Compounding Advantages: How to Design Businesses That Improve Over Time

Most executives think about growth linearly. They focus on metrics like MRR or customer acquisition, plotting steady upward trajectories on their pitch decks or business cases.

But the businesses that become really valuable aren't just growing—they're compounding.

The more I read, the more this topic is emphasised:

• Hamilton Helmer refers to this as "scale economies" in the 7 powers book.
• Paul Graham calls it "superlinear returns" in his essay on the subject
• Lesson 3 from Sam Altman looks for exponential growth that allows for strong compounding.
• Reed Hastings in the book That Will Never Work actively seeks these advantages out before they start.

The difference is crucial. Linear growth means adding. Compounding means multiplying.

• A company adding $1M in revenue each year grows linearly.
• A company whose core systems make each dollar of revenue progressively easier to acquire is compounding.

Consider two businesses. Business A builds custom software for enterprise clients. They're good at what they do and grow by hiring more developers and salespeople. Each new client requires roughly the same effort as the last.

Business B builds a collaborative design tool. Every new user makes the product more valuable for existing users. Each project uploaded becomes a template others can learn from. Their cost of customer acquisition drops as their product becomes more useful.

After five years, Business A might be bigger. But Business B is more valuable, because they've built compounding advantages into their core business model.

Amazon's review system

Not all compounding systems directly increase revenue to begin with. In fact they can seem insignificant at first.

When Amazon started collecting customer reviews, many thought it was just a nice feature. But those reviews became a moat that grew deeper with every purchase.

Each review made Amazon slightly more useful than competitors, bringing more customers, who left more reviews. Twenty years later, this "nice feature" is an almost insurmountable advantage.

Here's what's counterintuitive about compounding advantages: they usually slow you down at first. Building systems that compound requires extra work upfront. It's like the difference between taking a direct path up a mountain versus building a ski lift. The direct path is faster initially, but the ski lift, once built, changes the game entirely.

Types of compounding advantages

There are three main types of compounding advantages businesses can build:

1. Network Effects: The classic example. Each new user makes your product more valuable for all users. But modern network effects are more subtle than just "more users = better." The best businesses create micro-network effects within specific features or user segments.

2. Learning Systems: Your product gets smarter over time. This isn't just about collecting data—it's about building systems that automatically improve based on usage. Chrome's address bar predictions get better the more you use them. Every keystroke makes the system slightly more valuable.

3. Cultural Compounds: The hardest to build but most valuable when they work. These are processes where your company's culture becomes self-reinforcing. Stripe's dedication to API documentation attracts developers who care about documentation, who then make the documentation even better.

Combine them for maximum impact

The best businesses usually combine multiple types. Figma, for instance, has network effects (more users = more shared components), learning systems (their tools get smarter based on usage patterns), and cultural compounds (their community creates educational content that attracts more users).

But here's the important thing: you can't bolt on compounding advantages later. They have to be built into your business's core architecture. This is why many successful founders seem to build their businesses "backwards"—focusing first on systems that will compound rather than features that will drive immediate growth.

The stop tomorrow test

A good test for whether you're building real compounding advantages is to ask: "If we stopped all active development tomorrow, would our product still get better for users?" For most businesses, the answer is no. For the best ones, it's yes.

This applies beyond product development.

• Hiring: Each great hire makes it easier to attract more great hires
• Culture: Strong cultures become self-reinforcing over time
• Brand: The best brands compound in value as they become cultural shorthand
• Knowledge: Internal tools and processes that make the company smarter over time

The hardest part about building compounding advantages is that they're often invisible early on. For more on this you can read our post on the Innovators Dilemma.

When PayPal was building their fraud detection system, it looked like a cost centre. Only later did it become clear that this learning system was one of their most valuable assets.

This invisibility creates an opportunity. Most competitors will focus on visible metrics that show immediate returns. If you're willing to play a longer game, you can build advantages they won't even recognise until it's too late to catch up.

Compound interest or current income?

Here's a practical way to think about this: For every major decision in your business, ask

"Does this create compound interest or just current income?"

Many good features create current income—they solve an immediate problem. But the best features create compound interest—they get more valuable the longer they exist.

Look at Xero as an example. Going from raising a $15 million IPO to now a $24 billion market cap. This is what happens when you get compounding right. We need more of these businesses in New Zealand.

Remember: The goal isn't to build a business that's valuable now. It's to build one that becomes increasingly valuable over time, even when you're not pushing as hard. That's the real secret of the most successful companies—they built machines that improve themselves.