In the AI era, traditional business moats such as network effects and conversion costs are rapidly collapsing, and speed has become a new competitive advantage. This article explores how speed is the moat of the new era and its impact on the AI-driven business landscape.
Where is the moat in the AI era? This question has always been a widely discussed point in this wave of AI. Two years ago, I didn’t find the answer to this question, so after exploring AI entrepreneurship for half a year (you can read my 23-year summary article), I chose to be an advisor, contact more products and founders, and try to find the answer to this question.
So in the past two years, I have cooperated with nearly 30 companies with AI products going overseas, as their external consultants, helping to find PMFs and build growth channels, doing go-to-market, and also cooperating with independent developers to release more than a dozen small AI products.
After two years of front-line observation and actual combat, I am now ready to start an all-in business again, and at the same time found the answer to this question – Speed will become the only moat in the AI era.
I happened to see the latest article by Altimeter partner Ball that coincides with my views, as well as the latest trend report of Mary Meeker, who swiped the screen during the Dragon Boat Festival, all the trends prove one thing:Speed Is Everything。
So I have this article, combined with his article and my thinking, I hope to inspire everyone.
PS: Two years ago, when I summarized my experience in doing deep thinking circles, I also mentioned speed, so in the AI era, whether it is entrepreneurship or self-media, we pay attention to a “only fast and not breaking”.
Do you still believe in the “moat” of the product?
Network effects, conversion costs, proprietary data – these commercial defenses that were once regarded as guidelines are collapsing at an alarming rate in the AI era. Imagine a product feature that took 12 months to develop in 2022 and now only needs 3 prompts and a wrapper to replicate in 2024.
Those competitive barriers that once kept entrepreneurs awake at night and worked hard to operate are becoming papier-mâché walls.
I have observed the fierce competition in the AI space over the past two years, from the first CRM systems based on large language models being surrounded by swarming competitors, to the endless cloning of AI note-taking tools, and even vector databases and AI agent frameworks at the infrastructure level have lost their differentiation overnight.
This made me realize that we are at a critical juncture when the rules of business competition are being rewritten.
What does a product manager need to do?
In the process of a product from scratch, it is not easy to do a good job in the role of product manager, in addition to the well-known writing requirements, writing requirements, writing requirements, there are many things to do. The product manager is not what you think, but will only ask you for trouble, make a request:
View details >
The traditional “long-term moat” is becoming a myth, especially in today’s rapidly changing market environment. I became increasingly convinced that moats were never permanent fortifications, they were temporary barriers with a time limit.
At best, they are just a bridge – either to help the company get to the next defensible position or to watch its own defenses breach.
A true “long-term moat” is actually a superimposed combination of a series of small moats, each of which can buy you valuable time.
And how you use that time, how fast you execute, and how well you are evolutionarily adaptable determines whether you can always stay ahead of your competitors.
In the age of AI, this is more real and brutal than ever. In the past, the effective time window for moats was 6 to 12 months, but now it is only 2 to 3 weeks.
Models are rapidly evolving, infrastructure is constantly changing, and customer needs are redefining themselves in real time.
This speed of change makes traditional business strategies seem clunky and outdated.
How speed can become the moat of the new era
I found that in an AI-driven business environment, speed is no longer just one of the important factors, but the moat itself. The ability to build, release, learn, and adapt faster than everyone else is currently the only sustainable competitive advantage. In a world where open-source code is everywhere, product demos are readily available, and innovative ideas can be replicated with just a blog post, speed is the only element that can generate compound interest.
This speed is not a one-dimensional concept, but a comprehensive embodiment of multifaceted capabilities. The speed of execution determines how quickly you can turn ideas into reality, the speed of recruitment affects how quickly you can assemble a combative team, the speed of firing reflects your ability to correct wrong decisions, the speed of distribution is related to how quickly the product can reach the target users, and even the speed of decision-making becomes crucial. In this era of information explosion and fleeting opportunities, a slow decision-making process can mean missing out on an entire market window.
I especially want to emphasize that the speed at which a wrong idea is killed is just as important as the speed at which the right idea is executed. The opportunity cost of time has never been greater. When an idea or project proves to be in the wrong direction, the ability to quickly stop losses and readjust resource allocation is often more valuable than staying the course. Many companies die from stubborn persistence on failed projects rather than from a lack of good ideas.
In the cases I have observed, successful AI-era companies have shown amazing adaptation speed. Instead of panicking because a product feature is being copied by a competitor, they immediately start building the next advantage. This continuous innovation and iteration ability keeps them at the forefront of the competition. You don’t win because you have an impenetrable defense, but because you run faster than others.
