Ramp runs on Notion: How they built an AI operating system for work

Notion is an AI-powered workspace platform that combines docs, projects, wikis, and databases into a single system of record. It offers AI features including meeting notes, enterprise search, and autonomous agents that can automate workflows. Notion's developer platform allows companies to connect data from any tool and build custom agents without infrastructure overhead.

Ramp is a fast-growing fintech company that provides corporate cards and spend management software. In 2025, the company surpassed $1 billion in annualized revenue and reached a $32 billion valuation—growing at a rate 10× faster than the median publicly traded SaaS company. Ramp set an internal mission to be the most productive company in the world, using AI and operational excellence to scale output without proportionally scaling headcount.

Before adopting Notion, Ramp's work was fragmented across too many duplicate and legacy tools. Simple questions like "What is the latest policy?" or "Who owns this?" turned into scavenger hunts across docs, tickets, and chat threads. Without a reliable system of record, AI couldn't do much more than summarize fragments. Meetings dragged on, key context lived in disconnected tools, and the lack of a single source of truth slowed decision-making and made it hard to trust what was accurate.

Ramp chose Notion as their system of record to consolidate and connect work in one place, making it legible for both humans and AI. They built a stack with Notion at the center, using AI Notes and Search to capture and find everything, and Notion Agents to turn answers into action. Notion's AI meeting notes captured conversation context close to the work, AI search pulled in data from Slack, GitHub, and Ramp's own systems, and personal Notion Agents helped people draft, analyze databases, route requests, and build workflows.

Ramp's Head of Operations & Internal AI, Ben Levick, led the transformation with a philosophy to multiply impact, not add busywork. They started by using AI meeting notes to capture context and decisions without manual note-taking. Then they expanded AI search to pull in connected data from multiple tools. Finally, they gave everyone personal Notion Agents and encouraged the entire company to become builders. In mid-2025, Ramp began experimenting with autonomous Notion Agents, and within months had deployed over 300 custom agents. Agents were created in as little as three minutes between meetings, automating hours of manual operational work.

With Notion as their foundation, Ramp cut productivity-tool costs by roughly 70% and teams reported moving about 3× faster. In a single year, the company's valuation more than tripled to $32 billion as annualized revenue surpassed $1 billion—at a rate 10× faster than the median publicly traded SaaS company. Headcount scaled more modestly because output per employee kept rising. The company deployed over 300 Notion Agents handling everything from product Q&A to sales feedback categorization to customer advocacy mining. Agents could be built in three minutes, and hours of manual operational work disappeared. Employees experienced two "dopamine hits"—one when they automated work they didn't like doing, and another when they had more time in their day.