UX Research Geeks Podcast – Using Market Research for Better Context in UX

Listen to a meaningful discussion about the connection between UX research and market research in episode 8 of the UX Research Geeks podcast with Tina Ličková, researcher and strategist. We talked about the importance of formal education in research, the value of conducting a problem audit before deciding on fit-for-purpose research designs, what personas really are and why they can become useless and die quickly and why how market research methodology and demographic information can provide context often missing in UX research, among other topics. Give it a listen or read the transcript below. This episode was recorded on June 13, 2022.

What UX Researchers Miss by Ignoring Market Research

00:01:41
Tina: Why is it for UX researchers, so important, in your opinion to look into market research and what are we missing out on if we don’t do it?

Michaela: Well, you know, I always come back to the textbook definition of market research. I’m a trained researcher. I’m a formally trained researcher. I have a Master’s in Market Research. So, it’s not like I went to a boot camp and then figured it out. I have been doing this for a long time. And I always come back to it because sometimes people don’t know where that comes from.

The textbook definition of it is, the identification, use, I mean, identification, collection, analysis, and use of information to make business decisions. I’m not talking, remember, notice, I’m not talking about any methods in particular. I’m talking about what the goal is because research is always about information needs.

So, you have to separate research, business problem, and research problem.

The business problem has to do with, it is focused on action, what the business needs to do, what the decision makers, what the stakeholders need to do just to move forward, whatever the decision could be, in product development, can be sales, marketing, and the research is about how you translate that problem, business problem into information needs. So, what do I need to know to be able to help you, stakeholder, to make a decision?

So, that context of the business is something that really is important in any research, market research, UX research, because if we don’t connect it to it, the business is not gonna invest in research. That’s the basic idea. I mean, we are interested, of course, we are a voice for customers, for users, we want to improve their experience, but if that insights about the user experience, about improvement of the products, or whatever, doesn’t help the business to move forward, or at least stay in business, even nonprofit organizations need to make money to support their operations. If they don’t stay in business, then they won’t have money to pay your salary and to invest in the resources that you need as a researcher.

So, the context of the business outcomes, where research leads, and how is connected to the business goals is important. That’s at the core of market research, right? That’s the business problem. That’s the context, right? And remember, marketing, sometimes people get stuck in the terms because they have not studied the field. Market research is a multidisciplinary field. Product is one of the pillars of marketing, one of the Ps, the four P’s, it’s product, is price, is place, which is distribution, and it is promotion. And people always think of marketing as just promotion of the product, but really, if there is not a product, there is nothing to market, and we shouldn’t be marketing products if there is not a need for the product in the market. So, they’re all connected. And you cannot go out and talk about user experience and products without considering in what channels you are doing that, where are you selling, where the product lives, in this case, in UX, most of the products live in the digital channel. So that’s the place in marketing.

Sometimes you have combined stores and channels. Many companies now have both, you know, more than one channel where they promote and sell the products and services. And so that’s a piece that needs to be integrated into, and of course, pricing, the best product in the world, if it doesn’t have the right price that people can afford, it doesn’t matter how much, how good it is the experience. So, all of that is connected. And that sometimes is what is missed depending on how narrow the focus is, of the research, people just don’t see the connection. And so, I always say, just look at the context, business context, the business goals.

That’s a piece that UX research needs to kind of look at market research and say, Oh, because actually, UX research is just, I see it, as one of the disciplines in market research. Market research is a multidisciplinary field that has a lot, tons of specialties. People think that, when I always ask about market research, the first thing that comes to mind is surveys. Surveys are just one of the many methods. Market research is not the same thing as a method.

It’s a system of methodologies. There are a lot of categories around it you have to consider. Is it primary, secondary? Is it, depending on what the objectives are, you know, is it the problem at hand or is it somebody else who developed this research for another objective? Is it qualitative, is it quantitative?

So, as you start using a categorization of methods, and see how they connect to the business goals, the problems that we need to solve, you’ll see that it’s much more than surveys. Surveys are just one of them. Surveys together with observation, many times, they fall on the quantitative side of the equation. But it can be primary, it can be secondary. Observation includes not just in-person observation, which is actually on the qual side, but it includes, we do transactional passive data collection, and web analytics, all that is part of quantitative data that we collect. So it’s a big field, just many people don’t know.

Why UX Researchers Don’t Need to Reinvent the Wheel

00:09:50

Tina: Sometimes UX researchers are a little bit reinventing the wheel, and I will be really interested, why do you think so?

Michaela: Well, I’m always talking about that, because I see new names for old stuff. All the time. And one of, for example, one of these simple ones is, I ask, sometimes I find UXers in events, and I asked them, do you do any research? Yeah, we do. What is it? We talk to users. That’s research. And it’s like, what exactly do you mean by that? No, we just interview them. We do stakeholder interviews, we do jobs-to-be done interviews, we do user interviews, and if you really go deeper into what those methods mean, is really, they come from in-depth interviews, and in-depth interviews is actually a way of asking questions, oral questions, and that all you know, all of that is connected to survey methodology.

