- Home
- Skills
- UX & UI Design
- Customer Research and Voice of Customer
More screenshots
Works with the AI Agents you already use
Customer Research and Voice of Customer
Synthesize transcripts, tickets, surveys, and reviews into graded themes, verbatim quote banks, and evidence based personas.
$5
Customer Research and Voice of Customer
Example session with this skill installed
We sell invoicing software for solo consultants. Trial users are not converting. I have 6 exit surveys, 22 support tickets from trial users, and 14 public reviews of a direct competitor. No interviews yet. Why are trials not converting, and how confident are you?
- Read your context and instructions
- Compiled the customer research
Categorized tickets first: 9 confusion, 7 bugs, 4 missing features, 2 expectation mismatches. The confusion cluster centers on setting up tax rates.
Theme 1, High confidence: Time to first invoice blocks conversion. It appears in 3 independent source types, unprompted. Frequency: 4 of 6 exit surveys, 9 of 22 tickets, and 8 of 14 competitor reviews mention speed or setup. Intensity is high, with language including gave up and wasted my evening. Representative quote: I spent longer setting it up than doing the actual work it was meant to bill for (exit survey, trial user).
Theme 2, Low to Medium confidence: Price sensitivity. Most exit surveys picked price in multiple choice, then described time and confusion in their own text. The two answers conflict, and the conflict is the finding. Do not cut price on this evidence.
Jobs: get paid without spending Friday afternoon on admin (functional plus emotional). Trigger, seen in 3 sources: landing a first larger client. Exact language captured: I just need to send it and stop thinking about it.
Implication: fix the tax setup step in onboarding before any pricing change. Gap: no interviews with converted users, so the success path is unmapped. Next step is 5 interviews, 3 converted and 2 churned, focused on the first invoice session.
Connects securely to your tools. The creator never sees your data.
About this skill
This skill extracts what customers actually say and struggle with, in their own words, from the material you already have or can gather in public. It works in three modes: analyzing existing transcripts, surveys, tickets, and reviews; mining public reviews and communities; and designing interviews and PMF surveys when no signal exists yet.
Every theme is scored for frequency and intensity and labeled High, Medium, or Low confidence using a fixed rubric based on source count, whether it was mentioned unprompted, and consistency across segments. Contradictions and sample bias are reported, not averaged away, and verbatim quotes are preserved exactly for use in copy and product work.
Outputs include a synthesis report, a quote bank organized by theme, jobs to be done maps, and personas built only from at least 5 data points per segment. Fields without evidence stay blank.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for UX & UI Design.
This skill synthesizes and grades evidence; it cannot create signal that does not exist. Conclusions from fewer than 5 data points per segment are labeled Low confidence, and personas are not built below that threshold. It does not recruit participants, run live interviews, or access your CRM or support system on its own.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 25 days ago
Permissions
No special permissions declared or detected