Analyzing the Business Models of Independent Content Creators
Content creation may be the biggest lever available to ordinary people today—low barrier to entry, high ceiling, zero marginal cost of distribution. One person, one phone, one video can, in theory, reach a billion people.
But a lever is just a tool; direction decides the outcome. Looked at through the lens of business models, there are really only three ways to make a living from content creation: sell content directly, help others sell goods, or help yourself sell goods.
Monetizing Content Directly: The Purest, and the Most Fragile
The first is revenue sharing—the platform pays based on view counts, and the creator is essentially a content supplier to the platform.
| Platform | Revenue structure | Typical monthly income (1M views) | Notes |
|---|---|---|---|
| YouTube (long-form) | 55% ad revenue share | $800–3000 (about ¥6000–22000) | Depends on CPM (tech/finance is higher) |
| YouTube Shorts | Shorts ad pool | $100–400 | Clearly lower than long-form |
| Bilibili | Creator incentive + Charge | ¥1000–3000 | Incentives are low; needs brand deals on top |
| Douyin (domestic) | Creator program | ¥200–800 | Pure view-based income is very low |
| TikTok (overseas) | Creator Fund | $100–500 | Varies greatly by country |
The second is charging directly for content, following a logic similar to box-office revenue or paid knowledge: free content drives traffic, paid content monetizes, and the platform acts as an intermediary taking a cut.
| Platform | Model | Platform cut | Creator’s take-home share | Notes |
|---|---|---|---|---|
| YouTube Memberships | Channel membership | 30% | 70% | In-app purchases via App Store get another 15–30% cut by Apple |
| YouTube Super Chat | Tips | 30% | 70% | May likewise stack with Apple’s cut |
| Patreon | Subscriptions | 5%–12% | 88%–95% | Plus about 3% in payment processing fees |
| Substack | Subscription writing | 10% | 90% | Plus about 3% for Stripe |
| OnlyFans | Subscriptions | 20% | 80% | Fixed cut |
| Zhihu Yanxuan columns | Paid reading | 30%–50% | 50%–70% | Varies widely by contract |
| Xiaohongshu paid columns | Courses/content | 20%–40% | 60%–80% | Depends on contract tier |
| Douyin paid knowledge | Courses | 20%–50% | 50%–80% | Douyin mini-program cut is higher |
| Bilibili Charge program | Sponsorship | Platform takes nothing (or very little) | 90%+ | Mainly payment-channel fees |
| WeChat paid reading | Paid articles | About 30% | 70% | Includes payment processing fees |
Both of these models are pure enough—the creator produces content, the audience consumes it, and there’s no third-party product in between. But the price of purity is fragility: your income depends entirely on the platform’s distribution rules, and those rules can change at any time.
Advertising for Others: Renting Out and Burning Through Trust
A KOL’s essential role is much like that of a legislator—they make a living on influence. Content is the means; trust is the asset. The core act of monetization is just one thing: renting out the network of trust you’ve accumulated to brands.
Bloggers who dig deep into a vertical—phone reviews, car reviews—monetize by working sponsored placements into their videos. It’s a tightrope walk: you have to spend your followers’ trust to make money, but you can’t overspend it to the point where your reputation collapses. Building trust takes a year; overdrawing it takes a single video.
There’s also the big-streamer livestream-selling model. The streamer goes in thinking “make as much as I can,” the audience thinks “buying here is the same as buying anywhere,” and both sides get what they want—seemingly reasonable. But the problem is this: livestream sales involve hundreds or thousands of SKUs, the streamer can’t possibly vet the quality of each one, yet they’re staking their own credibility to vouch for every single product. When you frequently guarantee things you can’t control, a blowup isn’t an accident—it’s inevitable.
Advertising for Yourself: The Model with the Strongest Structural Advantage
Making advertising so good that the audience willingly subscribes to watch it is a remarkable thing.
For this kind of creator, the content itself is part of the brand’s marketing. The content is controllable, the product is controllable, and the brand moat can keep compounding—of the three models, this one has the best structural advantage.
The most typical form is the entrepreneur as a personal brand (IP). Musk to Tesla, Lei Jun to Xiaomi—the entrepreneur’s personal narrative carries over into the brand image, sparking not just the desire to buy but three deeper layers of psychology:
Mimetic desire: The person I admire uses this product, so I should too—exactly the same underlying logic as luxury goods.
Participatory desire: I share their vision and want to take part in it. Buying a Tesla to support Musk’s Mars program, buying a Huawei to support Ren Zhengfei’s fight against sanctions—paying money is the most convenient way to acquire a sense of participation. This desire carries a strong religious flavor.
Belonging desire: People naturally and spontaneously form clusters. When a brand is tightly bound to a particular circle, buying becomes a pledge of allegiance to an identity—buy Xiaomi and become a Mi fan, buy Huawei and become a patriot, buy Tesla and become tech nouveau riche, buy Hermès and become a wealthy socialite.
On top of that, anchoring a brand to a specific person naturally strengthens trust—if the product goes wrong, it’s me who goes wrong. Fans may even gain scarce opportunities to interact with the entrepreneur through their purchases, which is fundamentally the same as a fangirl buying merch for her idol.
There’s also a more elegant variant: turning the production process of the product itself into content.
A craft blogger films the making process into a beautiful video; the audience is drawn in by the content and places orders; and the process of making products for those customers becomes new content material—content drives consumption, consumption feeds back into content, and the flywheel spins on its own.
Online consulting works the same way. The consulting process, anonymized, is turned into content; viewers watch it, develop a need, and come for a consultation; the consultation becomes new material that attracts more viewers.
The beauty of this model is: you don’t have to deliberately hunt for topics—the consumer’s needs are the topics; you don’t have to deliberately hunt for clients—interested people are drawn in by the content. Content and business fuel each other, forming a genuine virtuous cycle.
