Analyze user feedback data to identify segments with sentiment scores, JTBD, and product satisfaction insights. Use when analyzing user feedback at scale, running sentiment analysis on reviews or surveys, or identifying satisfaction patterns.
Sentiment Analysis is a product-research skill for synthesizing large volumes of user feedback into structured, segment-level insight. Acting as an expert user researcher and feedback analyst, it takes a body of feedback for a given product or subject and organizes what it finds by user segment, sentiment, and business impact, so that qualitative data becomes an actionable picture rather than an undifferentiated pile of comments.
It accepts a range of feedback sources (CSV files, PDFs, survey responses, review data, and social-listening reports) and reads them directly to extract patterns, themes, and sentiment signals. The analysis follows a defined sequence: ingest the data into a working inventory, identify at least three distinct user segments or personas, run thematic analysis to surface recurring pain points and positive feedback per segment, assign sentiment scores on a -1 to +1 scale, assess impact by frequency and severity, and synthesize segment profiles. For each segment the output includes a profile with characteristics and proportion, the segment's jobs-to-be-done and desired outcomes, a sentiment score with satisfaction drivers, detractors, and an NPS proxy where applicable, top positive themes, top pain points with direct quotes where available, a product-segment fit assessment, and two to three highest-impact recommendations.
Use it when analyzing user feedback at scale, running sentiment analysis on reviews or surveys, or identifying satisfaction patterns across a customer base. The skill's best practices keep the output honest and useful: grounding every finding in actual feedback with cited sources, capturing both majority and minority perspectives within a segment, distinguishing feature requests from fundamental pain points, flagging segments with small or uncertain samples, and looking for cross-segment patterns and universal pain points to give a balanced view of product strengths and weaknesses.
CSV files, PDFs, survey responses, review data, social-listening reports, or other feedback sources, which it reads and analyzes directly to extract themes and sentiment signals.
It assigns an overall sentiment score from -1 to +1 per segment, along with key satisfaction drivers and detractors and a Net Promoter Score proxy where applicable.
At least three distinct user segments or personas, each with a profile, jobs-to-be-done, sentiment score, positive themes, pain points, and recommendations.
A segment profile, jobs-to-be-done, a sentiment score with drivers and detractors, top positive themes, top pain points with quotes where available, a product-segment fit assessment, and two to three highest-impact recommendations.
It flags segments with small sample sizes or uncertain sentiment, grounds all findings in actual feedback with cited sources, and captures both majority and minority perspectives.
Quick Setup:
.claude/skills/Repository
phuryn/pm-skills