Personas Are in Crisis

The persona is having a moment. Two recent pieces, one from Forrester and one from SwiftERM, have declared them outdated. On a recent episode of the Experience Strategy Podcast, Dave Norton, Joe Pine, and Aransas Savas took a closer look at that argument and arrived at a more nuanced, more useful conclusion.

The verdict? The persona isn't the problem. The way most companies are creating and using personas is. 

Where Personas Came From and What They Became

Personas didn't start as a marketing tool. They emerged around 1999–2001 as a design thinking technique, invented to keep engineers honest. Left to their own devices, product builders tend to design for themselves. Personas put a real human in the room, someone with different needs, constraints, and contexts, so the team couldn't forget who they were actually building for. 

That original purpose was powerful. Then personas migrated into marketing, merged with segmentation research, and something important got lost. Instead of keeping the customer's lived experience central, personas became proxies for market size. The question shifted from "who is this person and what do they actually need?" to "how many people like this are in our market?" 

The result is what we call segment personas: broad demographic and attitudinal composites that can describe millions of people but predict the behavior of almost none of them in any specific moment. "John, the upskilling family man," tells you something about a slice of your market. It tells you almost nothing about what John needs when he's standing at the pharmacy counter at 9 pm with a sick kid and an unfilled prescription. 

The Problem Starts Before You Ever Name a Persona

Most discussions about personas focus on how teams use them. But the limitations often begin much earlier, in how personas are built and what data shapes them. 

When personas are constructed primarily from demographic surveys and broad attitudinal research, they reflect averages. And averages, by definition, flatten the very variation that matters most in experience design. You end up with a persona that is technically grounded in data but practically disconnected from the range of real people it claims to represent. 

There is also a representation problem. Building a persona around a single archetype, one face, one story, one composite individual, creates unintended biases in how teams design and communicate. It can cause diverse swaths of your actual customer base that fit within the persona to become invisible. The people who don't look or sound like the archetypal persona are underserved, not by intention but by omission. 

Richer personas show a range of people who share similar jobs to be done, including the inner circles that influence their decisions: partners, children, colleagues, and caregivers. They incorporate underlying beliefs, information gaps, the data that actually informs customer decisions, and the specific life situations in which needs shift. The difference between a traditional segment persona and a well-built next-generation persona isn't just aesthetic. It's the difference between knowing who your customer is in aggregate and understanding what they need right now. 

The Frozen Customer Problem

This brings us to the deeper issue the podcast surfaced, and one that AI is making impossible to ignore: customers are not stable. 

Joe Pine put it this way: the progression of markets has moved from mass market to segments to niches to markets of one, and the next frontier is markets within one. Joe, on a business trip to Chicago, is a completely different market than Joe on vacation in France with his wife, even though it is the same person and the same credit card. Same values, same demographics, same persona category, entirely different needs. 

SwiftERM captured this well: while your team is building a persona for a particular customer profile, that customer has already moved on. They bought a yoga mat six months ago. For the last three days, their behavior shows interest in supplements and strength training. The persona captured who they were. It missed who they are right now. 

This is not an argument for abandoning personas. It is an argument for not asking them to do work they were never designed to do. 

What's Actually Missing: Situations (and a Word on Moments)

The most durable concept from the podcast is one Stone Mantel has been building toward for years: situations as a stable unit of design. 

It is worth pausing on language here. "Moments" is an intuitive word, and the right instinct. But 20 years of design thinking have loaded it with a particular pattern: retail moment one, retail moment two, retail moment three. That framing is discrete and product-out, not organic and customer-out. Situations carry the same meaning without the accumulated baggage, and they keep the focus where it belongs: on the customer's context, not the company's funnel. 

A situation, a Friday night with your family, a Monday morning before a big meeting, a hot stretch of summer weather, a prescription that needs filling before a weekend trip, is reliable in a way that demographic profiles aren't. People's behavior changes dramatically based on context, even when the person doesn't change at all. But the situations themselves recur. They can be studied, mapped, and designed for. 

Dave's research offers a concrete illustration: weather dramatically changed bathroom usage patterns at French gas stations. At Chicago train stations, the same variable barely moved the needle. Same behavior category, different situational market entirely. 

The design implication is significant. Instead of asking "what does our segment persona prefer?" the better question becomes: what situations does our customer find themselves in, and what do they actually need in each one? The more situations you can identify and solve for, the more value you create for customers and for your business. 

Situations Are a Key Missing Piece, Not the Only One

Situations are critical, but they don't work in isolation. The full upgrade to persona thinking involves layering in another dimension: modes. 

Customers are not just in different situations; they are in different states within those situations. Someone in planning mode needs entirely different support than the same person in urgent mode or indulgence mode. Modes are temporary but powerful. They shape what a customer is receptive to, what they want to accomplish, and how they will evaluate whether an experience delivered value. 

Building modes into your personas, alongside situational thinking, dramatically sharpens how you design for customers. It moves teams from describing who the customer is to anticipating what they need next. And it makes your personas actionable in a way that demographic composites never can be. 

The Opportunity

AI doesn't kill personas. It raises the stakes for using them well. When AI can update its understanding of a customer in real time, responding to what they just did, what they are about to do, and what the context suggests, a static demographic profile becomes a floor, not a ceiling. Companies that upgrade their personas to include situational thinking, mode-awareness, and a genuine range of real people will be the ones whose experiences actually feel like they understand the customer. 

The persona was always meant to keep the customer in the room. The next evolution is keeping the customer in the moment, recognizing that the same person has multiple needs, multiple contexts, and deserves more than a single story told about them. 

If this is a challenge your team is wrestling with, whether that's how to build next-generation personas, how to structure situational research, or how to translate all of this into actionable experience design, we would love to talk.

Click to schedule a time to chat.