Survey methodology is not about written questions. There are master’s degrees in survey methodologies. It is a big field too. You really learn about written surveys and oral surveys. And part of that comes from many of the developments in in-depth interviews, which is individual interviews, you have group interviews, which are also called focus groups. All that comes from psychotherapy, and psychology. Context inquiry, which combines observation and in-depth interviews on site. All of that comes from ethnography, psychology, sociology, and if you want to learn about that, and you change the names, you’re going to lose the connection, you know, you aren’t going to be able to go back and really learn the techniques. They have already been invented, a long time ago. It keeps evolving and developing. You don’t have to start from scratch.

If you study that, there is a whole field in what they call here polling. If you go to the American Association for Polling Research, AAPOR, that’s where a lot of academic institutions, and universities, which have research centers, federal agencies, the Census, the United States Census, you will learn tons about methodology regarding surveys. Of course, you have to ask questions differently when you do it in person or on the phone or in survey, but the core, how you structure questions, the things that you’re trying to do, there are some principles that are shared, that are common, and then you start doing adaptations to it. And when you write that type of questions, you have to consider different types of errors that you can introduce, the user, the respondent can introduce, think about the analysis you’re going to do later on with statistics.

So, there’s a lot of that already studied, and it keeps being developed. But if you’re going to reinvent something, start from, create something new, you better understand what’s the history, where that comes from, and then improve that instead of trying to start from scratch because you’re going to make a lot of errors, or you really don’t know, you don’t have necessarily the methodological background to really do it right many times. I have seen it. And it just, it’s a little bit lazy, in a way, because the means that you have to go and study, do the homework before you start trying to create new stuff.

Usually, innovation doesn’t come out of the blue, of nothing. Innovation usually tends to come from different ideas and evolution of things, instead of a new idea, totally out of the blue with nothing, no background, no history, nothing. That’s kind of a myth when it comes to innovation, and that applies not only to product, but also applies to the development of methods. If you think for example, now the hottest thing in UX is jobs-to-be-done, which doesn’t really come from UX. It actually comes from business, it’s not a new thing. It’s just an adaptation of in-depth interviews, focused on needs, making it a little more operational, trying to connect it to actions and outcomes. But at the core, we are asking questions about the needs of the users, right?

What is it you’re trying to do, regardless of the method that you are, or the tool, or the process? It is just, what is it actually you’re trying to accomplish? We have been doing that type of research for a long time, in product research in market research in the physical world. If you really want to learn about doing good product research, look at the product research done by consumer packaged goods, Procter and Gamble, Unilever, the big companies that put out there physical products, you’ll see the principles are really the same, even if the channel is different. You have the digital channel, but we do a lot of product testing with physical products. We send the product home to people, and they use it, and we talk to them, we do surveys, we observe and now with digital, it’s even easier now. Because before you had to have maybe someone there in the house watching, and right now people can just record themselves while you’re doing this. And it’s the same principle, right? We can see the interactions, we can see how they use it, we see the bigger context in which the product lives.

So, when you do, for example, usability testing you are very focused on that interaction and the navigation, what people are doing on the website, but you’re not thinking about what other things people do around that. What are the alternatives when that particular interaction comes in the context of all the things that people do to solve that particular problem?

Right now I’m planning a trip, and, of course, I’m interacting with many different websites where I’m trying, booking my trip, my hotels, trying to see where to go, but at the same time asking friends and finding other sources of information around how to do this, and the way I interact, my mental models, my expectations, my needs, are going to bleed into how I’m looking and using those websites, or the apps, or whatever it is that I’m using digitally. We want to separate users in how they do things for research purposes, but the reality is that people don’t separate that and when you do, even if you focus and you narrow something in the research, you still need to consider the context in which you are doing this, right? and what people are doing. So that’s part of the more holistic way of conducting research.

How UX Researchers Can Serve a Business

00:17:41

Tina: Where’s the opportunity for us to really serve the business?

Michaela:

That’s why I keep coming back to the connection to market research and marketing, right? So, the impact of the user experience, you know, the one that you are focusing on now as a user researcher, is really mediated by competing alternatives, by pricing, by brand positioning, by the access to distribution channels, you know, all the Ps, price, promote promotion. And all that is part of the business context that needs to be considered when we’re designing UX research. And even if you focus on a particular thing, you have to think about that user experience in that context. Even the results need to be always interpreted in that context.

This is what you learn when you do market research. Ideally. The goal is to help businesses to make business decisions. And that’s the piece sometimes that I’m missing, when I talk to UX researchers, because somehow it has evolved into making money is dirty, somehow. Well, you still need to make money to help the business make money to be able to survive, right? And so the UX researchers need to really understand the context, the business perspective.

