Hey, Megan, let's do a podcast.

Great idea. What should we talk about?

So I know that one of the most popular parts of the LLAMA workshops that you run on a regular basis is the portion on Personas. So I thought maybe we could talk about that today. But first things first. What is a Persona?

My definition of this has evolved over time as I figured out how to make this work for folks. What I'm currently talking about a Persona as a archetype or a representative of an entire segment of learners. And they're kind of a stand in so that you could talk about an entire segment that shares a lot of characteristics by talking about this one person archetype, the word sounds really close to stereotype. That gets people very, very uncomfortable, and understandably so. But this archetype is really helpful and we use this as a way of getting close to learners. But because we give them a name and some characteristics and we spend time talking about it, it seems like a fantastic way to connect with the humans of the learners rather than just have them be out there invisible while we're designing.

Cool. I know I've been in some Persona identification sessions in the past with different clients, and in some of those, the group really wanted to say, well, on average, we have some learners that do, or on average, the age is whatever. But that's really something you want to stay away from when you're doing learner Personas, right?

Yeah, totally. So here's the. It's so easy to say to talk about our learners in terms of averages or ranges, right? They could be 18 to 65 years old. They, some of them have GEDs and some of them have PhDs, and the average of that isn't necessarily a meaningful construct. When we can create Personas, a number of Personas, we then can see them as segments with differing needs. And it might be in terms of needs relative to their career or their prior experience, or how they're going to interact with a particular learning experience. And that ability to break down that average is really important. I mean, I talk to people, like sometimes they'll say, well, there's this many men and this many women.

It's. Well, the average of that is not meaningful. And they start to kind of get it.

So how many Personas would you make for a particular project?

It depends. It depends on a couple of things. One is the criticality of the project. Right. A higher stakes project. We're going to dig in a little bit deeper and maybe make a few more Personas. I generally find, though, we end up Creating anywhere from two, which is usually not enough, to five to sometimes eight or more Personas. And what I'm aiming for is to have Personas, to have the groups that they represent comprise about 80% of the audience.

And so often in that 3 to 5 range, I've got about 80% of the audience. And I use a lot of questioning. I don't actually make these. I pull these out from the client, from the subject matter experts, from the people who are close to the learners. Sometimes we have some members of the learner population themselves in that kickoff and pull that out from them. And that seems to be a generally good number over time, over like 15 years of doing this.

How do you decide what a Persona is?

This is interesting because when I open up this part of the kickoff session or our ignition session, there's usually this big awkward pause like, who are these people? And I tell this story. I tell this story both in our workshops that we teach on this as well as tell this story when we're actually doing this with clients. Imagine you decide to become a wedding planner. And I know it's just a terrifying job. I'm glad somebody does it. I'm glad I don't. But you know, you know, probably somewhere on your website you're like, oh, everybody goes home happy from our weddings. But everybody coming to a wedding has different needs.

And there's kind of pockets of those needs, right? And so I ask, I said real quick, don't ever think about it. Who's the person you have to make the most happy at the wedding? And what you get back is like, bride, bride, bride, bride, bride, bride, bride. And then fraction of a second later, father of the bride, whoever's paying for it. Usually it's about 8, 9, or 10 seconds before somebody mentions the groom or they cheat. And they very early say, bride and groom. And I'm like, that's two people. Is the average of their needs meaningful? And they're like, oh, yeah, no. And this is admittedly a very heteronormative description of a wedding.

But what it helps people get is that different people have different needs, right? What a bridesmaid needs out of a wedding is very different than what the bride needs. And the father of the bride and the kids who got brought to the wedding, who. Nobody really wants them there because, you know, they might be disruptive or whatever, right? You know, the. All this, right. Great Aunt Matilda from Idaho. Right? Everybody has different needs. And then I talk about that. And so people like, oh, they can get that, right? Because most of Us have been to weddings.

And then I stack on those are all role based Personas. Bride, groom, father of the bride. Right. And then I take bride and I can slice that Persona into different sets of needs and experiences. So there's first time bride and fifth time bride and clueless bride and casual bride and bridezilla. And each one of them, they all have the same role, but they have very different needs based on what they're trying to get out of the wedding. And so we then, as we kick off the process, I'll generally tell that story and then I'll be quiet for a little bit. And then somebody will name a role or a slice of a role and that gets the ball rolling.

