When 90% of AI companies in Silicon Valley are “cutting leeks”, the real dividends actually come from the AI entrepreneurs themselves. The author of this article uses front-line data to break the illusion: 300% of ToB AI’s revenue has skyrocketed, and 90% of its customers are other AI startups; ToC is difficult due to the lack of incremental users. The article gives the ToB growth template of “naming hegemony + high-profile publicity + portfolio fission”, as well as the ToC counterattack routine of “doing growth first and then making products”, bluntly saying that “now entrepreneurship is hard to brain, not resources”.
While engineers in the South Bay are still hard at work debugging models, entrepreneurs get new AI financing every 10 minutes at the Capital One café in downtown San Francisco.
Silicon Valley’s South Bay and North Bay are completely two worlds: engineers in South Bay discuss promotions and stock speculation every day, and entrepreneurs in San Francisco are looking for customers to make money every day. I am also a member of the vast entrepreneurs, and after chatting with them, I got a lot of first-hand information and shared it with my readers.
The surge in AI startups is the biggest dividend of AI entrepreneurship at present.
This year, the revenue of many AI startups has risen a lot, and the B-end that has done well has an average increase of 300% in this year’s revenue in my own statistics. Most people will think that it is a breakthrough in AI technology that has unlocked many new scenarios. After I took a closer look, I found that this was not the case. A closer look at their customer data found that 90% of the incremental customers are other AI startups.
After 10 years of interaction design, why did I transfer to product manager?
After the real job transfer, I found that many jobs were still beyond my imagination. The work of a product manager is indeed more complicated. Theoretically, the work of a product manager includes all aspects of the product, from market research, user research, data analysis…
View details >
If you think about it carefully, it is also reasonable, the core of any industry needs a large enough variable. Nowadays, the user base is unchanged, and the only big variable is that the number of AI entrepreneurs has exploded in the past year or two.
As shown in the figure above, YC statistics show that 2 or 3 years ago, less than 50% of startups were AI companies. Almost 90%+ of startups today are AI startups.
AI entrepreneurs prefer to use other AI products and are willing to pay for them.
Obviously, the crowd characteristics of AI entrepreneurs are: extremely optimistic about the future of AI and willing to replace all traditional tools with AI.
That’s the perfect AI customer crowd.
They are extremely concerned about any newly released AI products and are very willing to pay for AI products.
With the explosion of the number of AI entrepreneurs, the income of these tob AI companies will inevitably rise. AI ToC entrepreneurship is much more difficult. Because for the broader ToC population, AI entrepreneurs account for less than 1%, with almost no increase. This also explains why AI entrepreneurship is simpler than ToC entrepreneurship.
Let’s talk about the template-style growth routines of these AI companies.
Let’s talk about toB first.
As mentioned earlier, the biggest customers are AI entrepreneurs, so to get their attention, you must do what they like. Let’s talk about the formula first: “Preemptive positioning + high-profile publicity”
Now AI is still a blue ocean, just like the age of sailing, whoever seizes the territory first will belong to whoever owns it. The world’s first AI general-purpose agent, the world’s first AI programmer, the world’s first AI designer, the world’s first AI lawyer, and the world’s first AI XXX. As long as you dare to grab it, this territory is yours. It must be fast, and futures don’t matter. Even if the latecomers do better, they are not valuable. OpenAI is the best example, and the recent Manus is not bad. I believe that at the product level, other products abound that are better, but VW will only focus on the first.
Therefore, in terms of positioning, it is best to subdivide vertically enough, don’t grab the territory that others have already grabbed, and if the speed is not fast enough, give up. The highest priority is to get your special vertical first.
In the past, the content of product promotions did not have much traffic. But you’ll find that now on X/Twitter, producthub, and even LinkedIn, your AI product launch videos suddenly get a lot of organic traffic.
The core reason is still the characteristics of your customers/users (AI entrepreneurs) as mentioned earlier.
