TTL 7

“Small bets” are not enough to build large-scale software

👋 Good Morfternight, this is Paolo Belcastro, with the seventh issue of TTL: Tools & Thoughts for Leaders.

This is an odd one

I’m writing this from a windowless room in an airport hotel in São Paulo 🇧🇷

My first plane was delayed, and the second one was not, and that’s why I’m stuck here. Just another reminder that we live in a stochastic world, where everything is probabilities, nothing is ever certain.

Coincidentally, this is the topic of today’s newsletter.

We’ll be dealing with the uncertainty of payoffs in large-scale product management.

Every decision is a bet, bets are hard

A stochastic world, where everything is probabilities, nothing is ever certain, means that we can never know for sure if a bet we make is going to succeed.

Every decision is a bet.

We never know if we made the right decision.

In the field of finance, specifically trading and venture capital, this inconvenience 😊 is addressed by multiplying asymmetrical bets.

Asymmetrical bets are bets where if you lose, you don’t lose that much and if you win, you win a lot.

And therefore, every win will cover for many, many of the losses.

In the financial world, this approach was popularized by Nassim Nicholas Taleb, in his Incerto series.

That’s the core principle behind the strategy behind Universa (the investment management firm where Taleb is the scientific advisor) financial success. Lose often, win big a few times.

In the VC world, this approach is the standard, with some cases where it’s pushed to the extreme, like YCombinator.

Ok now, how does that translate to producing software?

Small bets are trending

If you log onto Twitter, the impression you will get is that the same principle of asymmetric bets translates to producing software quite directly.

Also, quite recently the movement has gotten a new boost of attention thanks to Lex Friedman’s podcasts with Pieter Levels, an indie-hacker that works alone on a portfolio of different software businesses.

I like these communities, but there is an obvious difference between indie-hacking and leading large teams.

Indie-hackers work alone on small-scale software.
Large teams work together on large-scale software.

So being an indie-hacker is a special situation in the sense that when you work alone, you do not plan to scale your product to a massive size, and that’s perfectly fine, every app doesn’t need hundreds of millions of users.

And so you can go from a succeeding bet to an actual product you’re happy with by investing a few more weeks or months of your work once you have confirmed that that’s what users want.

In other words, small bets are a great tool to minimize risk and maximize chances to succeed in building a lifestyle business.

While small bets are an effective way to reduce risk and can lead to incremental improvements, they are not sufficient on their own to drive large-scale innovation. To achieve transformative results in large-scale software development, these small bets need to be part of a broader, more ambitious vision.

Local vs. Global Maxima

When you’re working with a large team on a product designed for massive scale, relying solely on small bets can lead you to chase local maxima.

This means you’re making incremental improvements that refine what already exists but miss out on game-changing innovations. Imagine you’re on a hill, trying to reach the top. If all you do is climb upwards, you’ll reach the peak of that hill, but you’ll never discover the much taller mountain two kilometers away.

To reach that higher peak, sometimes you need to descend first before climbing again. In the context of large-scale product development, sticking to small, safe bets keeps you on that initial hill, never venturing towards the larger, more impactful opportunities.

In some scenarios, particularly in competitive markets, pursuing local maxima can provide crucial short-term gains. However, for large-scale, long-term success, it’s essential to pursue global maxima—those game-changing innovations that redefine the product landscape.

The blueprint for innovating on large-scale software

So, is there a way to use asymmetrical bets to drive large-scale, long-term innovation?

Yes, but you have to do a few other things first:

You need to start with a vision that’s big—something that could take years of your teams’ work to realize and fundamentally change your users’ lives.

↳ To do this, you need more intuition than people like to admit.

Then, you build a road map in a way where you know that the pieces you build, the small pieces you build along the way, lead to that larger vision but also bring inherent value by themselves.

↳ To do this, you require more experience than people like to admit.

And so now what you do is that you’re in a system where you keep building fairly small pieces of the product that you can test and get quick feedback loops. And at the end of the day, even if you end up abandoning the original vision, you shipped many features that added intrinsic value to the product and improved your customers’ experience.

You’re still doing small asymmetrical bets. The low risk is represented by the fact that each one is a relatively modest investment, and brings marginal improvements to the experience on its own. The oversized reward comes from the compounding factor due to all these small features, leading to a cohesive and revolutionary user experience.

If you eventually reach the place you wanted to go, or even a different version of that original vision, that’s where you have the big payout.

A good example of this is Apple’s Continuity

The vision for Apple’s Continuity is: to create a seamless ecosystem where all Apple devices work together effortlessly.

But it started with something as small as a Universal Clipboard, copy something on your Mac, paste it on your iPhone.

Then they let you use your iPad as an external monitor for your Mac, and your iPhone as a webcam. In the current macOS/iOS betas (available to everyone in a couple of weeks), there’s this feature called iPhone mirroring, where you can basically run an image of your iPhone on your Mac, so that when you are at your desk you can use both devices from the same screen.

These are relatively small features, but they add immediate value to the customer AND compound towards this “one, seamless experience”.

Now, this is culminating in the ability to use your other Apple devices seamlessly in the Apple Vision Pro. Now, the Vision Pro might be the future of computing, or an ephemeral product, but even if Apple stopped working on it, the continuity features they already shipped would have still added a large value to their ecosystem.

Leaving the Apple ecosystem, a close example to me is how we approach Jetpack with our team:

Our vision is for Jetpack to provide the best WordPress experience globally, allowing users to benefit simply by installing it.

On the way, we’ve broken Jetpack into modular components. This allows us to offer a strong set of individual products like Backup, Security Scan, AI Assistant, VideoPress, Social sharing, enhanced Search, Anti Spam. By providing these Jetpack features individually, each brings immediate value to your site.

We can then bundle them to provide a Security, Performance, or Growth experiences. Eventually, they contribute to our ultimate goal: a fully cohesive and consistent WordPress experience for all users—the vision towards which we’re working (and hopefully compounding).

Conclusion

While small bets play a crucial role in reducing risk and iterating on products, they are not enough to drive the large-scale innovation required in today’s competitive landscape. It’s the alignment of these small bets with a grand vision, supported by the continuous delivery of incremental value to users, that leads to transformative success.

That’s it for today.

If someone forwarded this to you, you can subscribe to get your copy next week. If you enjoyed reading this, please share it.

Here on TTL, we dig into practical leadership tips and effective strategies, with a particular focus on tech leadership and managing distributed teams (that’s what I do every day, add me on LinkedIn).

Whether you’re steering a tech startup or leading a remote team, these insights are designed to help you navigate the complexities of modern leadership.

I also publish on paolo.blog and monochrome.blog

Cheers,