If you are trying to build a moat today, I recommend thinking of it not as a castle wall, but as a race. Whoever accumulates the most advantages the fastest – product advantage, distribution advantage, talent advantage, infrastructure advantage – wins. This competitive model requires companies to maintain a high level of vigilance and action, and any form of complacency and slackness can be fatal.
The great changes in the SaaS valuation system in the AI era
When we talk about competition in the AI era, we cannot ignore the reaction of the capital market to this change. The valuation system of the SaaS industry is undergoing a profound restructuring, and this change directly reflects the shift in investors’ perception of the concept of moat. Judging from the latest quarterly report data, the median valuation multiple of the entire SaaS industry is 5.6 times, while the median valuation multiple of the top five companies is as high as 23.2 times, which shows the difference between the market and the different types of companies. Against the backdrop of a 10-year Treasury yield of 4.4%, these valuation multiples reflect investors’ mixed expectations for the future growth of the software industry.
What’s even more interesting is the strong correlation between growth rates and valuation multiples. When Ball delved deeper into the data grouped by growth rate, a striking valuation gradient was found: high-growth companies (more than 25% annual growth) had a median valuation multiple of 20.6x, medium-growth companies (15%-25% growth) had a median valuation multiple of 9.2x, and low-growth companies (less than 15% growth) had a mere 4.1x. This five-fold valuation difference clearly shows that the market now values the company’s growth rate more than traditional defensive metrics. This just confirms my point that in the age of AI, speed is everything.
Ball also pays special attention to the indicator of enterprise value divided by the next year’s revenue multiple (EV/NTM Rev/NTM Growth), which reveals which companies are overvalued or undervalued relative to their growth expectations. A company that trades at 20 times the next year’s revenue and is expected to grow by 100% is 0.2 times. The chart below provides a better understanding of how the market balances growth expectations with current valuations.
When it comes to free cash flow valuation, Ball observes an interesting phenomenon: the market only uses cash flow valuation methods for companies with free cash flow multiples between 0x and 100x. This filter creates a subset of which companies are free cash flow a relevant valuation metric. Companies with negative cash flow do not appear in this chart at all, indicating that the market’s valuation of these companies relies more on revenue growth expectations than cash generation capabilities.
Looking at the full picture of operational metrics, the performance of the entire industry is both worrying and promising. The median next year growth rate was 11%, compared to the previous year’s growth rate of 14%, and this downward trend reflects the intensity of market competition and macroeconomic pressures. But on the other hand, the median gross profit margin of the industry as a whole reached 76%, which is still impressive, showing the inherent advantages of the software business model.
What surprised me was the 45-month payback period for customer acquisition. In an era of AI where product lifecycles can only be a few months, this means that most companies’ customer acquisition strategies need to be significantly adjusted. Sales and marketing expenses account for 38% of revenue, R&D expenses account for 24%, and general overhead expenses account for 16%, indicating that even in these rapidly changing times, there is still a significant investment required to acquire and retain customers, and the cost of innovation is constantly rising.
Of particular interest is the median net income retention ratio of 108%, a metric that partly reflects the company’s ability to scale based on existing customers. In an era where products are easily replicated, the ability to consistently generate more revenue from existing customers becomes even more precious. This also explains why some companies are still able to maintain high valuation multiples in the face of fierce competition. At the same time, the median operating margin was negative 4%, but the free cash flow margin was positive 18%, which reflects the difference between accounting profit and cash generation capacity, and also shows that many software companies are actually generating positive cash flow despite book losses.
In the so-called “Rule of 40” metric, I see how software companies are finding a balance between growth and profitability. This metric adds up revenue growth to free cash flow margin, which should ideally exceed 40%. Companies that can maintain healthy cash flow while growing rapidly often achieve the highest valuations in the market. This further proves my point: in the age of AI, it is not only necessary to run fast, but also to run long.
Behind these data is a harsh reality: in the era of AI, the market’s criteria for judging companies are fundamentally changing. Traditional profitability metrics become secondary, while growth rate, cash generation and adaptability become key factors in determining a company’s value. Investors are willing to pay huge premiums for companies that have shown rapid growth and strong execution, even if they are not currently profitable in the accounting sense. This change in valuation method is the best validation of what I call the “speed moat” theory.
The law of survival in the new competitive environment
In this new competitive environment, I have summarized several crucial rules of survival. The first is to embrace a culture of rapid failure, where traditional business thinking encourages us to make perfect plans, conduct exhaustive market research, and build complete product relaunches. But in the age of AI, this methodology is outdated. The current success model is to quickly build a minimum viable product (MVP), bring it to market as quickly as possible, and quickly iterate and improve through real user feedback. I saw the first AI products to enter the market, although quickly surrounded by competitors who flocked to the market, some of them built a significant customer base and brand recognition by getting ahead of the competition.
