Teams across every organization need holistic context that lives across multiple systems (like Pendo, Slack, Google Drive, Confluence, Jira, and Salesforce). 

By connecting these data sources through MCP, they can ask questions that span those systems: leadership combining Pendo and Slack for NPS analysis; engineering combining Pendo, Jira, and customer feedback for prioritization; or support combining Pendo and Salesforce for case resolution. 

This breaks down data silos and enables decisions based on a holistic view rather than fragmented insights.

For the first time, your CEO, customer support rep, and product marketer can all tap into the same powerful intelligence—no specialized skills or constant app-switching required. 

And now, we’re democratizing access to product insights with Pendo’s MCP server, pulling behavioral data into LLMs like Claude, Cursor, and even Salesforce. Instead of logging into dashboards or running manual reports, teams can ask questions in natural language and get answers grounded in real user behavior. 

Sound like something your company could use? Here are 12 prompts to get started with Pendo MCP, from your CEO to your engineering team.

MCP prompts for the C-Suite

1. Updating board slides with real-time product data 

At quarterly board meetings, Pendo’s leadership team often needs to report on progress from last quarter's commitments and update the slides with performance data. 

To do this, they prompt their LLM, “Read the August 2025 board deck's product slides, then propose updates based on product usage from the last quarter." 

Your LLM of choice will return insights on overall adoption trends, product performance metrics, in-app engagement rates, internal champion activity, and power user distribution—everything needed for a data-driven board narrative.

2. Monitoring key business metrics, like MAUs, daily 

Pendo had a goal of reaching 1 billion active users. As our executive team tracked progress toward this goal, they asked Claude or ChatGPT: "How many people are using our product today? Has this increased or decreased from last week?" 

They also used this prompt to account for seasonality, understand how holidays will impact usage, and click deeper into trends. 

3. Analyzing customer sentiment across NPS detractors

Many companies, including Pendo, have a Slack channel that reads out every single NPS response we get in-app. But these survey responses can flood your inbox, making it hard to parse through feedback regularly. 

To understand core themes across detractors (or promoters), Pendo CEO and co-founder Todd Olson connects Claude to both Slack and Pendo, then asks: "Give me customers leaving poor NPS responses in the #nps channel." 

Claude surfaces key themes across detractors, turning a chaotic stream of alerts into actionable insights about what's most frustrating to customers. 

MCP prompts for engineering and IT 

4. Prioritizing engineering work 

Pendo's engineering manager, Tomer, needed a data-driven approach to prioritize work. He did just this with MCP by combining usage analytics, customer feedback, and bug reports. 

To do this, Tomer asked Claude, “Find pages with the greatest usage drops over the last 30 days and attach customer feedback with ARR data. Finally, add relevant Jira tickets.” 

As a result, Tomer got a comprehensive report showing: 

  • How the old dashboard page was impacting events
  • How much ARR it's impacting across accounts
  • How many customer-reported bugs was this driving 

MCP prompts for Product teams

5. Building data-backed cases for roadmap decisions 

Pendo PM Jenn needed to defend a feature investment when engineering wanted to shift resources elsewhere. To gather data, Jenn prompted Claude: "Show me usage data from Pendo for our custom reporting feature over the last quarter. Who's using it most, how often, and what's the retention impact?" 

She found that many high-value accounts use the feature weekly, and that accounts that adopt it in the first 30 days have 40% higher retention. Next, Claude automatically generated a business case in Google Docs proving that this feature drives growth worth investing in.

6. Discovering hidden investment opportunities

Jenn also needed to assess the platform for annual planning and wanted to find features with high usage but quality issues. To do this, she asked Claude, "Are there features with high bug counts in Jira but surprisingly strong adoption metrics in Pendo?" 

Claude found that one of our most-used features, by event count, had bugs in its core functionality—revealing a high-value investment opportunity that deserves attention. With this data in hand, Jenn could quickly have this fix prioritized on the roadmap.  

