Related methods
User journey
The user journey breaks a process down into steps to surface user needs and expectations for a given goal.
Read more about User journeyPersonas
UX deliverable
Personas to model behavioural archetypes of your target users
A format created by Alan Cooper in 1999, personas (or "personae") now hold a major place in UX. They're especially used to support decision-making and to pick which new ideas to pursue. A persona is a stereotyped representation of the target user group, used as a reference when expressing needs and understanding users. In marketing, personas are usually built around traits tied to buying behaviour. UX personas focus on behaviours around usage. From the data gathered from end users, you'll work to identify the behavioural variables that segment your target users. Four or five major dimensions will emerge, and they'll allow the definition of 4 to 6 main personas. These personas will stand as stereotypes of the users of your products or services. You'll accept the idea that personas are stereotypical profiles and, as such, can only represent a marginal slice of the user base. But the way their traits combine, having come out of rigorous behavioural research, gives designers the means to think through the user journey while constantly keeping the specifics that stand out in mind.
What are personas for?
- Simplifying complexity by offering a limited number of personas that reflect your users' needs
- Supporting empathy in the project teams through stereotypes
- Aligning teams on the primary and secondary behavioural traits
- Making informed decisions thanks to user knowledge
- Testing new ideas in relation to the needs gathered
Why build personas?
- Putting users' needs and expectations at the centre of design
- Simplifying complexity by offering a limited number of archetypes that reflect your target users' needs
- Synthesising usage data and user knowledge
- Personas are a guiding thread for project teams as they design products or services
How to define UX personas
01
User data
To build personas, you first need to gather behavioural data. Ideally, that data comes directly from real users. It can come through observations, individual interviews, focus groups or diary studies.
02
Behavioural variables
The data analysis should bring out behavioural variables. You can also talk about axes or dimensions along which the users you met seem to distribute. Generally, 4 to 6 dimensions are enough.
03
Behavioural archetypes (patterns)
Once users are placed on those axes, certain patterns start to take shape. That's what we call the behavioural archetypes. You then pick the most telling ones, and crucially validate them with subject-matter experts. This iteration helps target the most interesting ones to work on, and also adjust or even drop some archetypes.
04
Generating the user profiles
Once the archetypes are defined, all that's left is to enrich them with other information that fleshes out the profiles. You usually include expectations and frustrations, along with any other piece of information relevant to the product or service at hand.
En pratique
The usability and design of the final output are key characteristics of its quality. Far from being an expert’s flourish, these materials need to be usable in workshops, with profiles that don’t necessarily share the same sensitivity to usage as the user researchers.
The document needs to convey the most important information quickly and simply, and make it easy to remember. It also needs to support creativity. Its visual character should energise the teams gathering around it, make them want to consult it, even enjoy it.
This modelling work can also be carried out without the user-needs collection step. That situation comes up when resources (time and budget) are constrained, but the stakes around accounting for users justify the stereotypes being designed to reflect the (necessarily slightly biased) view of the business and the experts supporting the work. To set them apart from materials built on actual user data, these deliverables are called proto-personas.
Frequently asked questions
Yes, but they need to be set apart from personas built on actual user data: these are called proto-personas. That situation comes up when resources (time and budget) are constrained, but the stakes around accounting for users justify the stereotypes being designed to reflect the (necessarily slightly biased) view of the business and the experts supporting the work.
No, and that is intentional. Every real customer is a mix of several profiles. Personas are archetypes used to settle choices, not to classify individuals. Good practice is to cross-reference several personas to enrich the analysis.
The recommendation is to define 5 to 7 personas, no more. That is a deliberate methodological constraint. Beyond 7 personas, they become hard to remember and the teams' ownership of them suffers. The aim is to keep a workable tool, not an exhaustive one. Are personas there to generate ideas or to evaluate existing solutions?UsageMainly to evaluate and to challenge. The tool works better when it serves to test an idea or a solution that has already been formulated than when starting from scratch in ideation mode. So the problem needs to be framed ahead of the exercise. When can personas be updated?EvolutionAn annual review is recommended, for instance during a retrospective. The suggested method: list what worked, what did not work and what was missing, then recalculate the key indicators if needed. How do you keep personas alive over time? Personas are meant to evolve. A first quick round of feedback after rollout helps identify the immediate adjustments. A longer-term retrospective then makes it possible to assess how useful they really are, to spot the missing data and to understand how the teams have taken ownership of them.
A few of our case studies
SNCF and territories: rethinking the last mile in sustainable mobility
View the case study : SNCF and territories: rethinking the last mile in sustainable mobilityCurious to learn more about us?
Everyone talks about user experience, service design, ergonomics...
It's not always clear, and you'd like to find out, or sharpen what you already know.