Summarize Interview

What Is This?

Overview

Summarize Interview is a structured skill designed to transform raw customer interview transcripts into organized, actionable summaries. It applies a consistent template that captures Jobs to Be Done (JTBD) frameworks, satisfaction signals, and prioritized action items from discovery conversations. Rather than leaving teams with pages of unprocessed notes, this skill produces a clean output that product managers, researchers, and designers can immediately act on.

The skill processes interview recordings or written transcripts and extracts the most meaningful signals from customer conversations. It identifies what customers are trying to accomplish, how satisfied they are with current solutions, and what specific follow-up actions the team should take. The output follows a repeatable format, which makes it easier to compare insights across multiple interviews and build a reliable body of customer evidence.

This skill is part of the phuryn/pm-skills collection, a set of product management focused tools built to accelerate research synthesis and decision-making workflows. It is particularly valuable when teams conduct frequent discovery interviews and need a consistent method for capturing and sharing what they learn.

Who Should Use This

• Product managers who conduct regular customer discovery interviews and need to synthesize findings quickly
• UX researchers who want to standardize how interview insights are documented and shared across teams
• Product designers who need structured customer context before beginning solution work
• Customer success managers who gather feedback during account reviews and want to convert conversations into product insights
• Startup founders running early-stage discovery interviews who need a lightweight but rigorous documentation process
• Business analysts responsible for translating customer conversations into requirements or backlog items

Why Use It?

Problems It Solves

• Raw transcripts are time-consuming to review and easy to misinterpret without a structured extraction process
• Teams often capture different things from the same interview, leading to inconsistent or conflicting takeaways
• JTBD insights get buried in conversational text and are rarely surfaced in a format that informs roadmap decisions
• Action items from interviews are frequently lost or forgotten when they are not explicitly captured and assigned
• Comparing insights across multiple interviews is difficult when each summary follows a different format or level of detail

Core Highlights

• Applies the Jobs to Be Done framework to identify what customers are actually trying to accomplish
• Extracts satisfaction signals to distinguish between pain points and areas of genuine delight
• Generates structured action items with clear ownership potential
• Produces consistent output format across all interviews for easier cross-interview analysis
• Reduces synthesis time from hours to minutes for experienced practitioners
• Works with both verbatim transcripts and rough interview notes
• Supports discovery, validation, and continuous research interview types
• Integrates into existing product management workflows without requiring new tooling

How to Use It?

Basic Usage

To use this skill, provide the raw interview transcript as input. The skill expects plain text content from the conversation.

Input: [Paste full interview transcript here]

Expected Output:
- JTBD Summary
- Satisfaction Signals (positive and negative)
- Key Quotes
- Action Items

A minimal invocation using a prompt-based interface might look like this:

Skill: summarize-interview
Input: "Today I spoke with Sarah, a senior product manager at a mid-size SaaS company..."

Specific Scenarios

Scenario 1: Post-Interview Processing After completing a 45-minute discovery call, paste the auto-generated transcript from your recording tool directly into the skill. The output will identify the core job the customer is trying to accomplish, note where they expressed frustration or satisfaction, and list follow-up items for the team.

Scenario 2: Batch Synthesis When processing five interviews from a single research sprint, run each transcript through the skill separately. The consistent output format allows you to place all five summaries side by side and identify patterns across participants.

Real-World Examples

A product team preparing for a quarterly planning session used this skill to process eight customer interviews in under two hours, producing a consolidated view of the top three unmet jobs across their user base.

A UX researcher used the skill to prepare interview summaries before a design sprint, giving the team structured customer context without requiring them to read full transcripts.

Important Notes

Requirements

• A complete or near-complete interview transcript is needed for accurate extraction
• The transcript should be in plain text format for reliable processing
• Interviews should have a clear customer focus rather than being internal team discussions
• Basic familiarity with JTBD concepts helps teams interpret and act on the output correctly