Rémi KokabiNovember 4, 20266 min read
Mixpanel integration with lemlist through MCP, with prompts to copy
A free workspace adds three new active users this week. Someone on that team is close to championing the product internally. Nobody on your sales side sees it, because the signal lives in Mixpanel and the outreach lives in lemlist, and those two tools never talk to each other.
This article fixes that gap with a mixpanel lemlist integration built on MCP. You get working prompts to copy into Claude, ChatGPT, or Cursor, each one built around a real use case: catching active free accounts, saving accounts whose usage is dropping, and following up the moment someone adopts a key feature.
I’ll show you what to set up first, what the agent actually does behind each prompt, and what to check before you let it run on your full list. By the end, you’ll have a repeatable way to turn product usage into timed outreach, without building a pipeline or waiting on an engineer.
Why connect lemlist and Mixpanel through MCP
Product usage data is the strongest signal a SaaS company owns, and most of it sits in Mixpanel. When both MCP servers are connected in the same AI client, an agent can read that usage data and start the right outreach in lemlist without a CSV export, a Zapier step, or a data pipeline in between.
An MCP server is a standardized interface that lets an AI agent call actions inside a tool using natural language. Each server exposes a set of operations (querying data, creating records, launching campaigns) that the agent discovers and uses on its own. You can learn more about how this applies to outbound in the guide on MCP for sales.
Because both Mixpanel and lemlist publish their own MCP servers, the agent can query a usage cohort in Mixpanel and add matching leads to a lemlist campaign in the same conversation. No code, no middleware.
What you need before you start
Three things to have in place:
The lemlist MCP connected in Claude, ChatGPT, or Cursor. Follow the walkthrough on how to set up the lemlist MCP if you haven’t done this yet.
The Mixpanel MCP connected in the same client. Mixpanel provides an official remote MCP server at mcp.mixpanel.com/mcp, so AI assistants can connect to it without installing or running anything locally.
Permissions scoped to what each use case needs. Grant the agent read access to Mixpanel cohorts and events, and write access only to the lemlist campaigns you want it to touch.
Start with read-only prompts. Ask the agent to pull data and show you what it found before you let it create or update anything in either tool.
1. Sales assist on active free accounts
Free workspaces that adopt fast rarely hear from sales. By the time someone notices the usage spike in a dashboard, the window for a well-timed conversation has already closed. According to lemlist internal data, leads contacted the same day as a signal reply at 21.3%, compared to 8.1% after a week.
Copy this prompt into your AI client:
Find free workspaces in Mixpanel with 3 or more active users this week, get the decision maker at each company from lemlist, and add them to the sales assist campaign.
Here is what the agent does with that prompt. It queries Mixpanel for free-plan workspaces filtered by weekly active user count. Then it searches the lemlist People database for a decision maker at each matching company. Finally, it adds those leads to your sales assist campaign in lemlist, passing the usage count as a custom variable so your sequence can reference it.
Before running this on a full list, confirm the campaign exists, the sequence is reviewed, and the variable name in your template matches the one the agent sends.
2. Save accounts whose usage drops
Churn shows in product data weeks before the cancellation email arrives. A paying account whose active users drop by half in a month is telling you something, and waiting for a support ticket means waiting too long.
Copy this prompt:
List paying accounts in Mixpanel whose weekly active users dropped by half this month and start a check-in sequence for their admins.
The agent reads the usage trend in Mixpanel, identifies accounts that crossed the threshold, then pulls the admin contacts and adds them to a check-in campaign in lemlist. Your sequence can reference the drop percentage or the specific week the decline started, both passed as variables.
Before running this on a full list, verify the check-in campaign uses a tone that fits a retention conversation, and confirm the agent is filtering out accounts already in active support threads.
3. Follow up on a key feature adoption
Users who adopt a key feature for the first time are showing you they are ready for a deeper conversation. That moment is when an upgrade or expansion talk feels helpful instead of pushy. Acting on it the same day matters because speed to lead directly affects reply rates.
Copy this prompt:
Find users in Mixpanel who used feature X for the first time this week and add their team lead to the upgrade campaign.
Replace “feature X” with the actual event name in your Mixpanel project (for example, “dashboard_created” or “api_key_generated”). The agent queries that event, identifies the users, finds the team lead at each company through the lemlist People database, and adds them to your upgrade campaign with the feature name and adoption date as variables.
Before running this on a full list, check that the event name matches exactly what Mixpanel tracks, and make sure the upgrade campaign sequence is finalized.
Guardrails when an agent works in lemlist and Mixpanel
Letting an agent write in two tools at once saves hours, but it also means mistakes travel faster. Put these guardrails in place before you scale:
Review every campaign the agent creates or updates before it goes live. The agent can build the sequence, but a human should approve the copy, the timing, and the audience.
Test on 10 leads first. Run each prompt on a small batch, check the results in lemlist, and only then widen the scope.
Keep unsubscribes and exclusion lists active. lemlist enforces these automatically, but confirm they are on so the agent never contacts someone who opted out.
Limit who on the team can connect the MCP servers. Treat MCP access the same way you treat API key access.
Log what the agent changed. After each session, review the leads added, campaigns modified, and variables written. This is your audit trail.
These steps matter more as you layer in buying signals from other sources. The more inputs an agent acts on, the tighter your review loop should be.
Run it in lemlist
The lemlist MCP server works in Claude, ChatGPT, and Cursor. Once connected, the agent can do the following across your outbound workflow:
Search the People database with 650M+ contacts and 65M+ companies to find the right person at any account.
Enrich leads with waterfall enrichment across 14+ providers for verified emails and phone numbers.
Create campaigns and sequences across email, LinkedIn, calls, and WhatsApp from a single prompt.
Read the unified inbox and campaign stats to monitor performance without switching tabs.
Act on intent signals to prioritize leads showing real buying behavior.
Every prompt in this article chains Mixpanel data with one or more of these capabilities. You can find more workflows and ready-to-use prompts in the full list of lemlist MCP use cases.
Final takeaway
Product usage data in Mixpanel tells you who is ready to buy, who is about to leave, and who just hit a milestone. With both MCP servers connected, an agent turns those signals into outreach in lemlist without a single manual export. The prompts above are ready to copy, test on a small batch, and scale once you trust the output.
Launch AI outbound campaigns
Automate prospecting and personalize AI-powered sequences to book more meetings.