I have observed that companies that have succeeded in AI have one common trait: they are not afraid of failure, let alone being seen failing. They openly test new features, quickly adjust product direction, and don’t hesitate to abandon unsuccessful projects. This transparent and agile approach allows them to find market demand and product match faster than their competitors. As mentioned in the original text, “the ability to build, release, learn and adapt faster than everyone else is the only sustainable advantage at the moment”.
The second is to redefine the team building strategy. In an environment that changes so quickly, a team’s ability to learn and adapt is more important than experience. I’ve found that the most successful AI companies tend to prioritize hiring talent who are strong learners and flexible thinkers, rather than simply looking for experts with deep experience in traditional technology stacks. Because today’s professional skills may become obsolete tomorrow, but the ability to learn and adapt is timeless. Speed of recruitment is just as important as speed of firing – quickly recruit the right talent and quickly weed out the wrong employees.
Distribution strategies also need to be completely rethought. Traditional software distribution relies on building sales teams, partner networks, and brand awareness, which take time to build. But in the age of AI, distribution speed has become crucial. An AI product that truly solves user pain points can spread throughout the industry in a matter of days, while a mediocre product struggles to achieve sustained growth even with a strong sales team. In a world where open-source code is ubiquitous and product demos are within reach, word-of-mouth and viral growth are often more effective than traditional marketing.
Decision-making speed has also become a key competitive factor. In an environment where the window of opportunity may be only 2-3 weeks, any form of delay in decision-making can be fatal. I especially want to emphasize that the speed at which a wrong idea is killed is just as important as the speed at which the right idea is executed. The opportunity cost of time has never been greater. When an idea or project proves to be in the wrong direction, the ability to quickly stop losses and readjust resource allocation is often more valuable than staying the course. Many companies die from stubborn persistence on failed projects rather than from a lack of good ideas.
Finally, I think the most important thing is to stay sensitive to technology trends. In the era of rapid AI development, a new model release, a new open source project, and a new API interface could be a game-changer. Successful companies need to establish sensitive information acquisition and analysis systems that can quickly identify and assess the impact of new technologies on their business and quickly adjust accordingly. As I mentioned earlier, infrastructure-level tools such as vector databases or AI agent frameworks can lose their differentiators overnight, and this speed of change requires enterprises to adapt and transform quickly.
In the cases I have observed, successful AI-era companies have shown amazing adaptation speed. Instead of panicking because a product feature is being copied by a competitor, they immediately start building the next advantage. This continuous innovation and iteration ability keeps them at the forefront of the competition. You don’t win because you have an impenetrable defense, but because you run faster than others. This is exactly the heart of what I call the new moat concept: don’t try to build castle walls, but focus on winning this ongoing race.
My prediction of the future competitive landscape
Based on my observations and analysis of current market dynamics, I have several predictions for the future competitive landscape. First, I believe there will be a new business model that I call the “continuous moat model”. Successful companies do not rely on a permanent competitive advantage, but rather build a mechanism for continuous innovation and rapid adaptation to keep themselves at the forefront of technological and market changes.
This model requires companies to have several core competencies: high technology sensitivity to quickly identify and adopt new technologies; Flexible organizational structure that can quickly adjust resource allocation and strategic direction; Strong execution ability, able to quickly turn ideas into products; and keen market insight to anticipate and guide changes in customer needs.
I predict that in the next two to three years, we will see a new group of super companies rise not by having a specific technical moat, but by the ability to build a moat on an ongoing basis. These companies will show incredible growth rates and adaptability, and their valuation multiples will far exceed those of traditional software companies, as investors pay a huge premium for this ability to continue innovating.
At the same time, I also predict that a large number of traditional software companies will be eliminated or forced to transform. Companies that rely on traditional moats and cannot quickly adapt to changes in the AI era will find their market share being eroded rapidly. This elimination process will be faster and more brutal than most people expect.
For investors, this means that the criteria for evaluating companies need to be completely changed. Traditional valuation models focus too much on historical financial data and static competitive advantage, while new valuation frameworks require more focus on company adaptability, speed of innovation, and team quality. Investors who can accurately identify and invest in the new generation of super companies will reap substantial returns.
Eventually, I believe that competition in the AI era will enter a new state of equilibrium, but this equilibrium state itself is dynamic. Successful companies will be those that stay ahead of the curve in the face of constant change, not those that try to build static defenses. Speed, adaptability and the ability to continuously innovate will be the core indicators of a company’s value, and the traditional moat concept will completely give way to a new competitive philosophy.
In this new business world, the only constant is the change itself. Companies and individuals who can embrace this change and find opportunities in it will be the ultimate winners. And those who still hold on to the old business mindset are destined to be abandoned by the times. We are at an exciting turning point, where old rules are being broken and new rules are being written