Nice. So are there things that can go wrong? What sort of pitfalls might occur when you're developing Personas?

I think there are three common pitfalls. There's lots of pitfalls, but there's three common ones. One is when it's based on observation bias and when we can describe it in terms of a wedding, it's pretty clear that we're really focusing on and some of our primary Personas become those primary characters in the wedding. The bride, the groom, those kinds of things. Not the little kid who sits underneath the table and eats enough pickles to make himself sick. Right, but I actually tell this story. Right, but everybody remembers the little kid who sat under the table and ate all the pickles until they make himself. Right.

And so we have this observation bias thing. We all notice this. So we all like increase its importance and maybe miss out on some of the other important characters in this space because of this thing. So that's one concern. Typically subject matter experts or people who feel like they've been wronged or burned or have lots of traumatic lowercase t experiences with the learner population will often bring up these like, oh, but this. And this group does this all the time. So that's observation bias. Another is related to it, it's stereotypes.

So sometimes groups was like, oh, and we'll have milli- millennial. Can I tell you how many milli-millennial like Personas? We give them all names, right? And they make them easy to remember. And then they start ascribing all these very stereotypical behaviors to milli-millennial. And I really, I try to encourage people to think about the actual people, not the stereotypes. So I invite them to call each other on it. And the third one is just when we have the wrong people in the room. Because if you don't know the learners, all you've got is stereotypes. And that's not helpful.

Oh, can I add a fourth?

Sure.

Is when somebody says, hey, we should use AI for this.

Ah, yeah. So, yeah, thus far you've been talking about, like, being in a room and talking to people, and I guess a lot of people might want to know, like, and I think we've even seen articles written out there about this, how best to use AI to create learner Personas.

So my point of view is, and I am definitely bullish on artificial intelligence, right. But I think there are times when what we want is it's the actual insights from actual people. That's the whole point of a learner Persona is getting close to learners. And I think there's some value in using AI in a few parts of the process.

Right.

So I've taken notes from a learner Persona process and had them, you know, like, literally like sticky notes in Miro and had chatgpt process that image and create a paragraph that I can then use to describe a particular Persona. Magic. That was awesome. You can then say, given this Persona, what might their perspective be on different things or how might they engage in different things? I know Josh Penzel has done some work where he has made a whole cast of people to be a focus group in AI and he then asks the AI focus group to have a conversation about a particular topic. And he's carefully crafted each one of those Personas and they then have that air quotes conversation. And that's pretty powerful and interesting. If I had the opportunity to have the conversation with actual humans in a heartbeat, I would actually prefer actual humans for this part because the whole point is to get close to people.

Yeah, yeah. And bonus, it gives you a sense of whether or not the team that's creating that learning actually knows the learners or not.

Huge.

Right?

And it takes. So you have to be. When you're doing this, you're kind of processing it at two different levels, Right? You're collecting the information, but you're also saying, like, how much my professional opinion can I perceive that this group really knows about the learners?

So what do you do if you begin talking with that team and either one you find out that they really don't have any clue about their learners, or maybe they even start telling you things about learners that seem like, can that possibly be true? Or like you were talking earlier, you're just like, okay, yep, stereotype. Stereotype, Stereotype.

Yeah.

What do you do then?

Well, it takes a. I try to avoid rushing to a conclusion because they always know their situation better than I do. So even if they don't know anything about their learners. They know more about their learners than I typically do in this spot. Right. Just given my role in the room most of the time, what's interesting is I ask questions to try to get a sense for what's going on. And once I get that kind of funky sense that they don't know what's going on, they don't really know their learners, it's an opportunity to say, well, let's go grab some data about them. Since we don't have them in the room.

Let's come up with enough of a working definition today in this kickoff session. I keep saying room like it's someplace other than zoom or teams.

Right.

I don't think I've done a live kickoff in a really long time. Oh, I did last year. But to then. Right. So anything we do, whether it's in a zoom room or a room of actual humans, is what we call an assumptive learner Persona. We are making assumptions about them. And whenever I can get a data driven learner Persona, I will absolutely go out and collect data and do it. There was a team from North Carolina a number of years ago that did.

