User Personas and Journey Mapping
Let me lead with the controversial part: most personas are theater. A laminated card reading "Marketing Mary, 34, drives a Subaru, loves yoga and oat-milk lattes" tells you nothing about what to build, and it gives the team a false sense that they have "done the user work". I have sat through too many kickoffs where a beautiful persona deck stood in for actual understanding. So this note is about the narrow conditions under which these tools are genuinely useful, and the much wider conditions under which they are a waste of a designer's afternoon.
#What they are
Personas are composite, archetypal users built from research, a named and humanized stand-in for a segment, capturing goals, contexts, behaviors, and pain points. Journey maps chart a persona's end-to-end experience accomplishing something: the stages they move through, what they do and feel at each, and where the experience breaks. Done right, both are artifacts that carry research into the team's daily decisions, the "comprehension" half of Voice of the Customer.
#When personas help vs become theater
A persona is only worth making if it would change a product decision. Here is the honest test. Cover the name and the photo, and ask whether anything left on the card changes what you would build. If it is just demographics and hobbies, no. If it is goals, contexts, and behaviors that distinguish how this person uses the product, yes. Demographics are usually decoration. Behavior and goals are the signal.
| Personas help when… | Personas are theater when… |
|---|---|
| Built from real research, updated as you learn | Invented in a workshop from assumptions |
| Centered on goals, contexts, behaviors | Centered on demographics, hobbies, a stock photo |
| Used to settle "who is this for" in real debates | Made for a deck, then never opened again |
| Few, sharp, and distinct | A dozen overlapping near-identical cards |
Here is my rule. Build the fewest personas that capture behaviorally distinct ways people use the product, ground every one in research, and kill any you cannot tie to a decision. A persona built on assumptions does not reduce risk. It encodes your bias and then lends it the authority of a printed artifact, which is worse than having no persona at all.
#Where I find journey maps more useful than personas
In my experience journey maps earn their keep more reliably than personas do, because the act of mapping is itself diagnostic. Laying out the real end-to-end flow, including the parts before and after your product touches the user, surfaces the gaps, the hand-off cracks, and the moments of friction that a feature-by-feature view hides. The map points your funnel at where drop-off happens and your research at why. The value is in the making, not in the poster on the wall.
Two things separate a useful journey map from a decorative one:
- It is grounded in observed behavior, not in how you wish the flow went. Map what users actually do, warts and workarounds and abandonment, not the happy path.
- It crosses the edges of your product. The most valuable insights live in what the user does before they arrive and after they leave, the parts you do not control but that shape whether your bit works.
#How these relate to JTBD
I get asked whether personas compete with Jobs To Be Done. They do not, because they answer different questions. A persona describes who is doing the job, the job describes what progress they want, and the journey map describes how the experience of getting there unfolds. I use jobs for strategy and positioning, personas for grounding day-to-day design empathy, and journeys for finding where the experience breaks. The mistake is treating any one of them as the whole picture.
#How AI is changing it
AI will happily generate a full persona set and a journey map from a one-line prompt, and that is exactly the danger, because it produces plausible fiction at scale, the laminated-card problem industrialized. Used well, it does the opposite: it synthesizes real interviews and support data into evidence-backed personas, and drafts a journey map you then validate against actual behavior. The discriminator is whether the artifact is grounded in research or hallucinated from priors, and that judgment is yours to make. See AI in Discovery and Research.
#Continue Reading
- Voice of the Customer for the competency these artifacts serve.
- Jobs To Be Done for the complementary lens: who vs what vs how.
- User Research and Discovery for the research that grounds a real persona.
- Product Analytics and Metrics for where a journey map points your funnel analysis.
- User Experience Design for how journeys feed design decisions.