How PMs Use MCPs to Automate Meeting Prep & CRM Updates

Lenny Rachitsky - How This PM Uses MCPs to Automate Meeting Prep & CRM Updates#

Source: https://www.lennysnewsletter.com/p/how-this-pm-uses-mcps-to-automate Credibility: High (detailed first-person workflow demonstration with concrete examples)

What happened: Lenny published a walkthrough of using Model Context Protocol (MCP) servers to automate recurring PM workflows—meeting prep, CRM updates, and research aggregation. The key shift: MCPs move AI from "chat with context" to "agent with access."

Key workflow automations demonstrated:

Meeting preparation automation:

• MCP server connects to calendar (Google Calendar API)
• Agent automatically pulls meeting details, participant info, and related context
• Generates pre-meeting briefs with research on attendees and agenda topics
• Updates automatically as calendar changes—no manual prompting required

CRM updates without manual data entry:

• MCP server connects to CRM system (example used: HubSpot)
• After meetings, agent extracts action items and updates CRM records
• Maintains relationship history without PM manually logging interactions
• Pattern: agent observes workflows and updates systems autonomously

Research aggregation:

• MCP servers pull from multiple data sources (docs, Slack, emails)
• Agent synthesizes context across tools without manual copy-paste
• Delivers daily digests of relevant information filtered by specific topics

Why MCPs matter vs. traditional integrations: MCPs provide direct access to organizational data and systems, not just query capabilities. The agent can read calendar, write to CRM, and execute updates—not just answer questions about them. This is the architectural difference between assistants (passive) and agents (active).

Why it matters for PMs: This demonstrates AI transitioning from "I ask, it answers" to "it observes my workflow and handles parts autonomously." The MCP pattern—standardized connectors to organizational systems—means you don't need custom integrations per tool. For PMs evaluating AI productivity tools, the question becomes: does it integrate via MCP? If yes, workflow automation is possible. If no, you're limited to chat.

Critical questions:

• What's the security model when agents have write access to CRM and calendar?
• How do you debug when autonomous updates are incorrect?
• At what point does automation create more overhead (reviewing agent actions) than it saves?
• Which workflows benefit from automation vs. which require human judgment?

Action you could take today: Map your recurring PM workflows (meeting prep, status updates, CRM logging). Identify which tasks are high-volume, low-ambiguity, and involve pulling data from multiple systems. These are MCP automation candidates. If you use tools with MCP support (calendar, CRM, docs), prototype one automation before building custom workflows.

Teresa Torres - February 2026 Book Club: Reading Continuous Discovery Together#

Source: https://www.producttalk.org/cdh-book-club-february-2026/ Credibility: High (structured learning program from author of Continuous Discovery Habits)

What happened: Teresa Torres announced the February 2026 cohort of the Continuous Discovery Habits book club—a guided reading and practice program for PMs learning discovery methods.

Program structure:

• Four-week cohort starting February 25, 2026
• Weekly live sessions with Teresa for Q&A and practice
• Peer learning groups for applying concepts to real product work
• Structured exercises translating frameworks into daily practice

Why book clubs matter for PM craft: Discovery methods are experiential—reading alone doesn't build the skill. The cohort model creates accountability and feedback loops that turn concepts into habits. This is the pattern for PM skill development: frameworks + practice + feedback, not just frameworks.

Why it matters for PMs: This continues the pattern from recent updates (Jan 25, Feb 1): PM craft improves through structured practice, not just consuming content. For PMs wanting to improve discovery methods, the cohort format provides what solo reading can't—feedback on your application of the framework.

Critical questions:

• What's the completion rate for cohort programs vs. self-paced reading?
• How do you apply discovery methods in fast-paced environments that resist structure?
• Which frameworks actually transfer to your product context vs. which remain theoretical?

Action you could take today: If you're planning to improve discovery methods this year, evaluate whether structured learning (cohorts, workshops) would accelerate adoption vs. solo reading. The investment is higher, but the skill transfer rate may justify it.