Author: Frank, PANews

For some time now, the hottest topic in the tech and startup circles hasn't been a major company releasing a new model, but rather the nationwide phenomenon of "raising lobsters."

On the one hand, the "lobster farming" craze has driven business growth in related industries, with large-scale model companies and cloud server providers making a fortune. On the other hand, how much actual benefit Openclaw can bring to users remains a mystery. Although social media is full of such myths, a closer look reveals that most are fictional stories created to attract traffic.

Can you really make money raising lobsters? If so, who actually profits from it?

PANews compiled data from the TrustMRR data platform, publicly available case studies on social media, the project's official website, and reports verified from multiple sources. To distinguish between "verified real income" and "self-proclaimed myths online," a large number of rumors based solely on one-sided claims or lacking any evidence were eliminated.

TrustMRR, a data platform for overseas startups, shows on its OpenClaw category page that there are 153 recorded projects in this ecosystem, with a total revenue of approximately $358,600 in the past 30 days. Further analysis of the top 30 projects reveals that their combined revenue accounts for 97.3% of the total. If we dissect these projects and their underlying profit-making logic layer by layer along the "industry value chain," we discover a sobering truth: those who made money first were not the people who produced lobster products, but rather those who helped others raise lobsters and taught them how to do so.

But this is clearly not the real answer we're looking for: how exactly do those who actually use OpenClaw make money? PANews has compiled five money-making strategies for OpenClaw.

The first type: Selling "shovels" and providing agency services: Quick money by exploiting "cognitive gaps"

Whether overseas or domestically, the products that OpenClaw discusses the most and generate the most revenue are often not specific applications, but rather packaging tools and one-click hosting services.

OpenClaw is more like an infrastructure layer than a ready-to-use consumer product. This creates a high barrier to entry for non-technical users. Once complexity exists, services will inevitably emerge.

Of the approximately $350,000 in front-end revenue for TrustMRR over 30 days, “managed deployment” and “one-click cloud hosting” projects alone contributed approximately $120,100, accounting for 34.5% of the sample revenue.

A typical example is QuickClaw, which packages its underlying capabilities into a mobile app, priced at $3.99 per week or $49.99 per year, generating approximately $8,782 in revenue over the past 30 days.

In the Chinese internet, this logic is even more simplistically applied: "replica lobsters" on Xianyu (a second-hand marketplace).

According to media reports, the "OpenClaw deployment service" has seen explosive growth on platforms like Xianyu and Xiaohongshu recently. Remote installation typically costs between 100 and 300 yuan, while local on-site installation ranges from 400 to 1000 yuan. During a certain period, the daily transaction volume of related products and services increased by 150% compared to the previous period.

The essence of this logic is "profiting from information and cognitive gaps." Users are willing to pay to save 30 minutes of hassle, but this is a "window of opportunity" business. As official one-click deployment tools mature, the benefits of pure installation services will quickly fade away.

The second layer: Packaging the AI expert persona: When the "story" itself becomes the most expensive product.

If we take it a step further, another, even more valuable layer emerges in the OpenClaw ecosystem: it's not about deploying for you, but about training your agent. In the top 30 TrustMRR samples, templates, skill packs, and configuration-related projects contributed 26.4% of the revenue.

One of the most credible and well-documented business cases in this layer: FelixCraft.

In early 2026, overseas creator Nat Eliason launched an experiment. He named his OpenClaw bot "Felix," invested $1,000 in seed funding, and let it build its own business. Within a week, Felix generated approximately $3,500 in revenue through Stripe. In addition, the crypto community issued related MEME tokens for this agent on-chain and forwarded 60% of its daily transaction fees to it, enabling it to earn up to $100,000 worth of crypto tokens within a week.

As the most noteworthy case study in the research, Felix's case has several distinctive features. First, Nat Eliason granted the AI sufficiently high permissions, enabling the agent to autonomously post tweets, retweet comments, and interact with the community. Furthermore, before the product launch, Nat Eliason stated that he had invested significant time in building the agent's framework, including memory modules, security settings, workflow design, and more.

Nat Eliason himself admitted in a podcast interview that achieving profitability was somewhat unexpected. Essentially, Felix's main revenue source comes from packaging and selling products based on his training process and results. The revenue from MEME coins, on the other hand, is more due to the buzz and traffic generated by this story.

It's worth noting that Claw Mart (an Agent Skill marketplace), the top-earning project in TrustMRR's OpenClaw category, was created by the agent Felix, and its cumulative earnings have reached $71,300. Another key reason for this high earnings is that Felix's story as an agent, capable of autonomously creating projects and automating tasks, serves as the strongest endorsement for the product.

Felix's success reveals a high-level path to OpenClaw's commercialization: giving agents a lasting identity. When OpenClaw is packaged with a specific name (Felix), a marketable guide, a set of reusable skill sets, and a perfect narrative of "starting your own business with AI," it becomes a new type of personal brand with explosive dissemination potential. However, the core barrier to this approach doesn't actually lie in AI, but rather in Nat Eliason's powerful agent training capabilities and brand marketing strategies.

The third layer: Selling the myth of efficiency: Using AI to do the work and monetizing through "storytelling".

