The median U.S. business paying for AI on Ramp spent $15.70 per employee per month on AI in September, up from $6.32 in January, according to the monthly series published on the Ramp AI Index. By our arithmetic, that is 2.48 times the January level in eight months.
Adoption moved far less. Ramp says 55.9% of U.S. businesses on its platform paid for AI in September, against 45.52% in January in the same series: 1.23 times, or 10.3 percentage points, by our arithmetic. As we read the two series, more of this year’s growth shows up in spend per worker at firms already paying than in the number of firms paying.
Business Pill · Usage-Based Pricing
A short explainer of usage-based pricing, where customers pay for what they consume rather than a flat fee per seat. It teaches the general idea only and does not discuss Ramp’s data.
The key insight: As we read Ramp’s series, the median firm paying for AI spent 2.48 times as much per employee in September as in January, by our arithmetic, while the share of businesses paying for AI rose 1.23 times: spend per worker grew faster than the number of firms paying.
The index comes from Ramp Economics Lab; the page credits Ara Kharazian, Ramp’s lead economist. Ramp says its sample includes “more than 85,000 American businesses” and uses data from its corporate card and bill pay platform. The page says it covers data through September 2026.
Ramp defines adoption as “the share of businesses with spending on AI models, subscriptions, tokens, and APIs”. It says “Spend per employee covers only firms paying for AI”, and that for the top 10% and top 1% of those firms the figure is the median within each group.
AI spend per employee per month at the median firm and the top 10% of firms paying for AI on Ramp, and the share of businesses paying for AI, indexed to January 2026 = 100. Data: Ramp AI Index. The multiples are our arithmetic (September / January).Animated: median monthly AI spend per employee among firms paying for AI on Ramp, January to September 2026. Data: Ramp AI Index. The 2.48x multiple is our arithmetic.
The Top of the Distribution Moved Faster
The top 10% of firms paying for AI spent a median $813.43 per employee per month in September, up from $296.72 in January: 2.74 times, by our arithmetic. The top 1% spent a median $9,567.53, up from $3,390.40, or 2.82 times; that series is more volatile, peaking at $9,873.12 in July before falling to $8,568.17 in August.
By our arithmetic, the top-10% figure was 47 times the median firm’s in January and 52 times in September. The gap widened even as the median itself rose 2.48 times. We looked at the top 1% in how 1% of Ramp firms drive most AI spend.
Ramp’s series also tracks AI’s share of business spend excluding payroll. It was 3.23% in January and 5.77% in September, down from 5.96% in August and 6.27% in July.
Where the Tokens Go
Ramp’s token data, from its Token Spend Management product, runs later than the monthly index. In the week ending 23 October, Ramp says Anthropic took 60.5% of token spending among businesses on Ramp, up from 51.3% in the week ending 25 September; OpenAI’s share fell to 35.1% from 44.2%.
Over the same weeks, Ramp says total token spending rose 8.9% while token volume rose 37.1%. The blended price per million tokens for OpenAI and Anthropic fell 26.8%, to $0.56 from $0.76. Ramp says the token sample “is not representative of Ramp’s total AI spend volume, though results are directionally similar”.
On adoption by vendor, Ramp puts Anthropic at 43.8% of businesses in September and OpenAI at 40.6%, with OpenAI gaining 1.1 points from August. Large businesses were at 65.7% and small businesses at 50.0%.
Why It Matters
As we read the series, AI vendors’ revenue on Ramp is growing through deeper use inside paying customers as well as through new ones. That is the backdrop to our piece on five AI pricing moves pointing from seats to usage: when spend per worker rises faster than the number of buyers, pricing tied to usage tracks that growth more closely than a flat price per seat.
As we read the token figures, a 26.8% fall in the blended price per million tokens alongside a 37.1% rise in token volume left total token spending up 8.9% across the four weeks Ramp compares: cheaper tokens did not shrink the bill.
As we read the distribution, the top 10% of firms paying for AI (2.74 times since January) moved faster than the median firm (2.48 times), so the heaviest AI spenders on Ramp pulled further ahead in 2026 even as typical spending rose.
Ramp AI Index methodology
“Firms are considered to have adopted AI tools if they have a positive transaction amount for an AI product or service in a given month”
Three Implications
FOR AI VENDORS Ramp’s median AI spend per employee rose 2.48 times from January to September while adoption rose 1.23 times, by our arithmetic.
FOR FINANCE TEAMS Ramp’s series puts AI at 5.77% of business spend excluding payroll in September, against 3.23% in January.
FOR THE MODEL RACE Ramp says Anthropic took 60.5% of token spending in the week ending 23 October and OpenAI 35.1%.
The framework says: “lower cost per token multiplied by exploding token volume increases total spend.”
As we read it, Ramp’s token data shows that arithmetic over four weeks: the blended price fell 26.8%, volume rose 37.1% and spending rose 8.9%. The framework was published in June and does not discuss Ramp’s index.
What Is Not Established
Ramp’s sample is businesses on Ramp, not all U.S. firms. The spend-per-employee figures are medians among firms paying for AI, so the group changes as new firms start paying, and the 2.48x is not the growth of any single firm. Ramp says its results likely underestimate adoption and spend because of free tools and personal accounts. The January figures come from the monthly series on Ramp’s page, not from its summary table, and October’s monthly figures are not yet on the page.
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.
The median firm paying for AI on Ramp spent $15.70 per employee per month in September, 2.48 times January’s $6.32 by our arithmetic, while the share of businesses paying for AI rose 1.23 times. The top 10% of spenders grew faster still, at 2.74 times.
A note on sourcing. All figures are from Ramp’s AI Index page (Ramp Economics Lab), read on 10 November 2026: its summary table, methodology and the monthly series behind its charts. The multiples (2.48x, 2.74x, 2.82x, 1.23x) and the 47x and 52x gaps are our arithmetic on those figures.
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.