7. Investigating usage spikes and dips 

Finally, Jenn noticed a sudden spike in a feature she’s unfamiliar with and wanted to understand the issue’s root cause. To do this, she created a Claude Skill that checks visitor metadata, groups data by segments, reviews federal holidays, and searches Slack, Google Drive, and Confluence for relevant signals.

Then, Jenn prompted: "Let's use the Pendo usage investigator to investigate an increase in usage in the agent analytics nav item." Claude identified a 7x spike on Veterans Day and traced it to a Slack announcement that the feature moved from closed beta to open beta. 

8. Prioritizing daily product attention with automated usage summaries

Michelle Green, Director of Product at Cohely, faced a common problem: with hundreds of users active in the platform each day, she didn't know which Pendo sessions to prioritize. Manually reviewing session replays was time-consuming and often meant missing critical signals.

To solve this, Michelle built an agent in Dust.tt that runs every morning at 8 A.M., pulling yesterday's Cohely activity from Pendo. She then prompts: "What happened yesterday?"

The agent returns a daily summary showing activity worth celebrating, concerning patterns that need attention, and specific replays to watch, all contextualized against Cohely's ideal customer journey. With MCP, Michelle knows exactly where to spend her time each day.

9. Getting realistic design feedback grounded in user behavior

Lindsay Frank at Clinical Ink struggled with a common design challenge: feedback was slow, subjective, and disconnected from what users actually do in the product. 

To validate designs against actual usage, Lindsay connected Pendo MCP and Figma MCP to Claude, pulling production data like feature usage, behavioral patterns, and segment insights. She then mapped this data to defined user personas. 

Now, she prompts: "Review this Figma design file and give me feedback as [persona], based on their actual behavior in the product."

Claude returned context-aware design critiques grounded in real user behavior, flagging potential usability issues early in the design process before they reach development.

MCP prompts for Customer Success

10. Auto-populating quarterly business review decks 

A customer success rep at Pendo needed to prepare for a customer quarterly business review (QBR) by updating a slide deck with account data. To do this, they uploaded their standard template to Claude Artifacts and prompted, “Use the BSR template in Google Drive and fill in all the details for this customer." 

Claude then generated a complete deck based on the template, showing data points like: 

  • Active visitors
  • Events
  • Active apps
  • The renewal date
  • Usage hours
  • Strategic priorities
  • Adoption status
  • Opportunities for continued growth

11. Identifying at-risk customers before they churn

Stephanie Tanzar at ATC built a Customer Health Dashboard to help account managers proactively spot customers at risk of churning. Customer engagement data lived in Pendo, but it wasn't actionable.

To fix this, Stephanie connected Pendo MCP to pull login frequency, feature adoption, event engagement, and consistency metrics. Then, she built a health score algorithm (0-100) that flags accounts as Healthy, At Risk, or Critical. 

Account managers at ATC can now prompt, "Show me customers at risk of churn," and "Which new customers aren't adopting key features?"

The dashboard surfaces dormant accounts, customers who've stopped using core features, and new customers who need training. Now, ATC’s account managers know exactly whom to reach out to, vs. waiting around to discover problems until after a customer has already decided to leave.

MCP prompts for Marketing

12. Creating product-informed marketing campaigns 

A product marketer at Pendo, Tate, was planning the GA launch campaign for Agent Analytics and needed to understand historical user behavior within the product. 

To find this out, Tate prompted Claude, "Show me which Agent Analytics features our beta users engage with most. Then, create a launch brief in Google Docs that summarizes the top use cases and user behaviors." 

After this, Claude found that beta users wanted to benchmark the time to complete for agentic and non-agentic workflows, analyze conversations, and understand top use cases. Claude created a v1 launch brief and messaging grounded around these challenges, and gave marketing a deeper understanding of Agent Analytics users. 

AI tools are only as smart as the context you give them. Whether you're an executive preparing board slides, an engineer prioritizing bug fixes, or a support agent resolving cases, Pendo MCP brings product intelligence into your existing workflows. 

Ready to connect Pendo's MCP server to your workflow? Here's how to set it up today.