Following up on last week's article on forecasters, I want to zoom in on the one tricky issue when you go about plotting where the future will be: exponentials.

The problem(s) with exponentials

Very often, exponentials are a cheap sleight-of-hands used by many forecasters like with AR/VR, blockchain, or any unproven technology, that for the last ten years has been on the verge of accelerating next year...

Next year, we promise, next year will be the great unlock of [insert unproven technology of choice]. Don't dare miss out on the big opportunity. 

But truth to be said, we do face a lot of real exponentials, whether it's in biology (the growth of a bacteria population, the spread of covid) or business (the value created by compounded rates of interest, the way a digital network effect propagates).

And the most famous exponential law to date is undoubtedly in technology:

Moore's law is the observation that the number of transistors in a dense integrated circuit doubles about every two years. Moore's law is an observation and projection of a historical trend. Rather than a law of physics, it is an empirical relationship linked to gains from experience in production. - Wikipedia.

I wrote quite a few times about the difficulty our brain has in dealing with Moore's law. But why? Why are we afflicted with exponential blindness?

It's difficult to say.

I'd venture that our ancestors in African savanna never really had to deal with phenomenons showing compounded accelerations. A saber-tooth tiger will rapidly reach top speed mere seconds before snacking on you. Or, by the time you try to make sense of smallpox, the tribe is dead, making longitudinal epidemiology rather unpractical. In any case, accessing the part of our brain that compute these accelerations is still tough today.

Note that the most common representation of Moore's law is... linear!

No, we don't understand logarithmic axis either (transistor count), but the output is more comfortable for our brains. 

Exponential blindness

Even for us innovation specialists, exponential accelerations are just as painful to deal with. While they fuel so many disruptions and brutal market changes, we still fail to acknowledge them properly to the point that even when they have been steadily happening for many years, we continually fail to recognize them as such. It's just too frightening.

One typical case of exponential blindness is currently ongoing in the energy market. Ramez Naam was already writing about it in 2020 while trying to map the future of clean energy.

Whatever source you're using, the cost-efficiency of producing a kWh of solar energy has been exponential since at least 2010.

Do we understand or accept that as a fact? No.

Even the all-mighty International Energy Agency (IAE) is still applying linear models to this acceleration, consistently missing the mark by several years on where the market will be at.

Dealing with exponentials

No matter what your market is, you're probably dealing with an underlying exponential somewhere or just plain and straightforward Moore's law acceleration with your infrastructure technology. And if we admit that you don't deal with this part of your market's reality, I'd advocate for a simple thing to do: invest a fraction of your resources in exploring what this exponential could mean for your business.

  • In B2B SaaS, what if cloud computing becomes so fast and so cheap that, customer-side, you can run your currently heaviest applications on a simple TV with a $10 dongle?
  • In mobility, what if energy was one of the minor cost contributors to your business model?
  • In healthcare, what if you could produce personalized medicine at a current generic price?
  • Etc.

I'm not saying you should bet big on these incredible turnarounds, but explore them. Get close to a few startups, model a few practical scenarios, or just talk to industrials in other markets that have had to deal with similar tech infrastructure changes recently.

It's a low investment/high pay-off situation or, as we say, proper innovation.

The link has been copied!