011 min
Why teams that each do well can still lose people
Picture a design lead at a subscription app. This is an illustrative case. Three teams report good results. The signup screens test well, the welcome emails get opened, and the support team answers tickets within a day. Even so, many new subscribers stop using the app in their first week.
No single team can explain it, because each team sees one piece of the experience. A new subscriber pays on the website, waits for an email, opens the app on a phone, and cannot find the feature the website promised. No team is responsible for that whole sequence.
The problem is real. Researchers at the Norwegian research institute SINTEF and the telecom company Telenor studied how companies handle customer journeys. None of the companies they studied had a complete overview of the journeys their own customers took.
Journey mapping is the method built to look at that whole sequence at once, from the point of view of one person.
Gibbons names two benefits. Building the map makes people from different teams talk, and it leaves them with one shared picture of the experience. The finished map then explains the experience to everyone involved and gives the team a basis for decisions.
023 min
What a journey map contains
A journey map answers that problem by putting the whole sequence on one page, so a missing owner or a broken handoff between teams becomes visible. Sarah Gibbons of the Nielsen Norman Group wrote in 2018 that journey maps of any shape share five parts.
| Part | What it holds | In the subscription app |
|---|---|---|
| Actor | The one kind of person whose view the map takes | A new subscriber who joined to track their runs |
| Scenario and expectations | The goal, and what the actor expects to happen | Start using the app on the day of payment |
| Phases | The large stages that organise everything else | Choose a plan, pay, set up, first week |
| Actions, mindsets, emotions | What the actor does, asks and feels in each phase | Searches the phone for the feature the website promised, and feels confused |
| Opportunities | What the team learned, who owns the fix, and how to measure it | A clear first-run screen, owned by the onboarding team |
The actor is the one kind of person whose view the map takes. Gibbons gives a university as an example. A map about a student and a map about a faculty member would look different, so the university needs two maps. Actors usually match personas, which are written profiles of typical users, when a team has them.
The scenario is the situation the map covers, with the goal the actor has and what they expect. Maps suit scenarios that run through a sequence of events, such as shopping or taking a trip. They also suit scenarios that cross several channels. A new-patient journey at a medical practice is one case: it can run for days, weeks or months, and it touches the practice website, the phone, email, the office and a patient portal. The same method can map the journey of an employee or a patient, not only a customer.
Phases change from one scenario to the next. For buying speakers online, Gibbons suggests discover, try, buy, use and seek support. For buying a car, the phases are engagement, education, research, evaluation and justification.
Under each phase the map records what the actor does, what they ask themselves, and how they feel. Actions are a short account of the phase, not a log of every click. Mindsets work best as the person's own words from research. Feelings are drawn as one line that rises and falls across the phases, so the reader sees at once where the person is pleased and where they are frustrated.
Touchpoints and channels
Kate Kaplan of the same group explained in 2016 that a touchpoint is a moment when the actor deals with the company. A channel is the way that contact reaches them, such as a website or a physical store. She gives both special weight because they are where inconsistent messages and disconnected steps tend to appear. In the subscription app, the payment page, the receipt email and the first screen of the app are three touchpoints on three channels.
Pain points and opportunities
A pain point is a step where the actor is blocked, slowed or upset. Most pain points are friction, meaning extra effort or delay, that the people who build the product no longer notice. The last part of the map turns pain points into opportunities: what to change, who owns the change, and how the team will measure it. A map with no named owner for each opportunity tends to produce a good discussion and no change.
031 min
Where journey mapping comes from
Journey mapping developed alongside service design, and the earliest dated related method that sources name is the service blueprint. In 1984, the bank executive G. Lynn Shostack published the first description of a service blueprint, in the Harvard Business Review. A blueprint lays out a service as a sequence of steps, from the customer's side, with the staff work behind each step.
Gibbons treats the blueprint as an extension of the journey map. The map follows the customer. The blueprint adds the business's side: the people and processes that support each step.
Fail points and bottlenecks
Later writers made blueprints better at finding trouble. Zeithaml, Bitner and Gremler recommended in 2006 that a blueprint mark fail points and bottlenecks. A fail point is a step where something could go wrong and lower the customer's satisfaction. A bottleneck is a step where customers are likely to wait longer than they will accept.
Journey maps use the same idea when they mark pain points. The blueprint adds more detail, because it ties each mark to the staff work and the waiting time behind it.
042 min
How to build a journey map
A team builds a map in a fixed order, because each step needs the one before it.
