CEO Yang on Business Models: Not Just Making Money, but Building a Business That Keeps Growing

Over the years, I have met many entrepreneurs and worked with many companies.

When people talk to me about their businesses, I often hear: “Chairman Yang, my product is excellent.” “Our technology is impressive.” “This market is huge.” “If I had the funding, I could make it work.”

But I usually do not start by asking about the product. The first thing I want to know is: What is your business model?

I have always believed that a good product does not necessarily make a good business. Revenue alone does not mean you have a business model. A truly strong company must do more than sell something and collect money today. It must answer a more important question: Why does this business work today, why will it still work five years from now, and why does it become more competitive as it grows? That is what a business model is.

1. What Is a Business Model?

Textbooks usually define a business model as how a company creates, delivers and captures value. That definition is correct. But in an entrepreneur’s language, I would translate it into four things:

Value × Money × Scale × Moat

A viable business model must pass all four tests. Without any one of them, it may be no more than a business that looks promising.

2. Value: Someone Wants It

The first question is not what you want to sell, but why the market needs you. Many entrepreneurs spend a great deal of time describing their technology, products, features and specifications. They forget that these are not what customers ultimately care about.

Customers want to know: What problem do you solve for me? Do you save me time, reduce costs, increase revenue or lower risk? Or do you help me do something I could not do before?

That is why I often say: Start with what the market is missing, rather than what you have. Technology itself is not value. A product itself is not value. Solving a problem is value.

3. Money: You Can Make a Profit

The second question is more practical. Demand does not guarantee profitability. Many products are popular, yet their companies lose money. Some companies generate enormous revenue and still collapse. Why? Because revenue is not profit. Revenue is not a business model; the structure of profitability is.

When I examine a business model, I keep asking: Who pays? Why do they pay? How often, and how much? What does it cost to acquire a customer? What does it cost to serve that customer? How long will the customer stay? How much value can one customer generate for the company over their lifetime?

These questions reveal the real unit economics: Does each sale or customer relationship make money? If every additional customer creates an even larger loss, that is not growth. It is scaling your losses.

4. Scale: It Can Grow

This is one of the things I take most seriously. Some businesses can generate 10 million a year but begin to struggle at 100 million, let alone 1 billion or 10 billion. So I ask: Can the method that works at 10 million be replicated at 100 million? Can the method that works at 100 million then be replicated at 1 billion?

If a tenfold increase in revenue requires ten times the staff and ten times the cost, while management complexity increases twentyfold, the business model has limited scalability.

A truly strong model should lower unit costs as scale increases, improve system efficiency as the customer base grows, and gradually allow revenue to grow faster than costs. That is what I mean by scale.

Entrepreneurship involves more than selling something. The real challenge is turning a successful approach into a repeatable system.

5. Moat: You Can Defend It

Even when the first three conditions hold, I still ask one final question: If this business is so profitable, why do others not enter? That is the question of a moat.

Your moat may come from technology, patents, brand, data, supply chains, distribution, economies of scale, network effects, customer relationships, switching costs, industry know-how, or even regulatory and capital barriers.

But let me stress this: “We started earlier” is usually not a moat. If you can do it, others may well be able to do it too. A real moat means the more successful you become, the harder you are to catch.

This distinction matters. If growing from 100 customers to 10,000 merely increases revenue, that is scale. But if reaching 10,000 customers also gives you more data, better models, lower costs, a more complete supply chain, a stronger brand and a larger ecosystem—so that even well-funded newcomers struggle to catch up—that is a moat.

6. The Best Models Eventually Create a Flywheel

When I evaluate companies today, I look one layer deeper: the flywheel. A good business model does more than keep moving forward. After a while, the system itself begins to drive progress.

For example: More customers → more data → a better product → higher customer retention → greater scale → lower unit costs → more competitive pricing and service → more customers → more data.

At that point, the company has created a reinforcing cycle. Ordinary companies depend on people to drive growth. Excellent companies depend on systems. Great companies let their business models propel the next round of growth. That is the flywheel.

7. The Nine Questions I Ask About a Business

Whether it is a coffee shop, an AI company, a SaaS platform, a computing center or even a capital platform, I return to these nine questions:

1. Who are you selling to? Who uses it, who makes the decision, and who pays?
2. Why do they buy? What pain point do you actually solve?
3. What do you sell? A product, a service, a solution or an outcome?
4. How do you deliver it? Through company-owned operations, franchises, a platform, SaaS, APIs, distributors or an ecosystem?
5. How do you charge? One-time payments, subscriptions, commissions, licensing, usage-based pricing or success fees?
6. Does each unit make money? What are the gross margin, customer acquisition cost (CAC), lifetime value (LTV) and payback period?
7. Can it be replicated? Does the same logic hold from 1 to 10, from 10 to 100, and from 100 to 10,000?
8. Why can others not copy it? Where are the real barriers to competition?
9. Does it get stronger as it grows? Has it developed its own flywheel?

If these nine questions can be answered clearly, I can usually judge whether I am looking at a product, a business, or a company with a real chance of growing large.

8. AI Will Reshape Business Models

I often say that during the Industrial Revolution, we measured the power of machines in horsepower. In the AI era, we talk about computing power. But what businesses should truly pursue goes beyond computing power: it is intelligence.

AI is like electricity. Having electricity does not automatically give you air conditioners, refrigerators, televisions or factories. Likewise, having AI does not automatically give you a business model.

The real competition will be over who can turn AI into products, products into processes, processes into systems, systems into business models, and business models into organizational intelligence that can be replicated again and again.

When I evaluate an AI company in the future, I will not stop at asking which model it uses, how many GPUs it has, or how capable its agents are. I care more about how much value AI creates for its customers. Can the company charge for that value? Can it scale after it begins charging? And once it scales, does it build a moat of data, processes, knowledge and an ecosystem?

That is a real business model for the AI era.

9. CEO Yang’s Business Model Formula

Finally, I distill my view into four ideas: demand, profit, scale and defensibility.

• Value: Someone wants it.
• Money: You can make a profit.
• Scale: It can grow.
• Moat: You can defend it.

One level higher: Value × Money × Scale × Moat → Flywheel.

When these four elements begin to reinforce one another, a company goes beyond simply doing business. It starts building a business system that can grow, be replicated, accumulate strengths, and become more valuable over time.

My final definition is this: A product answers, “What do you sell?” Marketing answers, “How do you get people to buy?” Operations answers, “How do you do the work well?” A business model answers, “Why does this business work? Why is it profitable? Why can it scale? And why do you ultimately win?”

That is how I understand a business model.

— CEO Yang