Rémi KokabiNovember 4, 20267 min read
lemlist and Salesforce MCP use cases
A Tier 1 account goes quiet for 90 days in Salesforce. The signal sits right there in the account record. Nobody moves it into outreach because that means exporting a list view, cleaning a CSV, and building a sequence by hand before the moment passes.
This article shows you how to close that gap with a salesforce lemlist integration that covers both the native sync and MCP workflows. You’ll get prompts you can copy into Claude, ChatGPT, or Cursor today, built around real moments: launching a campaign from a list view, pulling contacts missing from an account, briefing an AE before a call. I’ll walk through what each setup does, when to use which one, and the checks to run before you let an agent act on your Salesforce data.
Expect no theory. Just prompts, workflows, and the guardrails to keep an agent from making a mess across two systems at once.
Native integration or MCP, what each one does
lemlist connects to Salesforce through a native integration that syncs leads, contacts, accounts, and activities between the two tools. You can import leads, contacts, and accounts from a Salesforce view or report, and push new lemlist leads back automatically. You map any lemlist field to any Salesforce field, standard or custom, and set the sync direction per field: one-way or two-way.
The MCP adds what the native sync does not cover. When both the lemlist MCP and the Salesforce MCP are connected in the same AI client, an agent can read data from both tools in one conversation, cross-check it, and act on what it finds. That means you can ask a single prompt to compare Salesforce opportunity stages against lemlist campaign activity and take action without switching tabs or exporting anything.
Here is how the two options compare:
Native integration | MCP | |
|---|---|---|
What it does | Syncs leads, contacts, fields, and activities between lemlist and Salesforce continuously | Lets an AI agent read and write in both tools on demand through natural language prompts |
When it runs | Automatically, on a schedule or trigger | On demand, when you send a prompt |
Who sets it up | RevOps or an admin, inside lemlist settings | Anyone who connects both MCP servers in an AI client |
Best for | Keeping records in sync across the team | Cross-checking data, running one-off workflows, and chaining actions across both tools |
The MCP does not replace the integration. Use both: the native sync keeps your data consistent, and the MCP gives you a way to act on that data in real time.
What you need before you start
Three things to have in place:
The lemlist MCP connected in Claude, ChatGPT, or Cursor. Follow the steps in how to set up the lemlist MCP to get it running.
The Salesforce MCP connected in the same client. Salesforce Hosted MCP Servers give AI agents a secure, governed way to interact with Salesforce data and automation using the open MCP standard, and any compatible AI client can connect to a Salesforce org and act on behalf of authorized users. You can find the setup steps in the Salesforce Hosted MCP Servers guide.
Permissions scoped to what the use cases below actually need. Limit the Salesforce MCP to the objects and fields each workflow touches.
Start with read-only prompts. Ask the agent to pull data and show it to you before you let it create, update, or delete anything in either tool.
1. Launch a campaign from a list view
Building a campaign from a Salesforce list view usually means exporting to CSV, cleaning up columns, deduplicating, and importing into lemlist. That manual step is where leads stall.
Copy this prompt:
Import my Salesforce list view of Tier 1 accounts with no activity in 90 days into a new lemlist campaign, and write a 4-step sequence for it.
The agent reads your Salesforce list view through the Salesforce MCP, pulls the matching accounts, and uses lemlist’s CRM import filters to add them to a new campaign. Then it creates a four-step sequence with email, LinkedIn, or call steps based on the channels available in your plan.
Before running this on a full list, preview the imported leads and review the sequence copy. Run it on a batch of 10 first to confirm the fields mapped correctly.
2. Exclude accounts already in a sales cycle
A prospect with an open opportunity in Salesforce should never receive a cold sequence. Sending one risks the relationship your AE already built and wastes a touch that could go to a fresh account.
Copy this prompt:
Before launching campaign X, remove every lead whose account has an open opportunity in Salesforce.
The agent queries open opportunities in Salesforce through the Salesforce MCP, matches them by account or domain to leads in the lemlist campaign, and removes every match. What remains is a clean list of accounts with no active deal.
Before running this on a full list, spot-check a few removed leads to confirm the account matching logic is correct. A mismatch on domain or company name can pull the wrong leads.
3. Find more contacts at target accounts
Account lists in Salesforce often have one contact per account, sometimes zero. That means your outreach hits a single person, and if they do not reply, the account goes cold. Account based prospecting works better when you reach the right group.
Copy this prompt:
For accounts owned by me in Salesforce with fewer than two contacts, find the VP Sales and Head of RevOps in lemlist and add them as contacts.
The agent reads your owned accounts in Salesforce, identifies the ones with fewer than two contacts, then searches the lemlist lead database (650M+ contacts, 65M+ companies) for the titles you specified. It enriches emails and phone numbers through waterfall enrichment across 14+ providers and creates the new contacts in Salesforce. You can learn more about how to find decision makers at target accounts.
Before running this on a full list, verify the enriched contacts on a small batch. Check that the titles match and the email addresses are verified.
4. Brief an AE before a meeting
Reps open Salesforce for the opportunity stage, lemlist for the outreach history, and their inbox for the latest reply. Three tabs, three logins, and the meeting starts in five minutes.
Copy this prompt:
For my meetings tomorrow, give me the Salesforce opportunity stage and every message we exchanged with the attendee in lemlist.
The agent reads your calendar, pulls the opportunity record from Salesforce for each attendee’s account, and retrieves every message exchanged in lemlist (emails sent, replies received, LinkedIn messages). It writes a short brief per meeting: deal stage, last activity, and what the prospect said.
Before running this across all meetings, test it on one meeting to confirm the agent matches the calendar attendee to the correct Salesforce contact and lemlist lead.
Guardrails when an agent works in lemlist and Salesforce
Giving an agent write access to two systems means you need clear boundaries. Follow these:
Review every campaign the agent creates before you launch it. The agent can build it, but a human should approve the sequence, the lead list, and the sending schedule.
Test on 10 leads first. Run any new prompt on a small batch, check the output, then scale.
Keep unsubscribes and exclusion lists active in lemlist. The agent respects these, but only if they are turned on.
Decide who on the team can connect the MCPs. Limit access to people who understand what each prompt changes.
Log what the agent changed. After each session, review the actions taken in both tools so nothing slips through.
Run it in lemlist
Every prompt above runs through the lemlist MCP server, which works in Claude, ChatGPT, and Cursor. The MCP server exposes actions across lead search, enrichment, sequence building, and campaign management.
Here is what the agent can do once connected:
Search the lead database (650M+ contacts, 65M+ companies) using ICP filters like role, industry, company size, and tech stack
Enrich leads with verified emails and phone numbers through waterfall enrichment across 14+ providers
Create campaigns and sequences across email, LinkedIn, calls, and WhatsApp
Read the unified inbox and campaign stats to surface what is working
Act on intent signals to time outreach to the right moment
You can explore more workflows in the full list of lemlist MCP use cases.
Final takeaway
The native Salesforce integration keeps your data in sync. The MCP lets an agent cross-check both tools and act on what it finds in a single conversation. Together, they turn Salesforce data into outreach without exports, CSV cleanup, or tab switching.
Turn leads into pipeline
Find prospects, automate follow-ups, and generate more sales opportunities.