Anthropic’s IPO roadshow video says its biggest customers have barely started: “On average, spending from these Fortune 500 customers is less than 0.1% of their total expenses”, according to the video’s official captions. “Even among the top ten, it’s still less than 1% of their total expense line”.
The same video says more than 50% of the Fortune 500 spends more than $100,000 with Anthropic, more than 250 companies by our arithmetic. A slide posted on X by Tanay Jaipuria on 11 November shows knowledge work at “Early agents”, with its “Threshold ahead”; Jaipuria summed it up as “still very early in agents outside of coding”.
Business Pill · The Model and the Harness
An evergreen explainer on why products built on the same model differ. It teaches the general idea only and does not discuss Anthropic’s customers.
The key insight: Anthropic’s IPO roadshow video says Fortune 500 customers spend less than 0.1% of their total expenses on Claude, and under 1% even among the top ten, while more than half of the Fortune 500, more than 250 companies by our arithmetic, spend over $100,000 with it. A slide puts knowledge work at early agents, behind coding and cyber.
The captions, which do not name speakers, read: “More than 50% of the Fortune 500 spends more than $100K with us. And that’s true for more than 30% of the Global 2000”. By our arithmetic, that is more than 600 of the Global 2000. The video is on Anthropic’s site for US investors.
The financial segment returns to the figure: “our Fortune 500 customers spend less than 0.1% of their total expenses on Claude”. It frames the market as “knowledge work, a $40 trillion market that we have just begun to penetrate”.
Our IPO timing piece quoted these lines in an update box. The slide adds the timeline the captions describe.
Reach versus depth: the share of companies spending more than $100,000 with Anthropic, and Claude’s share of customers’ total expenses, as stated in Anthropic’s IPO roadshow video captions. Values are plotted at the stated bound.
What Was Posted on X
Some slides from the Anthropic roadshow video
– big emphasis on fungible compute, ability to use compute efficiently and onboard new chips quickly – still very early in agents outside of coding – long-term target of 35% adjusted operating margins
Tanay Jaipuria posted on 11 November at 01:37 UTC four images the post calls slides from Anthropic’s roadshow video, noting that it is “still very early in agents outside of coding”. One image shows knowledge work at “Early agents”, behind coding and cyber; this piece sets that slide against the video’s captions on Fortune 500 spending and the S-1/A’s customer counts.
Coding First, Then Cyber, Then the Rest
The slide, titled “Making fast progress across knowledge work”, draws three curves. Coding is marked “Nov ’25 Threshold crossed” and cyber “Apr ’26 Threshold crossed”, five months apart by our arithmetic. Knowledge work sits at “Early agents” today, with “Autonomous fleets of agents” drawn beyond the threshold.
Its footnote reads: “Illustrative curve. Threshold = the release after which long-running, multi-agent autonomous work became the frontier default”. The slide gives no date for the knowledge-work threshold.
The captions describe the same order: “Coding is the example we’re most familiar with”, then “After coding came cyber”, then “Knowledge work is on a similar curve”. As examples they name scientific research, investment analysis and underwriting, and say Claude helped AIG “cut down time to quote by 55%”.
How Customers Move Beyond Code
The video describes a customer journey that “starts with a single use case”: “Often that’s Claude Code for developers and engineers, and it soon expands to adjacent roles”. It ties that path to “an unprecedented net revenue retention rate of approximately 900%”.
The S-1/A filed on 10 November counts more than 11,900 customers with $100,000 or more of trailing revenue at 30 September 2026, up from more than 950 a year earlier, about 12.5 times by our arithmetic. More than 1,900 were at $1 million or more and about 250 at $10 million or more. Anthropic’s run-rate is covered separately.
The filing says adoption of Claude Cowork, the agent for non-coding work, “outpaced the adoption Claude Code achieved at the equivalent point following its research preview release in February 2025”. It puts knowledge workers at “a population that is estimated to exceed one billion in 2028”.
The Structural Read
As we read the video, the 0.1% is its headroom argument: reach across the largest companies is already wide, the spend inside each is small, and the video calls knowledge work a $40 trillion market it has just begun to penetrate.
As we read the slide, the order matters more than the shape: Anthropic itself labels the curves illustrative and dates only the coding and cyber thresholds, so the knowledge-work step is presented as ahead, with no date.
Anthropic IPO roadshow video (captions), 10 November 2026
“On average, spending from these Fortune 500 customers is less than 0.1% of their total expenses”
Three Implications
FOR REACH More than 50% of the Fortune 500 and more than 30% of the Global 2000 spend over $100,000 with Anthropic, the video says; more than 250 and more than 600 companies, by our arithmetic.
FOR DEPTH Claude is under 0.1% of Fortune 500 customers’ total expenses on average and under 1% for the top ten, the captions say; the video does not define the measure.
FOR THE FILING The S-1/A counts more than 11,900 customers at $100,000 or more of trailing revenue, up from more than 950 a year earlier, and says Claude Cowork’s adoption outpaced Claude Code’s at the same point.
The framework describes two curves, capability and product: “They are not the same curve, and the gap between them โ what the platform makes possible versus what the product surfaces โ has a name.”
As we read it, the roadshow slide draws that gap for knowledge work, with early agents today and the threshold ahead, and the 0.1% line puts a spending number on it. The framework does not discuss the video.
What Is Not Established
The video does not say how it measured customers’ total expenses, which year it used or whether the 0.1% is a simple or weighted average; the S-1/A text we searched does not give the figure. The slide calls its curves illustrative and gives no date for the knowledge-work threshold. We read the slide as a frame posted on X by Tanay Jaipuria, not from the video file. Anthropic was not contacted for this piece.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Anthropic’s roadshow video says Fortune 500 customers spend less than 0.1% of their total expenses on Claude, under 1% even among the top ten, while more than half of the Fortune 500 spend over $100,000 with it. A slide puts coding past its threshold in November 2025, cyber in April 2026, and knowledge work at early agents.
A note on sourcing. Video quotations are from Anthropic’s own captions file for its IPO roadshow video; slide text is read from a frame of the video posted on X by Tanay Jaipuria on 11 November 2026. Customer counts are from Anthropic’s S-1/A filed with the SEC on 10 November 2026. Company counts and ratios are our arithmetic. Anthropic was not contacted for this piece.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
