Spellbook's Word Integration Beats Chat
Scott Stevenson, CEO of Spellbook, on building Cursor for contracts
The real moat in legal AI is workflow ownership, not raw model access. Spellbook wins by living inside Microsoft Word, where lawyers already redline contracts, and by turning AI into concrete actions like flagged issues, track changes, playbook enforcement, and market comparison charts, instead of a blank prompt box. That matters because its core buyers are high volume in-house teams, where faster contract turnaround directly helps revenue move and cost come down.
-
Spellbook is used by about 4,000 customers, and about 60% of revenue comes from corporate in-house teams. Those buyers process heavy contract volume, care about speed more than billable hours, and often start with one user before expanding across the organization.
-
The product is not just drafting help. It reviews agreements against company playbooks, edits documents with track changes in Word, and can compare a lease against market norms using charts. Those are workflow products tied to specific contract jobs, not generic legal chat.
-
This is also why Spellbook sits differently from Harvey, Legora, and CLM vendors like Ironclad and Icertis. Harvey and Legora skew toward lawyer facing assistants, while CLMs own post drafting workflow. Spellbook is taking the contract review and redlining seat where daily user habit gets formed.
The market is heading toward contract systems that start in Word, then pull in email, Slack, intake, and approval workflows around that drafting surface. The companies that win will look less like chat apps and more like always on contract infrastructure, where AI does the first pass before a lawyer even opens the file.