Product Research
Product Research is a research-ops skill that helps you plan and synthesize user research with methodological rigor. It maps research goals to the right method, computes defensible sample sizes with stated confidence, and clusters coded observations into insights while flagging single-source anecdotes. It produces study plans, saturation rationales, and governed insight repositories.
✨ What it does
- Maps research goal and product stage to an appropriate method via study_designer.py.
- Provides method-based sample size guidance with confidence labels via saturation_planner.py.
- Clusters coded observations by tag and ranks by cross-participant recurrence via insight_synthesizer.py.
- Flags single-source findings as ANECDOTES, never promoting them to insights.
- Includes an onboarding questionnaire to customize defaults like insight source-threshold.
- Offers an opt-in autoresearch bridge to optimize coding/clustering of evidence.
🎯 When to use it
- When you need to choose the right research method for a product goal (generative, evaluative, or validation).
- When you need a defensible sample size or saturation rationale with an explicit confidence level.
- When you have raw coded observations and need to synthesize them into insights without over-claiming.
- When setting up or auditing a research repository and need insight-vs-observation discipline.
🚀 How to use
Trigger: /ar:setup
Trigger the skill by asking for help with product research planning or synthesis, or by running its scripts directly. It expects you to provide a research goal, product stage, and optionally a profile (e.g., b2b-saas). For synthesis, provide coded observations in JSON. Example prompts:
Help me plan a usability test for our beta consumer app.
Synthesize these observations into insights, requiring at least 3 sources.
📄 Output: Study plan skeletons, saturation guidance with confidence labels, and synthesized insights with anecdote flags.
📦 Add this skill to Claude Code
# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills
# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/research-ops/skills/product-research .claude/skills/product-researchSkill source: research-ops/skills/product-research/SKILL.md
⚠️ Good to know
Requires honest goal framing; saturation guidance is method-based, not a power calculation; coding quality is upstream of synthesis.
❓ FAQ
What is the default insight source-threshold?
The default is 3 independent participants; you can raise it for high-stakes or heterogeneous populations.
Can this skill generate personas or journey maps?
No, that is handled by the product-team/ux-researcher-designer skill; this skill focuses on method and repository discipline.
Does the autoresearch bridge fabricate evidence?
No, it optimizes coding only and never fabricates evidence; it runs only on demand.
🤖 Overview, features, install steps and FAQ were generated from the project's SKILL.md on Sep 4, 2026. Always check the original source before running commands.