the past year. Contemplating optionality is *not* about making decisions based
on hope. It is one of many mental models in your arsenal to improve your
decisions. Each model gives you a new vantage point to see the problem from to
help you come to a better decision.