What to do when you are asked to “refresh the personas”
(Oh, and btw I know why personas usually don’t work.)I got a call today from yet another person in yet another large company (a super-big one this time) who has been tasked to ‘refresh the personas’ that were created several years ago by collecting new data and creating new posters. Let’s unpack that:
Personas were created several years ago. That usually means a consulting company made big bucks charging to do ‘persona research’ and create these personas that they then ‘delivered’ in some form, probably a presentation or two, a fancy report, and maybe some posters.
The personas are no longer working. No news here…personas either work great and last almost forever for a company, or they fail fast.
People in power in the company still think that personas have got to be a good idea. They know that the organization needs to be more ‘customer-focused,’ that the ‘old’ personas aren’t working ‘anymore’ (they probably never did), and that, in theory, personas are a great way to establish and maintain customer focus.
Some poor dude that’s somewhere in the middle of the great management totem pole has been assigned the job of ‘refreshing the personas.’ He probably has a budget, but relatively little political power. Yes, it could be some poor gal. In this case it’s a dude. So, I’ll use ‘he.’
This well-meaning dude is all for customer focus. He loves the idea. He’s got the money to spend. He starts calling around. He calls trusted colleagues, he calls consultants he knows or has used in the past, he even (in this case) finds the name of a ‘persona expert’ or two and reaches out to them on Linkedin for a quick chat.
He then has a series of calls that fall into two categories:
Category 1: Persona-doin’ agencies: “Yes! We will take your money and we will refresh those personas and it will be awesome and absolutely will work this time for sure! It will be expensive but worth it! We will collect so much data you will freak the f*** out and omg no one will be able to dispute the validity of the personas!”
Category 2: Colleagues and friends: “I get it, but I don’t know what to tell you. In my experience, personas fail all the time and I don’t know why. You’ve been tasked to refresh them, and that’s cool, because the idea of personas is cool. But I don’t have good advice for you because the ones I’ve worked on always fail too.”
So this guy reaches out to me, and we talk for fifteen minutes, and here I am an hour later all revved up again and slapping myself mentally for not writing up and publishing the process I created after years of trial, error, frustration, and victory.Why am I all amped? Because the colleagues and friends are right: most of the time, personas don’t work. And the consultants trying to sell expensive research projects that will create new personas piss me off completely (inevitably, the new personas won’t work either.)
The difference between me and them? I know why.
Ok, here’s a big clue: if you or someone in your organization thinks your personas need to be ‘refreshed,’ they probably never ‘worked’ in the first place. Here are other things that are probably true about your non-working, need-refreshing personas:
They were probably created by some team or an external consulting group and then ‘presented’ when they were done. In other words, they were thrown over the wall as completed entities. This never works well.
They were probably created ‘for’ the entire company, a big bunch of products or services, or a big division in the company, and not for any individual projects or products. This never works well.
They were never explicitly associated with clear, measurable business goals. Which wouldn’t have made sense anyway if either of the two bullets above is true. This makes personas quickly irrelevant and easy to ignore.
The reason all of this is true is easy to summarize, but takes some unpacking: There is no deliverable on earth that can displace closely-held, dearly-believed, fairly basic, not-quite-universally-shared assumptions about who key target users or customers are, and what the company or product should be offering to those users or customers. I italicized ‘assumptions’ because that’s become a dirty word and I don’t think it should be.
The only assumptions that can truly hurt your product are the ones you don’t know about. If you don’t know about them, you can’t see differences between assumptions closely held by key stakeholders and execs. Trust me, they exist. It’s these mismatched, invisible assumptions that kill great experience design.
Unaligned assumptions running amok in the minds of executives are a much bigger problem than lack of data. In the real world of day to day product design, execs and their assumptions have huge impact multiple times a day. Data has impact only if it sinks in and displaces assumptions. And it almost never does. Which means you can pile tons of data onto your execs, present it a gazillion creative ways, create deliverables and posters until the cows come home, and none of it will stick.It doesn’t matter how much data you collect to create your personas if the personas don’t stick.Of course I think data is important — critical even. Of course you need to talk to real users to create user-centered products (at some point during the design and development process). But I think there’s a bigger problem than lack of data in companies and projects within those companies: lack of shared focus and a shared vision of what you are building and why. Data does not help to solve this problem unless you do some work to contextualize any data you collect.
Ad hoc personas, or, as I’ve suddenly decided to call them, Alignment Personas, are created during a collaborative workshop with key stakeholders.