- Choose one actor and one scenario, and write down the goal the actor has.
- Write the start and end points: the situation the person is in at the start, and what has changed for them at the end.
- Collect research about the real journey. User interviews and diary studies, in which people record their own activity over days, suit journeys that last days or weeks. Analytics and support tickets show where people drop out or ask for help.
- List the steps in order, online and offline, and group them into phases.
- Add what the person thinks and feels in each phase, in their own words.
- Mark the pain points. Then write each opportunity with an owner and a measure, and rank the opportunities by how much they help the person and how hard they are to build.
Do the work with people from different teams. Kaplan says the act of mapping is often worth more than the finished picture, because it puts the owners of different steps in one room. Finish the analysis before you design the picture, since an attractive graphic can hide a weak analysis.
A team can draw three kinds of map. A current-state map shows how the journey works today. A future-state map shows a journey the team plans to build. A day-in-the-life map follows the actor through a whole day, including time spent away from the product.
A user flow shows one step in more detail
Some steps of a journey happen inside one product, such as signing up for alerts. A user flow is a diagram of the interactions needed to finish one task in one product. The team can draw a user flow for that step and keep the journey map for the larger picture. A user flow works as a closer look at one area of the high-level journey, so the journey supplies the context and the flow supplies the detail.
051 min
A government service mapped on a wall
The steps above work with simple tools. The teams that publish GOV.UK, the United Kingdom's government website, use a wall and printed pages. In a 2019 post, Katherine Dunn, Matt Clear and Gabrielle Acosta describe four stages for improving a service's content: scoping, auditing, journey mapping and content mapping.
Their example is a service for people who want to marry in the United Kingdom but do not live there. The start point is the wish to hold the wedding there. The end point is being married there.
The team puts the start and end points on a wall, writes the tasks between them in order, and sticks a printed page next to each task it supports. Online and offline tasks go on the same wall. An application is a typical case. Its tasks are to download the form, print it and fill it in, send it, and receive confirmation.
Then the team looks at what the wall shows. A task with no page is a content gap. A task with two pages that say the same thing is a duplicate. A page that serves several tasks should be split so that each page covers one task. For duplicates, the team's advice is to agree which page stays and take the others off the site.
062 min
A shopping mall that counted who met each touchpoint
A map can also go wrong in the other direction, by treating every touchpoint as if every customer meets it. Mark Rosenbaum, Mauricio Losada Otalora and Germán Contreras Ramírez tested this in 2017 on one of the world's largest indoor shopping malls.
Which touchpoints shoppers actually meet
Mall managers first listed every point where a shopper could meet the mall. They found 60 touchpoints. The researchers handed questionnaires to 100 shoppers in different parts of the mall. Each shopper said which of the 60 they had met that day.
The percentages show how many shoppers said they had met each touchpoint or planned to. The ten most common touchpoints did not reach the same share of shoppers. Reach ran from 95 percent for recommending a store to 46 percent for hearing comments from other shoppers.
The managers then acted on the touchpoints that most shoppers met. Because 89 percent had visited the mall before, it placed customer ambassadors at each entrance to collect feedback. Because 74 percent had seen advertising on a highway billboard, it replaced its signs with three-dimensional boards that have moving parts.
Rows that name who acts
The authors also gave each touchpoint a set of rows. The rows cover what the shopper needs, what employees must do, what employees need in order to do it, how the mall is designed, and how departments can together improve the touchpoint. The authors argue that a map's rows should show which departments, from marketing to operations and information technology, contribute at every touchpoint. That is the same link to service blueprints that appears above.
072 min
Why the emotion line matters on a journey map
The emotion line is easy to treat as decoration, but memory gives it a job.
How endings shape memory
People do not remember an experience as an average of its moments. In 2003, Donald Redelmeier, Joel Katz and Daniel Kahneman tested this on 682 patients who were having a colonoscopy, a medical examination of the colon. Half of the patients, chosen at random, had a short extra interval added at the end of the procedure, with the instrument left in place. Patients who had the longer procedure rated its final moments as less painful. They also rated the whole experience as less unpleasant, and they were slightly more likely to come back for another colonoscopy.
The setting is medicine, not product design, so treat the result as a reason to look closely at how a journey ends, not as a rule. On a map, it suggests two places to check first: the lowest point of the emotion line and the last step. Teams call the moments that matter most to a person's decision the moments of truth, and those are the first places to look. This is the idea behind the peak-end rule, which says people judge an experience mostly by its most intense moment and by its end. In the subscription app, a smooth payment followed by a confusing confirmation email may be remembered as worse than the payment deserved.
