How to use the Customer Journey Map agent skill

What is an agent skill and what does this one do?

An agent skill is a folder containing a SKILL.md file with instructions that an AI agent loads when a task matches the skill's description. The skill tells the agent how to approach a specific kind of work, so you do not have to write those instructions yourself each time.

This skill, from the pm-market-research collection in phuryn/pm-skills, is licensed under MIT. Once it is installed, the agent can turn a product or customer scenario into a map that moves through the journey from awareness to advocacy. For each stage it documents touchpoints, user actions, thoughts and questions, emotions, pain points and opportunities, then identifies critical moments such as the aha moment and churn triggers, and recommends prioritized improvements.

• Stages: Awareness, Consideration, Acquisition, Onboarding, Engagement, Retention, Advocacy (adapted to the product)
• Output: a markdown analysis with a journey map table (stage, touchpoint, user action, emotion, pain point, opportunity)
• If you provide files such as interview transcripts, survey data, analytics, support tickets or existing journey maps, the agent reads them first
• For a visual map, the skill suggests building one in Miro or FigJam using the analysis

Install the customer-journey-map skill

Download the ZIP for the skill from the directory page and unzip it. The skill folder should contain SKILL.md. Then place it in one of these locations, depending on whether you want it for yourself or for one project.

For a personal install, put the folder in ~/.claude/skills/. For a single project, put it in the project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings → Capabilities → Skills.

• Personal: ~/.claude/skills/customer-journey-map/
• Project only: .claude/skills/customer-journey-map/ inside your project
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP

Trigger the skill with natural requests

You do not need a special command. The skill's description says it applies when you are mapping the customer experience, identifying friction points, improving onboarding or visualizing the user journey. Asking for those things in plain language is usually enough to load it.

Name the product, the customer and the goal. The more specific the scenario, the more specific the persona and stages will be. If you have research files, mention them and provide them so the agent can read them before it writes the map.

• "Map the customer journey for our free trial sign-up and find where people drop off."
• "Create a customer journey map for a freelance designer using our invoicing app, focusing on onboarding."
• "Here are my interview notes and support tickets. Map the journey and suggest the highest-impact fixes."
• "Identify friction points in how new teams adopt our analytics dashboard."

What the output looks like

The skill asks the agent to define a specific persona with jobs to be done, then walk through each stage. Each stage gets touchpoints, user actions, thoughts and questions, emotions, pain points and opportunities. Emotions are rated on a scale or shown with emoji indicators, as the skill instructs.

After the stage analysis, the agent identifies the aha moment, moments of truth and churn triggers, then recommends prioritized improvements. These cover which pain points most affect conversion or retention, which quick wins are available now, and which changes need deeper investment.

What you need and safety notes

The source does not list any required tools or packages, and the skill does not run scripts. It does use the network: the instructions say the agent can use web search to understand the product when you give it a URL. Treat that as a reason to check what the agent fetched before trusting it.

The skill reads files you supply, so only provide material you are comfortable sharing with the tool you are using. Keep your own review in the loop for anything that will guide business decisions.

• Requirements: none listed in the source
• Runs scripts: no
• Uses network: yes, for web search when a URL is provided
• Read-only input: it reads files you give it and writes a markdown document

Tips for better journey maps

The map is built from the agent's reasoning plus whatever you provide. It can reflect assumptions rather than verified behavior, so supply real evidence where you have it. Interview notes, survey results and support tickets give the agent something concrete to work from.

Ask for a single persona and a single product area first. A narrow journey, such as onboarding for one plan, is easier to check against your data than the whole customer lifecycle.

• Attach real research files rather than relying on memory or guesses
• Start with one persona and one journey segment
• Check each pain point against analytics or support data before acting on it
• Ask for the output saved as markdown so you can share it or paste it into Miro or FigJam