Pendo investigation

When this applies

Use this when Pendo already shows feature adoption, a guide outcome, or a poll, and the reason is still open. The connector is https://mcp.usercall.co. The entry tool is create_study. The shared loop is on Ask users why.

Signal

Keep the Pendo numbers and the exact guide or feature. A poll comment that says "confusing" is a signal with almost no content. NPS is a score. Neither one is the cause.

Example: The billing-page guide was dismissed by 60% of viewers in the last 30 days (2,100 views, 1,260 dismissals). Feature adoption for the new invoice download is 11% among accounts that saw the guide.

Unknown

The open question is why people dismiss the billing guide and then skip invoice download. They may already know the page, distrust the guide, or fail to find the control the guide points at. The dismissal rate has not separated those.

Interview affected users

Interview people who dismissed that guide, or who saw it and did not download an invoice. Leave the rest of the account out. The study should quote the behavior, not the poll score.

create_study({
  key_research_goal: "Why do people dismiss the billing-page guide and skip invoice download? Ask what they came to the page to do, what the guide seemed to be for, and where they went next.",
  business_context: "Subscription billing. Pendo, last 30 days: billing-page guide dismissed on 1,260 of 2,100 views (60%). Invoice download adoption is 11% among accounts that saw the guide. Interview people who dismissed the guide or saw it and did not download an invoice.",
  target_interviews: 8,
  duration_minutes: 12
})

Share interview_link with that group. Poll get_study_status with study_id until the study is complete.

Evidence

get_study_results({
  study_id: "<study_id>",
  format: "summary"
})

Set the themes next to the 60% dismissal and the 11% adoption. Quotes, when you need them, come from format: "full". A second poll that still only collects "confusing" will not add this evidence.

Stop here

If the guide targets the wrong URL, or the download control is absent for that segment, the Pendo setup is the finding. Interview people after the guide can actually be completed.