Of all the ways to make money, the one with the highest approval rating is probably: use OpenClaw to replace manual labor, and the money saved becomes the profit.

In the content creation industry, this has already become a reality. Developer Oliver Henry named his agent "Larry," who is fully responsible for the TikTok account. Larry automatically calls upon large models to generate images, write titles, and upload drafts. Henry only needs to spend 60 seconds a day selecting background music and clicking publish.

Oliver Henry stated that Agent Larry helped his videos surpass 500,000 views within five days, generating $588 in revenue (this refers to the paid revenue from the two apps he recommended in his videos). Additionally, Larry generated $4,000 in revenue by issuing MEME tokens. Interestingly, Oliver Henry's tweet telling this story has already reached 7.1 million views, similar to Felix's story mentioned above—the story itself seems to have more commercial value than the agent.

In China, Cheetah Mobile founder Fu Sheng assembled a team called "Sanwan" with eight agents, increasing the frequency of their WeChat official account updates from a dozen articles a year to daily updates. They also achieved a record-breaking 1 million+ reads on their account, attracting widespread attention. The post that garnered over a million reads still tells the story of how agents work.

In other words, whether the content produced by Agent itself can become a viral hit remains to be seen. The viral hits that have already occurred are mostly stories about Agents making money or improving work efficiency. Telling the story of "crayfish" is currently the biggest topic of discussion in the content creation field.

Fourth layer: Deep industry customization: Break free from the tool-centric competition and earn "service premiums".

If the profit from OEM packaging is based on the "entry barrier," then taking it a step further and packaging "crayfish" as a product for personalized needs is another matter entirely.

RoofClaw is a typical example of this type of project. TrustMRR shows that it earned approximately $49,800 in the past 30 days, with cumulative revenue reaching $1.8 million. It positions itself as "personalized customization and delivery of MacBook Airs equipped with the Openclaw system." In other words, its business strategy isn't simply to pre-install a "crayfish" (a virtual extension for RoofClaw), but rather to encapsulate this "crayfish" within a MacBook, along with customized services to mold this "crayfish" into an agent tailored to your specific needs.

This type of service may be the real business opportunity for "crayfish" in the future. Users may not actually need a "usable" "crayfish" app, but rather a "crayfish" app that has been fully customized to their specific needs. Behind this demand is the sale of in-depth services tailored to the agent.

To put it bluntly, we can foresee that a large number of companies will rely on agents in the future, but how to train or "manage" these agents will become an unavoidable necessity.

Fifth Layer: On-Chain Transaction Legends: The Most Tempting Poisoned Apple and Traffic Bait

On social media, the most effective way to spread the word about OpenClaw is always through the myth of getting rich overnight.

Currently, the few accounts verifiable through on-chain data are accounts like 0x8dxd on the prediction market Polymarket, which is a high-frequency trading bot. Numerous posts on social media speculate that it relies on OpenClaw for its high-frequency trading, but PANews' analysis shows that the actual controller behind this high-frequency address has never made any such claims. These stories of "OpenClaw helping me design an automated trading program and earning $100,000 a month" are mostly advertorials designed to attract users to adopt their copy trading bots.

The reason for listing this case is to serve as a reminder that, consistent with the findings of previous PANews research, Agents and high-frequency trading bots are not the same thing at all, and people are always misled and fantasized about them due to their mystique.

Final thought: The person who teaches you how to make money is the real winner who never loses.

After analyzing the entire ecosystem, we discovered a phenomenon more thought-provoking than any single case: sharing "how much money I made with OpenClaw" on social media is itself a very stable business.

When a post claiming "I earn 50,000 yuan a month using OpenClaw" goes viral, traffic becomes the bait. The author then smoothly directs viewers to paid communities, consulting firms, or paid product links. "Boasting about income" is the top of the customer acquisition funnel, and the "money-making myth" is the most powerful marketing material. This forms a perfect self-fulfilling cycle: selling money-making case studies—gaining traffic—monetizing traffic—then sharing money-making secrets as a mentor—leveraging even greater resources.

Essentially, it has indeed spawned a new business chain: the bottom layer is contract manufacturing and infrastructure, the middle layer is skill packages and workflow replacements, and the top layer is industry solutions and consulting services. If you understand business, marketing, and have traffic, OpenClaw can drastically reduce your costs and increase your productivity.

Many people are sharing how OpenClaw has optimized their workflows and enabled many convenient functions, but it's far from a get-rich-quick scheme. The "herd effect" it triggers is the core of this traffic story. When you desperately push your way through the crowd to the front, you find there's nothing there, and you're the one being waited for.

(P.S.: The term "crayfish" was not used in this article.)

The first type: Selling "shovels" and providing agency services: Quick money by exploiting "cognitive gaps"

The second layer: Packaging the AI ​​expert persona: When the "story" itself becomes the most expensive product.

The third layer: Selling the myth of efficiency: Using AI to do the work and monetizing through "storytelling".

Fourth layer: Deep industry customization: Break free from the tool-centric competition and earn "service premiums".

Fifth Layer: On-Chain Transaction Legends: The Most Tempting Poisoned Apple and Traffic Bait

The second layer: Packaging the AI expert persona: When the "story" itself becomes the most expensive product.

US stocks opened lower, with all three major indexes falling.

Popular Articles