A map made only of feelings
Some teams respond by drawing only the emotion line. Rosenbaum and colleagues warn that a map built only on feelings becomes an empathy-mapping exercise, which tells managers little about how the service system works. An empathy map is a diagram of what one person thinks and feels. It is a useful tool, but it answers a different question. Keep the emotion line, and keep the touchpoints and the departments in the picture as well.
082 min
When journey mapping misleads
A finished map can mislead in four ways, and each has a cause that is easy to check.
A map drawn from beliefs
Kaplan's rule is that a map should be a truthful account, built from research. A map drawn only from what the team believes records the team's views, and it can look as convincing as a researched one. A related mistake is mapping the company's internal process instead of the person's goal. The steps then describe what the company does, not what the person experiences.
Real journeys differ from the plan
The SINTEF researchers report that most customers meet deviations from the planned journey. In their 2016 paper, Ragnhild Halvorsrud, Knut Kvale and Asbjørn Følstad name four kinds of deviation: extra touchpoints nobody planned, touchpoints in an illogical order, touchpoints that fail, and touchpoints that are missing. They found these by combining interviews, diary studies and records of what customers did.
A map of the planned path is a starting guess. To find the real deviations, compare it with what customers actually did.
People loop back after buying
Some maps run in a line from first contact to purchase. Court, Elzinga, Mulder and Vetvik of McKinsey argued in 2009 that real buying is less linear. Their model is circular. A trigger starts the process. People first consider some brands, and then they evaluate actively. During active evaluation the number of brands a person considers can go up rather than down, and the experience after buying shapes the next decision. They based this on a survey of 20,000 people across three continents. For a map, this means the last phase may need to lead back to the first instead of stopping.
A map that nobody updates
A map shows one moment in time. When the product, the channels or the customers change, the map stops matching what happens. Kaplan advises keeping the map in use, for example by bringing it up in meetings, so people notice when it no longer fits.
091 min
Journey mapping compared with nearby methods
Several other diagrams look like a journey map, and teams often use the names as if they meant the same thing.
| Method | Whose view it takes | What it is for |
|---|---|---|
| Journey map | One actor in one scenario | Understanding the whole experience across steps and channels |
| Experience map | A general person in a general experience, not tied to one business | Understanding a human behaviour before choosing a product |
| Service blueprint | The business, laid over the customer's steps | Finding which people and processes support each step |
| Process map | The company | Showing internal steps in order |
| User flow | One user inside one product | Designing the screens and responses for one task |
| User-story map | The team planning a release | Breaking a feature into pieces for release |
Gibbons separates the last row from the first by purpose. Journey maps are for discovery and understanding, and user-story maps are for planning and building. She also advises against arguing over the words. Agreeing on what goes inside the map matters more than agreeing on its name.
?5 questions
Questions people ask
Is a customer journey map the same as a user journey map?
Do I need special software to make a journey map?
Should a journey map show emotions?
What is the difference between a touchpoint and a phase?
How often should a team update a journey map?
§10 sources
Sources
Sarah Gibbons, "Journey Mapping 101", Nielsen Norman Group, 2018:
Kate Kaplan, "When and How to Create Customer Journey Maps", Nielsen Norman Group, 2016:
Kate Kaplan, "User Journeys vs. User Flows", Nielsen Norman Group, 2023:
"Service blueprint", an encyclopedia entry that summarises Shostack (1984) and later work:
Show all 10 sourcesShow fewer sources
Mark Rosenbaum, Mauricio Losada Otalora and Germán Contreras Ramírez, "How to create a realistic customer journey map", Business Horizons 60(1), 2017:
Katherine Dunn, Matt Clear and Gabrielle Acosta, "Improving content through journey mapping", Inside GOV.UK, 2019:
Robert Langkjaer-Bain, "Purchase behaviour in need of a rethink, says McKinsey", Research Live, 2009, a news report on the McKinsey paper "The consumer decision journey":
Donald Redelmeier, Joel Katz and Daniel Kahneman, "Memories of colonoscopy: a randomized trial", Pain 104, 2003:
Christina Benjaminsen, "Many mistakes made on Customer Journeys", Norwegian SciTech News, 2017, an interview with a SINTEF researcher:
Ragnhild Halvorsrud, Knut Kvale and Asbjørn Følstad, "Improving service quality through customer journey analysis", Journal of Service Theory and Practice 26(6), 2016:
