The IPO file of Anthropic reveals its strategy for retaining corporate customers
TechCrunch
58m ago
Ai Focus
As disclosed in the S-1 document, customers who have spent at least $100,000 over more than half a year also use both the Claude platform and first-party applications such as Claude Code or Cowork. The company states that this integrated product strategy is driving adoption among corporate customers and helping to increase customer retention and average spending per customer.
Helpful
No.Help

Anthropic is building a similar cross-product ecosystem, a model that has helped tech companies such as Amazon and Uber develop strong user loyalty. Its S-1 documents show that this strategy is indeed effective. Today, more than half of the customers who spend at least $100,000 per year also use the Claude platform as well as first-party applications like Claude Code or Cowork; whereas at the end of last year, this proportion was around 30%.

"The speed and depth with which our corporate clients have adopted Claude is strong evidence that Claude is at the forefront of AI," the company stated in a S-1 document submitted to the U.S. Securities and Exchange Commission on Monday."

As of September 30th, Anthropic has recorded over 11,900 customers with an annual consumption of at least $100,000; over 1,900 customers with an annual consumption of over $1 million; and approximately 250 customers with an annual consumption of $10 million or more.

Anthropic attributes the customer growth momentum to "significant usage expansion, as customers gradually recognized the value of deploying Claude throughout the organization," which drove its net revenue retention rate to around 900% by the end of the third quarter.

In the document, the company describes Claude as not a "static product for a single service use case," but rather as a cutting-edge AI with the ability to evolve rapidly. It covers models, developer platforms, as well as the application layer that users directly interact with, which is essentially equivalent to "dousing cold water" on the entire traditional SaaS industry.

The document states: "Progress at any one level will have a synergistic effect among the three levels, accelerating our ability to deliver greater intelligence." The company views this "integrated approach" as the reason why more customers are using both the Claude platform and first-party applications simultaneously.

This tactic is not unfamiliar. Uber initially sold taxi services, later expanding to food delivery and freight shipping. Amazon, on the other hand, first attracted customers with cloud computing services, then extended its offerings to storage, databases, and other products, making it increasingly difficult for users to leave their ecosystem. Similar to these companies, Anthropic does not have to sell each product to completely new buyers: a company that starts with API can deploy Claude Code to its engineers and then promote Cowork to a wider range of employees.

The value of these application users is much higher. The documents show that in the third quarter, among the Claude Enterprise customers using Claude Code, the top 5% of customers with the highest spending had a monthly average expenditure about 20 times that of ordinary enterprise users who did not use this coding tool. This finding indicates that there is still significant room for increasing spending per user, and it also explains why Anthropic is eager to introduce its application products to those customer accounts that initially switched from API.

Anthropic indicates that the company amplifies this natural diffusion by investing in the Applied AI team. The internal Applied AI team collaborates with startups and enterprises to help integrate Claude into customers' existing technology stacks; recently, the company also launched an independent collaboration project called “Ode with Anthropic” with Blackstone and other private equity firms, which involves deploying engineers in advance to create customized AI tools using Claude.

Anthropic also relies on third-party distribution partnerships with ultra-large cloud service providers AWS, Google Cloud, and Microsoft Azure. According to the "joint market promotion method" mentioned in the document, Anthropic maintains direct contractual relationships with customers and pays fees to each cloud service provider. The company states that this allows them to "avoid channel conflicts" and "set prices" on their own.

As this AI giant seeks to continuously "reshape our business" by launching "new product categories that are intended to far exceed the economic value of existing products," the Anthropic strategy, which has begun to take shape, may become crucial. The company states that, compared to typical enterprises, it will devote less effort to "consolidating existing products and business models" and instead focus more on defining new products.

Despite impressive revenue growth in the first half of the year, reaching $16.3 billion, a significant increase from $1.2 billion in the same period last year, the document also points out unfavorable factors. Cloud partners that help sell Claude are increasingly viewing Anthropic as a competitor, which may lead to difficulties for Anthropic in obtaining the necessary computing power. The document states that Amazon, Google, and Microsoft may prioritize meeting their own needs or giving preference to their invested competitors.

Customer concentration is also a risk factor.

The risk factors section states: 'If there are any adverse changes in the demands of our largest customers... it may have an adverse effect on our business and prospects.' The document also mentions: 'Therefore, if the spending of one or more of our largest customers significantly decreases, or if we lose these customers, and this impact is not offset by new customer relationships or increased business from existing customers, it may have an adverse effect on our business, financial condition, and operating results.'

Anthropic did not immediately respond to TechCrunch's request for a comment.

Tip
$0
Like
0
Save
0
Views 16
HKWDB reminds readers to view blockchain rationally, stay aware of risks, and beware of virtual token issuance and speculation. All content on this site represents market information or related viewpoints only and does not constitute any form of investment advice. If you find sensitive content, please click“Report”,and we will handle it promptly。
Submit
Comment 0
Hot
Latest
No comments yet. Be the first!
Related
Autonomous excavator startup Bedrock Robotics raises $600 million, valuing at $4.1 billion
Bedrock Robotics Completes $600 Million Series C Financing, Valued at $4.1 Billion. The company claims that its autonomous excavators have been put into commercial operation at multiple construction sites and will use the new funds to accelerate commercial expansion.
TechCrunch
·2026-10-27 01:48:47
15
Bitcoin falls below $80,000 for the third time in three days; Solana tests the $100 mark as market focuses on the Fed's decision
Before the Federal Reserve meeting, traders reduced their risk appetite, and Bitcoin tumbled nearly 2% to around $79,300 in early trading on Monday, falling below $80,000 for the third time in three days; Solana dropped to around $100. Meanwhile, Strive continued to increase its holdings of Bitcoin, while Galaxy Research believes that now is the appropriate time to buy and hold Bitcoin.
SolanaFloor
·2026-10-27 01:48:45
16
India’s CoinDCX confirms a loss of $16 million, days after CEO claimed it was a “false alarm”
Indian cryptocurrency exchange CoinDCX confirmed that it lost 15.99 million USDT, approximately $16 million, in an attack on its wallet infrastructure. The company stated that customer funds are still safe and that it will bear this loss itself; previously, its co-founder and CEO Sumit Gupta referred to the related leakage reports as "false alarms."
Decrypt
·2026-10-27 01:48:43
16
A new way to fuel the betting craze with AI bonds and ETF
As the infrastructure financing of AI increasingly relies on bond issuance, at least five issuers have recently applied for ETF focused on AI-related loans. Analysts say that the risk pricing in the credit market of AI may be more attractive than that in the stock market, but there are significant differences in the holdings of different funds, and they may face the risk of liquidity freeze when the market is under pressure.
Yahoo
·2026-10-27 01:38:06
17
Verizon Stock Price Falls Despite Adding 406,000 New Pay-After-Use Users and Reaching a $2 Billion AI Agreement with Amazon
Verizon On Monday morning, the stock price fell by about 2.5%, despite the company announcing 406,000 new postpaid mobile users in the third quarter and revealing a deal with Amazon Web Services worth over $2 billion for artificial intelligence infrastructure. The company's revenue fell slightly short of Wall Street's expectations, but adjusted earnings per share and free cash flow were stronger than anticipated.
Coinpaper
·2026-10-27 01:18:07
19
View More