They like to scroll through X/Twitter, follow and try out new AI products, so there is an increase in this part.
So the template is: naming hegemony: (the world’s first XXX) + high-profile announcement video (find a better video studio) + founder real appearance (speak English, AI lip-sync)
As long as the demo is fancy enough, almost all of them can have good organic traffic. It doesn’t matter if the product is not done, make a waiting list, anyway, it’s all futures. Those who have money can also find some self-media bloggers to forward and heat, which completes the first step of cold start. With the initial user, the next big money can be raised, and then through the portfolio company of the investment institution, you can continue to expand high-quality customers and income, and the whole will be operational.
This is simple to say, but in fact it adds leverage to the company and the founder, if the product is a pseudo-demand, or far from the demo expectation, it is also very fast.
Another core point here is that if you can become a part of the workflow in an AI startup, then your value will be limitless. For example, Cursor is the world’s fastest-growing AI company, and the core is that almost every startup uses it to write code.
Cursor went from $100M ARR to $300M ARR in 4 months
X
Let’s talk about ToC.
ToC is much more difficult because there is no increment at the demand level. The only variable is on the supply side, where the cost is significantly reduced: two days is enough to copy a competitor with Cursor.
The rate of technology depreciation > the speed at which financing arrives
The three elements of previous entrepreneurship: product, technology, and growth. The cost of the first two is almost reduced to zero. Product, just reproduce. technology, can use cursor to almost 70%, the early financing is enough, and then recruit people when there is money.
I haven’t touched code for a few years now, and I want to say that I will never write code again in my life. I didn’t expect the appearance of Cursor to save me, a “lazy person”.
The first time I used it, I developed a complete client app in one day, and I aha.
The second time, it took two weeks to develop a complex app involving the front and back ends. Now I am very proficient in using cursor, and some companies have even invited me to teach their employees how to use it. Production efficiency is not a matter of improvement, it is a huge difference between having or not having it. In the past, in China, you had to finance at least 50-1 million to barely make a demo, but now, you can do it alone.
Therefore, growth becomes the only thing that matters.
We have a lot of experience, and have helped 8 apps rush to the top 5 of the US list, and companies in need can consult me.
This topic is also relatively big, and I will have the opportunity to talk about it in another article in the future. Today, let’s talk about the most important point:
Growth first, then product.
The order determines the priority and also determines the direction of your thinking. There is no shortage of good products in this era, what is lacking is how you let users find your products. Let me give you an example and you will understand.
Some time ago, there was a very popular product called Rizz, which mainly helped users pick up girls. When I paid attention to it, it used about 15 TikTok accounts to do 550 million traffic, with an average of 450,000 views per organic streaming video.
After some research, he found that this is how it has grown: first of all, he observed that texting stories on TikTok were very popular for a while, and this type of theme had hundreds of thousands of views as long as it was posted. This group of creators was unable to monetize at that time, and almost all of them were playing with a group of children. He thought, can these contents be productized and monetized? That’s when I thought of Rizz.
So I made a batch of content myself, and also mobilized this group of children to help make content (the cost is extremely low) and quickly achieved an amazing result of one million downloads, I estimate that it may have cost thousands of dollars, and there are 300,000 dollars a month.
This is a typical example of growth first and then product. You have to think of a popular theme and content by yourself, which is too difficult, and it depends on the face of the platform, whether you are willing to help you push it. But if you can identify a content form that has been verified and can explode, and it just so happens that this theme is not commercialized, it is an excellent business opportunity.
There are a lot of good things that are out there, and you just have the ability to find them.
So you can look at your company/product and see if you have really borrowed the above-mentioned AI dividends. Many things that are said in the news media are actually not needed by you, that is, they have become traffic from others.
The good news is that the threshold for starting a business has become lower, and you don’t need to raise too much money to start a company first.
The era of not relying on resources and purely working hard has begun.
Share.
