Costly Mistakes in Persona Creation: Marketing vs. UX Personas

The first personas I created were utter crap.

Okay, okay, they weren’t utter crap, but they could have been a much better tool if they had been created with quality research.

During my early years of UX design, I was leading UX while learning UX, so there are a number of things I did that I now see as limited and/or flawed. My first attempts at persona creation fall among those.

I created assumption-based marketing personas even though I was telling myself that they included research and I could do the work later to make them more research-based. The time never came later.


What are personas anyway?

There are dozens of countering definitions out there. Some that apply to marketing, some that apply to UX, some that apply to neither. So I chose a definition that involves research because ideally personas should involve research. When done well, personas are:

“A way to model, summarize and communicate research about people who have been observed or researched in some way. A persona is depicted as a specific person but is not a real individual; rather, it is synthesized from observations of many people.”–Shlomo Goltz, Smashing Magazine (A Closer Look At Personas: What They Are and How They Work: Part 1)

Businesses often think about trying to speak to everyone, to be inclusive, and making sure they aren’t alienating people, so how does that work with narrowing the focus with something like a persona?

“To create a product that must satisfy a broad audience of users, logic will tell you to make it as broad in its functionality as possible to accommodate the most people. When you design for your primary persona, you end up delighting your primary persona and satisfying your secondary persona(s). If you design for everyone, you delight no one. That is the recipe for a mediocre product.”–Eeva Llama, UXBooth (Creating Personas)

Good personas work by providing focus, establishing empathy with the end user, and overall increasing the quality of communication.


The flaws of my first personas.

Here’s what was flawed in the research of my early persona-creation:

  • The research conducted included focus groups, interviews, and stakeholder knowledge of our users. Both the focus groups and stakeholder knowledge were flawed methods of gathering research as focus groups can create group-think and stakeholder knowledge is based on memory/limited knowledge and not on the evidence at hand.
  • For the generative interviews, we had too small of a sample size for each user type, limiting the ability to find meaningful patterns. We only were able to talk to 1–2 users from a given user type in some cases.
  • Our generative interview questions were focused on uncovering issues with the brand and people’s perceptions of it. So they lended themselves to persona creation, but weren’t necessarily the ideal way of gathering full insights.
  • I was creating the personas, but wasn’t present for the interviews. Plus, we were discussing sensitive subject matter that could affect people’s jobs and wanted honest feedback from the interviews. This led us to not record any sessions. So I was relying on what I was being told about those interviews instead of being able to look at the findings myself.

So let me walk you through the process I used to create my first personas.

First, I read a ton of articles about how to create personas, but didn’t realize that there was a difference between marketing and UX personas. (I’ll explain the difference later.)

The articles I read framed persona creation around:

  • demographic information
  • buying motivations and preferences
  • concerns
  • marketing messages
  • media habits

The Persona Activities

These articles recommended finding photos that matched the demographic information of the market segments and gathering a small group of people together to craft the story of an individual pictured. So that’s what I did.

I led six different persona activities with departments from around the organization. Each of these meetings included 6–12 people, so we ended up running this activity with roughly 50 stakeholders. Each department focused on a different user related to their understanding of a specific user type. (i.e. the Alumni & Development department understood the donors and grads better than any other area of the organization, so we had them focus their efforts there). That was at least my logic for justifying why we did things this way.

I started each session with an overview of what personas are and answering questions. I primed people to think in terms of the pareto principle (also known as the 80/20 rule) and explained the difference between personas and stereotypes. Explaining that personas are different from stereotypes because while stereotypes are typically negative in nature, personas are focused on positive attributes that build empathy with our end users. While the empathy part is great, I was fooling myself for thinking that assumptive-personas were much better than stereotypes.

Everyone broke up into groups of two to three and grabbed a photo of someone they thought exemplified a user group they were very familiar with based on their job. Then they started filling in the demographic info., bio/backstory, goals, hobbies & interests, challenges, biggest fears, hesitations with using our company’s product/service, and feelings about our company.

After each group created a persona, we shared the results back with the whole group for feedback. That allowed us to see overlap and narrow down who we thought those 20% of users were. Then we did another round of persona creation. The second round typically proved much more useful as people were now primed to think this way and could do a better job of honing in. However, the second round also included making personas of people that we didn’t have as current users. These were people that the organization was hoping to target in the future. Risky business.

