 When you want to create personas, one of the first questions that comes to mind is which persona should I make? And the second question is how many do I actually need? In this video, you'll learn how to answer these two questions so you can get back to work quickly. Let the show begin! Hi, I'm Mark and welcome to the service design show. This show is all about helping people like you build organizations that put people at the heart of their business. And a way to do that is by using personas or customer avatars or behavioral profiles or there are many names out there that basically do the same thing or strive for the same purpose and that is making better design decisions. This video is not about the difference between all those different terms for sake of simplicity. We'll use the term personas from now on because the challenge stays the same, the question stays the same and that is which persona should I make and how many do I actually need? So now that we've gotten that out of the way, what is the big problem with personas? Well, the big problem is that people are looking for the best way to segment them, the most ideal way to segment them. They are looking to represent the truth. That's impossible with personas. It doesn't exist. Stop wasting your time and the video. Let me explain that. So the big misconception when it comes to personas is that they have to be statistically representative of the population. They have to represent every segment of your customers in the equal amount that they have to represent the truth and that is completely the wrong way to think about personas and people are wasting a lot of time and energy trying to actually achieve that because unless you have the resources to actually make statistical representative personas, which you never will, you'll end up in endless debates what the right way is to sort and segment these personas and you'll keep revising your segmentation along the project which just creates more confusion, slows you down and doesn't get you anyway closer to the goal you're trying to achieve. I much rather take a more pragmatic approach towards personas rather than the scientific one. So this is my recommended approach for you to create personas in your next project. The first thing you need is to have data, not just any data, data that tells you something about the needs, desires, anxieties, objections, fears of the people you're trying to serve. A lot of qualitative data. We're not going into how to get the data. There are other videos on this channel about that, but you need qualitative data. And once you have the data, that's the moment a lot of people end up struggling because you see all these quotes, you see all these photos, you see all these stories. How are we going to make sense out of this? How are we going to actually sort this into personas that are going to be useful in our project? And the way you're going to make sense out of this is by answering a really simple question and that is what is a meaningful segmentation in our context, in our project? And is the answer to that question subjective? Totally, absolutely. Are you going to be able to create different segmentations? Yeah, absolutely. Is one segmentation going to be better than the other? No, not necessarily. Is one segmentation going to be more helpful? For sure. So let's use an example to clear things up a bit. Imagine that you're trying to improve the employee experience and for that, you've interviewed 15 employees. So after doing the interviews, you've got a bunch of stories, quotes, data, empty maps, a lot of soft data. Imagine that every piece of data is a different piece of candy. Yes, candy. So every quote, every photo is a piece of candy. And then we have sweet candy, sour candy, hard candy, soft candy, large candy, small candy, chocolate, cakes, and there are different types of candy in front of you on the table. But one thing is really important. We agree that everything on the table is candy. Now, how are we going to actually segment and sort all these different types of candy that are in front of us? Well, we can sort based on round candy versus square candy. We can sort based on flavor. We can sort based on size. Is one segmentation better than the other? Not per se. We have to ask the question like, what is it that we are trying to achieve and which segmentation is going to help us to make better design decisions? So in this case, if we are building a machine that has the aim or the goal to sort candy by size and weight, I don't know, or maybe by color. We're going to build a machine that's going to sort candy by color. Then it's obvious that we're going to sort by color, and we're not going to sort by if they're made of chocolate or something else. But if we are going to create a cookbook that uses candy as the main ingredient, I don't know, I just came up with that, then we're going to probably sort the candy based on flavor, right? Makes sense to sort and own flavor. Both, both segmentations are sort of, okay, they're right. But one serves the purpose that we're trying to achieve much better than the other. And the same applies for you and your own project when you're trying to make personas think of what you are building, what are you going to, what are you trying to achieve and which segmentation makes sense in your context. So let's go back to our employee experience example and think about what if we're trying to design the ideal day at work, the best day at work. Are we going to sort based on what motivates people to come to work? Are we going to sort based on the environment people feel comfortable in? Or are we going to sort based on how people prefer to collaborate? All three are valid segmentations. For all three, we could make a group of personas. But you have to ask yourself to make this selection, which segmentation is the best for you? What is it that we're actually going to create? Where are we striving? What are we striving towards? That's the only and most important criteria to actually make a segmentation. You have to consider personas as a tool for you to make better design decisions. I cannot stress that enough. Personas aren't meant to be statistically representative or snapshot or maybe they shouldn't represent reality or they should represent the reality but they don't have to be 100% accurate. They have to be a tool that you can use to design solutions that actually serve the people you're designing for. Keep that in mind. Personas need to be a pragmatic tool and think about what kind of tool do you need to actually design this solution. Hope that makes sense. I have to credit my friend Daniel for coming up with the candy example and we've used it in multiple workshops and it's a great way to explain to people why we're sorting personas the way we're actually sorting them or why we're creating the personas that we are creating. So thanks for that, Daniel. Now, there was one other question that we asked at the beginning of this video and that is how many personas do I actually need and that question is really easy to answer. You need to create somewhere between three and five personas. Period. Why and why this simple and straightforward answer? Well, think about it. You're creating personas because you want to cater for the different needs of the different types of people, right, for the different needs. Now, how many different variations of your solution are you going to actually be able to implement? It's not realistic to think that you're going to be able to make 10 different variations of a service. You should be happy if you can already make three different variations for the three different types of personas. If you won't be able to create more solutions, you don't need more personas. It's as simple as that. So anywhere between three and five at max in most projects, you'll be done. There's a lot more to say about personas. And if you want to learn how to create personas that actually work, check out this video where we talk more about that. And if you haven't done so already, click this button, which will subscribe you to this channel. And I'll see you in this next video.