They went out, they interviewed their audience. They even made little videos and scrapbook. They were like a video scrapbook full of images and sayings and actual video clips from their work. It was fantastic. And you could connect in with their Personas in really, really very human ways. It was super cool.

Neat. So Personas are useful at the beginning of a project to kick it off to kind of like start to get your mind in and wrap your mind around like, who is this for? Who are we building this for? And that's great. But I feel like also Personas are only useful if you use them beyond that stage as well. So we're. We don't want to create Personas at the beginning and then check that off the list. Yep, done. We've. We've done our due diligence here, created our Personas.

We need. They can't sit on a shelf somewhere or sit on the server. We need to do something with them. What is it that we are going to do with those Personas throughout the project?

Yeah, yeah. And I've always, I've always wanted to make like big cardboard cutouts of our Personas so they like, so they're. They're a little bit more human. We. And we've. We've had posters of some of ours too. Like you remember Gemma and Hermann from that Vitamin Angels project a million years ago, right? Where we. We see them in a little bit and see them around.

I think one of the biggest things is when we're reviewing our iterations of our project as we go, we want to be reviewing the iterations of the project with people who fall into the target learner Personas of the work that we're doing.

Mm.

And, you know, so sometimes when you're. You're pilot testing something, you're just going to grab whoever you can, which makes absolute sense. It's hard to get people sometimes to pilot test stuff. But if you're pilot testing with people who aren't like your target learner Persona, you're going to miss something important or your data is going to come back and it's funky and you might. You risk skewing to meet this other Persona that's actually not where you are targeting from the beginning.

Yeah. Yeah, that makes sense. So I have a little bit of a confession. I was doing some Google research prior to talking about Personas, and I was looking up, you know, like, what does the word Persona come from? And so I was looking into the history of that word, and big surprise, it's from Latin.

All the cool words are.

Like, all the words, Pretty much them. All of them. And originally, the Latin word Persona means mask, and it was in the theater stage. So it was like, I'm gonna put on this mask or put on that mask. And so I was thinking about that in terms of, like, our review process. And could you read through this Persona and then be like, okay, I'm gonna put on my mask. I am going to review this as if I am Gemma or Alice or Hermann and pretend that I am them while I'm taking this training so that I can see is this meeting my needs as this person. But then you need to, you know, employ people who are very good at, like, being character actors in your.

In your organization, but.

Or use AI.

Or use AI. That's an interesting, interesting idea.

So you hear a funny story about Personas.

Absolutely.

Okay. So in our company, TorranceLearning, Steve, who's been with us for eight years, is retiring. And one of the things we're doing as we're kind of refitting, how we get some of our accounting work done in light of Steve leaving, is looking at all of the steps and the processes that we do in accounting. So we're writing the procedures for this. I actually total aside here. I think of procedures as like, entry level instructional design. Right. You have to break tasks down in very, very specific, concrete, followable things.

You can test it the Whole nine. Right. And so I'm working with our accounting team, who are not instructional designers, and trying to coach that writing and experience and stuff. And I said, so what I want you to do is to think about, what if you were writing this so that I could follow these instructions? And there's like, this really awkward silence. And then I just, like, busted out laughing. It was like, oh, that's what you want, right? So, but that's the power of a procedure and also evidence and how little I can make my way around QuickBooks. So, yep, there you go.

So how'd that go, Megan?

That was kind of fun. It was. I had to resist kind of just getting into my mode of delivering the LLAMA workshop, because that's one of my favorite pieces of the workshop.

How many times have you done that now?

I have no idea.

Thousands.

I was just going to say we should count. I don't even know how we would count. But it's also interesting because right now, my students at Penn State, we're talking about learner Personas, and they just had a big discussion topic about learner Personas and stuff. So I was also thinking about them as we were doing this. So. Shout out.

This is Meg Fairchild and Megan Torrence, and this has been a podcast from Torrence Learning. Tangents is the official podcast of Torrence Learning, as though we have an unofficial one. Tangents is hosted by Meg Fairchild and Megan Torrence. It's produced by Dean Castile and Meg Fairchild, engineered and edited by Dean Castile, with original music, also by Dean Castile. This episode was fact checked by Meg Fairchild.