How CMS rolled out Harvey to thousands of lawyers after a 12-month pilot
International law firm CMS piloted Harvey for twelve months, then scaled it across the firm for contract review, document analysis and knowledge synthesis. Harvey reports high adoption.
The problem
CMS leadership expected generative AI to change legal work and saw that clients already had similar tools. Simply handing lawyers a generic tool wasn't seen as enough.
How they did it
- Ran an initial trial, then a twelve-month pilot with 300 lawyers across several practice areas.
- Tested Harvey on live matters: reviewing contracts, summarizing documents and pulling insights from curated knowledge sources.
- Focused on hands-on onboarding and sharing real use cases rather than abstract training.
- Let lawyers share their own examples to build momentum across the firm.
Results
- Harvey reports 95% adoption across 3,000+ lawyers during the initial rollout.
- The deployment later scaled firmwide to 7,200 lawyers across 50+ countries.
- Qualitatively, the source says teams review much larger document sets faster. It gives no time-saving figures.
As reported by the source (Harvey customer story); AgentGid did not measure these figures.
This suits large firms with a dedicated rollout team and the patience for a year-long pilot; CMS started with 300 lawyers before scaling. Harvey has no public pricing and is sold only through sales, and the 95% adoption figure is vendor-reported with no time-saving numbers. A smaller team could try Spellbook (Custom, with a 7-day free trial).
The agent used here
Similar use cases
How Oxford PharmaGenesis ran a 500-paper literature review in a week with Elicit
Health-science communications consultancy Oxford PharmaGenesis used Elicit for screening, data extraction and reporting in literature reviews, with experts checking the results. E…
Elicit reports answering 40 research questions across 500 papers in under a week.
How an estate-planning lawyer saves drafting time with Spellbook
An estate-planning lawyer at CunninghamLegal uses Spellbook daily to draft clauses, write client memos and double-check agreements. Spellbook reports he saves time every day.
Spellbook reports savings of 15-20 minutes per clause drafted.
Run LangChain's Open Deep Research multi-agent pipeline locally
LangChain's open-source deep research agent, built on LangGraph, runs research sub-agents in parallel, compresses what they find and writes a final report. It works with many mode…
On Deep Research Bench, the default GPT-4.1 configuration scored 0.4309 and the GPT-5 configuration 0.4943.
