Legora makes Salesforce data actionable in Claude with Claudeforce

More sellers, more context to find

Legora builds AI-powered legal software for law firms and corporate legal teams. They’re growing rapidly, expanding their sales organization from roughly 100 to more than 650 sellers. 

Equipping new sellers with the knowledge and customer history they needed wasn’t as simple as pointing them to one source. It spanned Salesforce, Gmail, Notion, Gong, and other systems, and sellers had to search account records, emails, calls, campaign activity, and opportunities to piece together current context. And with Legora scaling so rapidly, traditional training and static resources couldn't keep pace with their complex internal sales processes and system requirements.

Legora needed a simple way for sellers to find accurate, up-to-date information in seconds — whether learning an account, preparing for a meeting, or working an active deal.

Salesforce in the flow of selling

Instead of having sellers chase down the latest context, Legora now delivers everything they need to them within Claude, a tool they work in every day. Claudeforce brings live Agentforce Sales data to Claude via Model Context Protocol (MCP) servers, giving account executives (AEs) and business development reps (BDRs) a conversational way to research customers, prepare for meetings, manage opportunities, and keep CRM records current.

For example, if an AE is preparing to meet with a law firm, they can ask Claude to get them up to speed, and it pulls together the account history, past emails, customer calls, campaign activity, and open opportunities into a briefing. The seller can ask follow-up questions to dig deeper without ever having to switch over to the CRM.

After the meeting, that same conversation becomes a way to take action. The AE can ask Claude to log call notes, update annual recurring revenue (ARR), or move the opportunity to a new stage. Claude shows the proposed changes for approval before writing them to Agentforce Sales and offers an option to undo the update afterward.

“A rep at Legora doesn’t even write up their own call notes — that’s automated,” said David Eckstein, CFO. “They walk into a partner meeting at a law firm or a Fortune 500 company, and they already know the practice mix, the deal history, and who’s in the room because the research was done before they even sat down.”

Managers work the same way, asking Claude to inspect pipeline, evaluate deal health, spot patterns across their teams, or generate quick forecasting views. These workflows handle the searching, synthesis, and CRM administration, while sellers stay focused on customers and remain responsible for decisions and record changes.

“Claudeforce provides near-immediate access to insights that historically required someone to open a Salesforce report or dashboard and interpret the data,” said Cory Gottlieb, GTM Systems Manager. “Reps used to manually review reports with thousands of records or export the data to Excel and build a chart to identify trends. Now, they can ask for the exact analysis they need and get it in a fraction of the time.”

One connection to seller context

Behind the conversational experience, MCP servers connect Claude directly to Agentforce Sales. This gives Claude access to the live Salesforce data that powers Legora’s sales process, including Accounts, Leads, Opportunities, roughly 10 ARR fields, and their seven-stage opportunity lifecycle.

Einstein Activity Capture and Sales Engagement connect sales activity to the sales records so Claude can also draw from past emails, calls, meeting transcripts, contacts, and calendar events. Campaign history and inbound lead details, such as job role, geography, and law firm type, add more context.

When a seller asks a question, Claude uses MCP to find the relevant Salesforce data and reason across it before responding. The connection also supports create, read, update, and delete actions when sellers need to change a record.

“When I think about Claude and ChatGPT, that’s about how people work and Salesforce is what they leverage to get that work done,” said Eckstein. “I would push back on any framing that says the AI layer will eventually replace the system of record. To me, it makes the system of record more valuable because the agent is only as good as the context you give it, and that context has to come from an authoritative layer.”

When updating the record, Legora built a custom confirmation skill. Claude shows the proposed changes for human approval, writes the approved values to Salesforce, and then offers an immediate rollback if the seller needs to undo the update.

“We’re seeing a lot of Salesforce usage through Claude shift from simple read-and-lookup workflows toward record creation, updates, and even deletion,” said Cory Gottlieb, GTM Systems Manager. “Those write-back actions now account for almost 20% of our headless tool-call volume.”

More selling, less CRM work

“Updates go into Salesforce via natural language now, or a pipeline review comes directly from the system rather than someone building slides from scratch,” said Eckstein. “To me, manual work is wasted mental calories. AI should handle the busy work — that’s AI’s job. The human layer adds the judgment and the sales strategy on top. That is the future of Salesforce and Legora.”

Legora’s custom workflows and skills within Claude take on the manual work of gathering and organizing customer information. Instead of spending hours searching across systems before a call, AEs can pull Gmail threads, customer calls, campaign activity, and opportunity history into a comprehensive meeting briefing in seconds.

That automation gives sellers more time back throughout the sales cycle. With less manual research and CRM data entry, AEs and BDRs can manage larger workloads, spend more time qualifying leads, and stay focused on active selling and longer customer conversations.

The approach also supports Legora’s rapidly growing sales organization without adding CRM administration at the same rate. New and tenured sellers alike can quickly find current customer and product context, act on it, and update Salesforce conversationally — spending less time clicking through systems and more time moving customer relationships forward.