Customer Research

• You have a collection of customer interview or sales call transcripts and need to extract recurring themes, objections, and exact language for copy or positioning.
• You're building or refreshing customer personas and need them grounded in real data rather than assumptions.
• You want to understand why customers churn, convert, or choose competitors by analyzing support tickets, NPS verbatims, or win/loss notes.
• You have no existing research and need to gather VOC from online communities — Reddit threads, G2 reviews, app store ratings, or niche forums — to understand how your ICP talks about the problem space.

• You already have finished research insights and need help writing conversion copy from them — use the copywriting skill instead.
• You want to act on VOC findings to improve a specific landing page — use the page-cro skill instead.

Supports two distinct workflows: analyzing assets you already own (transcripts, surveys, tickets, reviews) and actively gathering research from online sources. Most engagements combine both — the skill establishes which mode applies and adapts accordingly.

Every research asset is processed through a consistent extraction lens: Jobs to Be Done (functional, emotional, social), pain points, trigger events, desired outcomes, customer vocabulary, and alternatives considered. Exact verbatim quotes are preserved — not paraphrased.

Guides sourcing from the right online communities based on ICP type — B2B SaaS buyers, SMB founders, developers, B2C consumers, or enterprise. Includes per-platform extraction tips for Reddit, G2, Hacker News, app store reviews, LinkedIn, and more.

Every insight is labeled High, Medium, or Low confidence based on how many independent sources support it, whether it was mentioned unprompted, and whether it's consistent across segments. Prevents over-indexing on single anecdotes.

Personas are built only when sufficient data points exist (minimum 5–10 per segment). Each persona includes trigger events, top pains, desired outcomes, objections, vocabulary, and how to reach them — with explicit guardrails against invented or averaged details.

Outputs can take the form of a research synthesis report, VOC quote bank, persona documents, Jobs-to-be-Done map, competitive intelligence summary, or a research gap analysis — depending on what the user actually needs.

Upload 10–20 customer interview transcripts and the skill clusters them by segment, extracts recurring pains and trigger events, surfaces the most representative verbatim quotes, and produces structured persona documents ready to inform messaging or product decisions.

Before writing a homepage or email sequence, use this skill to mine G2 reviews, Reddit threads, and support tickets for exact language customers use to describe their problem. The output is an organized quote bank, segmented by theme, that copywriters can pull from directly.

Feed in churned-customer NPS verbatims, cancellation survey responses, and support ticket history. The skill segments churn by cause (price, fit, features, timing), identifies the highest-signal patterns, and flags what messaging or product changes might address each.

When entering a new market or refreshing positioning, use the skill to mine competitor reviews on G2 and Capterra — especially 3- and 4-star ratings — to identify switching triggers, unmet needs, and language competitors' customers use that you can address directly.