Do 10x projects work?

The opinions stated here are my own, not those of my company.

A “moonshot” is all the rage. 10x engineering. Big ideas. Big execution. Dream.

Yet do moonshots actually work?

Obviously yes, you’ll say. After all, the phrase was coined after the original moonshot: literally going to the moon.

The literal moonshot

At the same time, many moonshots we have today seem to fail. In particular, there are two that I’ve recently one into.

Makani

Makani was founded in 2006 to tackle the problem of renewable energy. They decided that the best thing to do was deploy large kites on offshore platforms to generate energy from crosswinds. This was a moonshot project to solve climate change.

There’s a whole documentary they published last year at the time that they shut down the company. They made great advances and had many setbacks. Yet ultimately their problem was cost effectiveness.

Makani was never working in a vacuum. Wind turbines were invented a long time ago and had garnered interest from power companies. Around 2006, this known and proven tech was expensive and inefficient. However, these companies and innovators slowly chipped away at these problems.

There was no moonshot, but rather progressive incrementalism. Rather than 10x engineering, there were a series of small but reliable 10% changes. If you make 24 changes at 10%, you’ll get to about 10x.

1.1²⁴ = 9.85

https://www.smbc-comics.com/comic/energy

Makani was outcompeted. Not because they had a bad idea, or bad execution. Fundamentally their attempts to do a moonshot was not sustainable compared to incremental progress. I’m sure it was a great company. Everyone was driven by a great purpose. I would’ve liked to work there.

Yet what did they accomplish at the end of it all? Had they focused on incremental improvements on wind turbines, they may have ended up with a larger impact on the problem. They could’ve found a few of the 10% improvements that have made onshore wind power generation economically viable.

Onshore wind electricity prices have declined by over 70% in the last decade

Medicare for All

The political fight to expand Medicare to every American has been going on for decades. Is this a good idea? That’s not important in this context. Regardless of public polling or political opposition, it hasn’t happened.

Proponents of M4A will say that this will finally take care of the dramatically increasing healthcare costs that Americans are facing. Medical debt in this country is astounding. Despite this hang wringing, these proponents have not actually managed to do anything. Theory doesn’t put money in pockets or food on tables.

There are a group of healthcare system experts calling themselves the 1% Steps. As the name implies, their approach is through making incremental improvements and reforms. One policy, out-of-network billing, has already been passed by Congress and signed into law.

It’s estimated this change is 5% of commercial health spending, about 1.7% of total health spending. It’s a good improvement in reducing patient costs, and they have many other ways that reduce 1% of costs each. Altogether, they represent some simple but effective reforms.

Even if you think Medicare for All is better, it was this incremental approach that already worked. This policy reduction has done much more in reducing costs than advocating for Medicare for All, and in a much shorter time span.

COVID-19 Vaccines

Moonshots are not impossible. One just happened. As of 2019, the fastest vaccine ever developed took 4 years. Pfizer and Moderna were able to do get their vaccines developed, approved, and deployed in under a year. Even more incredibly, the development time was really no longer than a weekend.

mRNA vaccines are excellent vaccines, a breakthrough technology that not only has been an amazing development for COVID-19 will fuel the next decade of biotechnology.

HIV, Malaria, and even Cancer are the next targets for mRNA vaccines. While the research had stagnated, it came back with something to prove. Johnson & Johnson and AstraZeneca have more common adenovirus vaccines that are being scrutinized around the world. Beyond that, they are not nearly as effective as mRNA.

https://xkcd.com/2400/

What’s in a moonshot?

Moonshots, 10x projects, out-of-box thinking. They can work. In fact, this article has pointed out two successes and two failures. How can we tell which is which?

Here’s a few points that may differentiate them:

  • The current technology is insufficient, regardless of cost. Cost can easily go down with production scale, but you can’t easily do that with time (for vaccines) or weight (for rockets).
  • Experts can explore possibilities with enough funding (often requiring government levels of funding). But when involving non-experts, like politicians, you get too many competing voices to allow for innovation.
  • As a corollary, you do need some sort of external pressure to solve a problem with a sense of urgency or deadline or too much time will be taken exploring perfection.
  • A moonshot needs to be broad enough to allow for a platform for future innovation. It can’t be specific, as a moonshot invention will change our relationships with the world. The Apollo missions miniaturized electronics. mRNA vaccines are broader than COVID-19.

If the technology exists already, but isn’t good enough, you probably shouldn’t use a moonshot. If you can do 24 changes of 10% improvements to accomplish a single 10x change, don’t do a moonshot. It’ll be much easier to pivot and adapt to new things at each step rather than hitting a roadblock and having to scrap all your work.

How do you know when to do which? I don’t know. Breakthrough technologies can happen at any time, but they usually don’t. And we can’t throw a billion dollars or engineer-years into every random idea. Being able to determine what’s worthwhile and what’s not is basically what the field of economics is all about, but economy is far from a precise science.

So I don’t have a good answer. But maybe it’s a good idea to take a step back at what you’re doing and wonder if you could solve it in a different way by leveraging something that already exists. If it doesn’t work, at least you wouldn’t have spent much time or money.

Social Media Expert -- Rowan University 2017 -- IoT & Assistant @ Google