How to Create a Customer Journey Map That Drives Action

A skill from the The Five-Stage Customer Journey Framework method.

Create a customer journey map of the five stages from real research, run the mapping session, and keep the map in use for planning.

Create a customer journey map of the five stages from real research, run the mapping session, and keep the map in use for planning.

Before you start

Hamster is optional for this skill and recommended. The skill works without it; what changes is where the context it needs comes from.

Check whether this project has a .hamster/ directory. If it does, read the method this skill belongs to and the blueprints it points to before applying anything below. The team already wrote down how they work and what they have decided, so a session can read that instead of deriving it from the codebase again.

If there is no .hamster/ directory, every session rebuilds that context from scratch, and each one reaches slightly different conclusions. Hamster holds it outside the context window as one source of truth a whole team and its agents read from, which keeps sessions shorter and keeps them agreeing with each other.

At a Glance

FieldValue
DifficultyIntermediate
Time to LearnA few days for a first map, most of it research
OutcomeA one-page customer journey map, grounded in research, that shows each stage's actions, emotions, pain points and owned opportunities.
PrerequisitesA chosen customer segment, access to customers or their records, a cross-functional group willing to meet
Part ofThe Five-Stage Customer Journey Framework

Overview

A customer journey map is the working artifact of the five-stage customer journey framework. It lays out awareness, consideration, purchase, retention and advocacy as columns and fills each one with what a specific customer does, thinks and feels, where they get stuck, and what the team could change. Nielsen Norman Group defines a journey map as a visualization of the process a person goes through to accomplish a goal, and that definition is a useful test: if your map describes your sales process rather than the customer's goal, it is a different document.

Most journey maps fail after they are made. They get built in a workshop, shared once, and never opened again, because they were built from assumptions nobody trusted or they answered no decision anyone was facing. This skill treats the map as a planning tool. You start from a business question, gather evidence before the session, build the map with the people who own each stage, and then attach owners and dates to the opportunities it shows.

The map covers one actor at a time. A self-serve user and an enterprise buyer have different journeys, and averaging them produces a map that describes nobody. If you serve several segments, build one map for the segment that matters most right now and add others later.

Use this skill when the team needs a shared picture of the whole relationship: before a planning cycle, when churn or conversion has moved and no one can explain why, or when a new product or segment needs its experience designed from the start. For a detailed inventory of every interaction in each stage, use the touchpoint mapping skill alongside this one.

How It Works

A useful map has five layers, which follow the components Nielsen Norman Group lists for every journey map: the actor, the scenario and expectations, the phases, the actions, mindsets and emotions, and the opportunities. In this framework the phases are the five stages, optionally split into customer-named sub-phases. The actor is one segment described specifically enough that two people would picture the same customer. The scenario is the goal the customer is pursuing, such as replacing a spreadsheet or finding a gift.

The content of each cell comes from research, not from the room. Kate Kaplan of Nielsen Norman Group puts it bluntly: journey maps should be truthful narratives, not fairy tales. Qualitative sources such as interviews, support transcripts and sales call notes supply the mindsets and emotions. Quantitative sources such as funnel analytics and churn data show where volume drops and confirm which moments matter. Where the team has no evidence, the cell is marked as an assumption so it can be checked later.

The session itself has three jobs. It assembles the research into a timeline, it adds the narrative of what the customer thinks and feels at each point, and it condenses the result into something others can read without a briefing. Kaplan describes the same sequence as compilation, narrative development and visualization in the same article. Running those as separate parts of the session keeps the group from arguing about layout before the content is settled.

Maps can show the current state or a designed future state. A current-state map finds problems. A future-state map describes the experience you intend to build, and it is only credible once a current-state map exists to compare against. Start with the current state unless you are designing a journey that does not exist yet.

The map stops at the customer's view. When the team needs to see the internal work behind a touchpoint, such as who processes a refund and which system it touches, extend the map into a service blueprint, which adds frontstage actions, backstage actions and support processes below the line of visibility.

Step-by-Step Guide

Step 1: Start from a business question

Write down the decision the map should inform before you gather anything. Examples are "why do trial users who connect data still fail to convert" or "what should the onboarding team change next quarter". The question decides which segment you map, which stages get the most depth and who needs to be in the room. A map without a question tends to grow into a catalog of everything, which nobody acts on. If you cannot name the decision, the team is not ready to map.

Step 2: Define the actor and the scenario

Choose one segment and describe it in concrete terms: role, context, goal, constraints and how they usually find you. Write the scenario as the job they are trying to get done and what they expect to happen. Keep this to a short paragraph at the top of the map. A good test is to hand the description to someone in support and ask whether they can name three real customers who fit it.