Making Sense of What We Created

By the time all of these persona activities were complete, we ended up with dozens and dozens of personas that we needed to sort through. So I created a small group of people that represented different areas of the organization and we found patterns.

We grouped the personas into user types. Then we charted out which personas tied into business for various aspects of the organization. That helped us focus on who the most important users were and which persona categories adequately filled needs throughout the organization.

Although not ideal, with robust offerings of the organization I was creating these for, we ended up with 7 personas to represent the organization. These 7 personas were met with various degrees of acceptance. At the time, I thought those that weren’t accepting them were being closed minded, but now I realize that they may not have been an effective tool outside of marketing efforts and those opposing them were doing so for good reason.


Marketing personas aren’t all doom & gloom.

So although there are limitations with marketing personas that I’ll go into in a minute, here is what was successful from this persona creation:

  • Stakeholder buy-in for personas as a tool
  • More buy-in for the larger branding initiative that I was leading for the organization
  • Focused internal thinking around users & created alignment (some focus is still better than no focus)
  • Deepened empathy and understanding of users
  • Started & continued the conversation of users
  • Shifted marketing to think about specific users instead of just demographic information

So how could this be done better?

  • Create separate personas for marketing and for UX.
  • Use research to craft a persona from the get-go
  • Conduct the research personally or have the person who did conduct the research present for persona-creation
  • Establish research goals beforehand that meet the needs of how the research will be used short-term and long-term
  • Find patterns in the research to inform persona creation — use real user behaviors, motivations, goals, quotes, etc.
  • Condense user types down if they share the same goals
  • Simplify the persona information down as much as possible to present it in digestible chunks

Since this method of persona creation used flawed research, I’ll call it an assumptive-based marketing persona method. If these personas truly would have been research-based, we would have been able to deepen the quality and usefulness of these personas as a tool. I was only fooling myself when I thought I could turn an assumptive-based persona into a research-based persona later on. I’d already created the story for this assumptive persona so I was already attached and would be looking for ways to match the research to the persona I’d already created.

“People are biased to interpret the evidence in ways that are consistent with their desires. That means that people may ultimately come to believe that the weight of evidence supports the position that they already wanted to believe was true. And they will believe this without recognizing that their own desires influenced the evaluation of the evidence.” –Art Markman Ph.D., Psychology Today (You End Up Believing What You Want to Believe)

By already creating the user story, it was going to be nearly impossible for me to break my bias towards what I already believed to be true. Starting with research first helps to not form attachments or biases towards an outcome before we begin. This is another reminder for us to not start solutioning too early in the UX process.

Also, thinking that these personas could work for all areas of the business (both for UX and marketing) was naive. I now realize that it is OK to have marketing personas and UX personas and that they are very different from each other. Or perhaps there is a way to create a marketing & UX persona hybrid…


So what is the difference between marketing & UX personas?

Let’s start by defining UX personas (also know as design personas):

“Design personas focus on user goals, current behavior, and pain points as opposed to their buying or media preferences and behaviors. They are based on field research and real people. They tell a story and describe why people do what they do in attempt to help everyone involved in designing and building a product or service understand, relate to, and remember the end user throughout the entire product development process.”–Eeva Llama, UXBooth (Creating Personas)

Did you catch the parts about field research, real people, telling a story, and creating a product or service?

Now let’s look at Marketing Personas:

“Marketing personas focus on demographic information, buying motivations and concerns, shopping or buying preferences, marketing message, media habits and such. They are typically described as a range (e.g., 30–45 years old, live in USA or Canada), and explain customer behavior but do not get to the why behind it. Marketing personas are good for determining what types of customers will be receptive to certain products or messages, or for evaluating potential ROI of a product. What they are not good for is for defining a product or service — what it is, how it will work, and how it will be used; or for prioritizing features in a product or service.”–Eeva Llama, UXBooth (Creating Personas)

So essentially the reasons for creating UX and marketing personas are different and therefore, so too are their outcomes. UX normally has a more specific focus on creating a product and is used to mitigate risk in the design process through deep understanding of the needs of a user. Marketing Personas are essentially used as tool for the business by helping them identify where to invest marketing efforts and how to communicate with prospective customers.

Both types of personas can be great tools, but it’s important to know the difference so you can make the best decision about which option is best for you, your organization, and the challenge at hand.