The main goal of the workshop is to get all the assumptions out of brains and into the light, where everyone can see them. Once you can see them, you can align them. And once you realize you need to align them, you realize that they should be aligned according to business goals. Which in turn makes you realize that you don’t have clear, measurable business goals (which, by the way, are much easier to create for individual projects and products than they are for entire companies or divisions).Once you have a nice set of Alignment Personas, you can quickly identify any areas where data would help. You can approach the data collection process as one designed to test hypotheses — and the ‘hypotheses’ are the personas you just created. You can go out there an do research to your heart’s content — do the personas you think exist actually exist? Do they care about what you think they care about? Do they have goals that are similar to the goals you think they have? Does data validate or invalidate the assumptions that went into the personas you just created? Viola. Now you have data-driven personas.
The dirty little secret is that most of the companies I work with never get to the deep data collection/validation part. Why? Because the Alignment Personas we create end up being really obvious (the annoying yet beautiful part of any UX work is that, if you do it right, the result looks like it was obvious from the outset). They also help execs get so much more aligned that a lot of political infighting stops (whether the execs realize it or not) and decisions that everyone agrees on start happening much faster. And results start to happen too: products and projects start looking better, and start producing better results, without any ‘data’ at all. Entire organizations start running themselves cleanly and efficiently towards designs that everyone agrees make more sense given the clear and measurable business goals that were articulated as part of the process.It’s kind of magic.
Oh, and another thing: crisp, clean Alignment Personas tend to be evergreen. They never need much refreshing. And that’s because they illustrate and personify the key DNA of the company: why the company or product was created, which is usually because someone had an idea to do something that would solve some specific set of problems for some specific set of people.Alignment Personas bring that ‘specific set of people and their specific and solvable problems’ out into the light.Anyone who argues I’m not user-centered can have a field day with this, I know. However, I challenge you back: prove to me what I’ve seen isn’t true. Prove to me that politics and assumptions aren’t killing your company or organization’s ability to create great, easy-to-use, problem-solving designs.My own version of Maslov’s pyramid attached to the workplace:
Data isn’t useful — it simply isn’t used — unless it has a place to land.
And that place to land can’t be created with data. It can only be created by creating alignment through a deceptively simple set of exercises to articulate who your execs think you are building things for, and why.
So, what do I do if I’m asked to ‘refresh the personas’ that we already have?
1. Ask why. More specifically, ask “what problem are you trying / wanting to solve with the personas? And why do you think personas will solve said problem?”Keep asking why several times until you get to a specific answer. Examples of non-specific answers that won’t help you:
To get more customer- or user-focused.
To get everyone on the same page.
To get more data into our decision-making.
Follow every one of these with “why?” until you get to something more like this: “I don’t think this specific section of our website/aspect of our product/part of our services really makes sense, and I can’t find a good reason that it is the way it is, but I also can’t get anyone to understand it’s a problem. I’m hoping personas will help me communicate/solve this problem without getting me and everyone around me fired — and it would be even better if the personas could help us solve the problem too.”
2. Ask for specific reasons why the old personas won’t help with this problem, and for specific reasons why the old personas aren’t ‘good’ anymore. You’ll probably hear things like:
The market or data or users or customers have changed, so the personas don’t represent them anymore.
No one pays attention to the personas, so they must be wrong.
I don’t trust that the old personas really do represent the data.
I have no idea how the old personas were created, or what they were created for, so I don’t know that they will apply to the problems I’m trying to solve.
3. Go find out more about the ‘old’ personas. Who created them? For what reason? How did they do it? How much did the persona effort cost? Did they work at all, ever? If so, how, for what project or department? In what specific ways did they ‘work?’ You’ll probably find at least one of the following:
Someone in a position of power read or heard about personas and thought they were a great idea and funded a project — often without super-specific reasons for doing so.
Someone without tons of power did a skunkworks project to create personas and try to drive them ‘up’ in the organization, without real success.
Marketing created personas ‘for’ marketing, or a product team created personas ‘for’ a product or products.
Someone hired an agency to create the personas — usually with a hefty pricetag and as a deliverable after a major data collection and analysis project.
4. Take this information and do an assessment. Is there any reason to think that ‘refreshing’ the personas in some way will really solve the problem that the stakeholder has? Personas are not magic. Fresh data does not make them magical. They are a tool that either works or it doesn’t. A hammer doesn’t work on a screw (just as personas won’t solve a problem that can’t be solved with a persona). And a hammer doesn’t help much if your goal is to ‘build a house’ (just as personas won’t help with a huge, unspecific goal like ‘get more customer-focused).Pull out that old quote: the true definition of insanity is doing the same thing and expecting a different result.
Personas are amazing, killer, hugely impactful tools. They can be built to live almost forever. But they only work if they are created and applied to solve specific, persona-solvable problems. And they only work if they actually make people’s lives easier in the organization — especially the lives of the muckety-mucks.
Therefore, that’s your job: Go figure out why the personas didn’t work before, and why they are wanted now. If you can do that, you’ll be well on your way to a new set of personas created and deployed in a new way to solve a well-defined set of problems.