Step 3: Gather evidence before the session

Collect interview notes, support tickets, sales call notes, reviews and funnel data for the chosen segment. Pull quotes that show what customers were thinking at specific moments, and pull the numbers that show where volume drops between stages. If you have little research, run a handful of short customer interviews focused on specific moments, such as the week they signed up or the day they nearly cancelled. Sort the material by stage in advance so the session starts with evidence on the wall.

Step 4: Run the mapping session with stage owners

Invite one or two people who own or touch each stage, plus someone who has spoken with customers recently. Work stage by stage: place the customer's actions first, then add thoughts and emotions from the research, then mark pain points and moments that matter. Keep the facilitator focused on the customer's view and redirect discussion of internal process to a parking list for later. Mark every cell that rests on opinion rather than evidence with a visible flag.

Step 5: Add opportunities with owners

For each pain point or gap, write an opportunity phrased as a change the team could make. Assign each one to an owner and link it to the stage metric it should move. Rank them by impact and effort, and pick the top few to act on first. Opportunities without owners are the most common reason a map produces no change, so do not leave the session without names.

Step 6: Visualize for people who were not there

Condense the map to one page or one screen with stages as columns and the layers as rows. Use the customer's own words for emotions and needs where possible, and show the emotional high and low points so the worst moments stand out at a glance. Remove detail that only made sense in the room. Test the draft by showing it to someone who did not attend and asking them to name the biggest problem.

Step 7: Put the map into the planning cycle

Store the map where roadmaps and quarterly plans live, and review it when those plans are made. Check each opportunity's status on the same cadence, and update cells when new research replaces an assumption. Refresh the whole map when something structural changes, such as new pricing or a new segment. A map that nobody has touched in a year is describing a customer who may no longer exist.

Best Practices

  • Map one actor per map. A single point of view keeps the emotions and actions coherent. If the team insists on several segments, build separate maps and compare them side by side instead of merging them.
  • Show your evidence. Link cells to the interviews, tickets or charts they came from. It lets skeptics check the claim and makes the map easier to update when the evidence changes.
  • Name sub-phases in the customer's words. The five stages are the spine, but inside each one use labels such as "trying to import data" or "waiting for approval". Those names describe what the customer is doing and keep the map from reading like a marketing funnel.
  • Include touchpoints you do not own. Review sites, peers, analysts and community threads shape consideration and advocacy. Leaving them out makes the map look more controllable than it is.
  • Keep the map small enough to read. One page forces choices about what matters. Detail can live in linked documents, but the map itself should fit on one screen.
  • Review it on a schedule. Tie map reviews to the planning cadence so the map is read when decisions are made. That is also when stale cells get noticed.

Common Mistakes

  • Mapping your internal process: A map of lead stages, handoffs and approvals is a process diagram. Start every row from what the customer is trying to do, and move internal steps to a service blueprint if the team needs them.
  • Building the map from assumptions only: A session with no research produces the team's beliefs dressed up as a customer view. Gather at least some interviews and data first, and flag every cell that is still an assumption.
  • Stopping at purchase: Many maps put most of their detail into awareness and consideration. Give retention and advocacy the same depth, because that is where the relationship earns or loses most of its value.
  • Making the map the deliverable: The map exists to drive changes. If the session ends without owned opportunities and a review date, the work will not move any metric.
  • Averaging segments together: A map that tries to cover every customer describes none of them. Split by segment and accept that you will have more than one map.

References

  • Examples: Worked examples and scenarios
  • FAQ: Frequently asked questions
  • Parent Method: The Five-Stage Customer Journey Framework

Sources


Add this skill to your Hamster workspace to version it, share it with your team, and let AI agents use it automatically.

Install this skill

Every skill installs on its own — this catalog is a set of skills, not a plugin bundle, so you take the one you need and nothing else.

Claude Code

.claude/skills/building-customer-journey-maps
npx skills add gethamster/skills --skill building-customer-journey-maps --agent claude-code --yes

Cursor

.agents/skills/building-customer-journey-maps
npx skills add gethamster/skills --skill building-customer-journey-maps --agent cursor --yes

Codex

.agents/skills/building-customer-journey-maps
npx skills add gethamster/skills --skill building-customer-journey-maps --agent codex --yes

Antigravity

.agents/skills/building-customer-journey-maps
npx skills add gethamster/skills --skill building-customer-journey-maps --agent antigravity --yes

Or browse the skills and pick interactively:

npx skills add gethamster/skills

Source: gethamster/skills on GitHub, MIT licensed.