I know there is a lot of discussions now about business outcomes. And that goes back to the initial point. I have clients that come to me and ask for, we want to do the testing of this particular app, in this particular product. And as we discussed what is it that they want to discover on the app, I always ask, are you sure that your users need this app at all? And many times, they don’t know. And I know that it’s more discussion now about discovery research and even when you do discovery research in UX, it tends to be narrowed to a particular digital channel and a particular customer journey. The view of the journey depends on how close you are looking at the interactions.

I look at it in a more holistic way with all the potential touchpoints across different channels, but you need to start at the beginning by understanding the needs of the users and the needs have different levels too, so you can have this market level need which usually falls in the market research type of research, more traditional research, and then there is the actual need at the interaction level, the micro-interaction with the product, with the service.

Traditionally, in market research, when someone says, a company says we need to acquire new customers or retain customers, that’s the main outcome many companies have, business outcomes, they need to grow, make revenue or save money, and that is done by acquiring new customers or retaining customers. In that context, the first question always is, Which goal? Which way should we go? Should we really develop a new product? Should we redesign existing products? Or is this a product development issue at all? Because it might be just an issue of awareness or brand position. You might have a great product, there is a need in the market, but nobody knows about you. Nobody understands your value proposition. So, you have to first understand what the problem is.

And if it is a product development thing, like people not liking your products anymore because a competitor came and it’s getting into your category, then you need to go understand what are the needs of the product at the highest level, at the key benefits, people don’t care about features, people care about the benefits that they get out of those features?

Initially, you start many times, with a general concept testing, if there is a need for this app, does the app needs to be digital at all?

There are a lot of questions that will have an impact on business decisions at the very high level. Once you have established that need, then you can go into trying to figure out the product configuration, the features, the specific needs, that are part of the hierarchy of needs on that user experience. But you have to start first at the beginning. And I have had clients come to me saying we want to do this. And then when they say, Well, we really don’t know, they have to go back two steps and say, Well, first let’s, let’s establish that this is needed. And then we can go and get deep into that.

Many times UX researchers are too close to the product. That’s their responsibility. That’s what they are tasked with and many organizations don’t know how they should be integrated into the bigger goals of marketing and the business, so many times now, it falls on the researchers to figure that out for themselves in trying to open and collaborate and see what is the connection of their work to the rest of the business goals so they can really show the value of the research they do. Even if they just do this little bit, that little bit is connected to something bigger. And that’s the piece that sometimes needs to be happening.

The key question, many times is they should keep asking, even if I do my usability testing, and I’m trying to figure this out, it’s under the condition that, there’s always the caveat, this is going to be working great, assuming somebody needs this product. Has that question been answered? If not, then we need to try to find the answer for that one. That includes many different types of research. It doesn’t have to be necessarily qual or quant, ideally, it should be both, before you get to the usability piece.

But you have to understand, if you start with the business outcomes and the business goals, that question will become obvious many times. But what happens in UX research in many organizations is that they are not connected to business outcomes, they don’t share those business goals with the marketing team. I mean, there are many companies that don’t have any market research and marketing team in-house. They just focus on UX only, and that’s the piece that’s hard to find the strategic value of the research when you don’t have this bigger context of marketing.

How to Direct Stakeholders to Consider User Needs

00:23:21

Tina: I think many UX researchers really want to point out if this product or this feature or whatever new coming up is needed, but what would you say, and you said one example was like doing the usability study but still pointing out okay, what are the needs, what is the context, but what it will be or what you do in order to turn the direction of the client and of stakeholders towards like, is this really needed?

Michaela: I usually start, when somebody comes to me and says, we want to do usability testing, and this is the project, I always go back and try to do what I call a problem audit. I wrote recently a blog post about that, because that’s, I think that’s a critical error. A lot of companies and teams don’t spend a lot of time on the problem definition. That’s the main problem, right? The main problem is we don’t know what the problem is, many times. So, I flip that question around before we start talking about methods. Like they say, in English, you are putting the car in front of the horses. Let’s talk about what are the decisions you’re going to be making. Let’s figure out if that’s the real problem that you have. And you will be amazed how many, even business stakeholders don’t really have a clear idea of what the problem is, they think is one thing, and as you start asking questions, and that’s where the business acumen is important.

You need to as a UX researcher, you need to understand the business, not just your little piece of the product, what is the business? Where’s the business going? What is the problem you’re trying to solve? Where are you as a stakeholder? If I come back to you, I always say. if I come back to you with this little piece of data, what are you going to do with it? And that’s where it kind of the bulb lights up. I don’t know if I can do anything with it. So well, that’s probably not then the question that we need to be asking.

And so you have to do, what is called, actually, that comes from also Psychology, a Laddering technique, which is I have seen now used in the jobs-to-be-done type of technique. Again, going back to the roots, you keep asking the why several, several times. So why will you need to do that? And what is it you’re going to be doing? Where are you going with that? And what are the consequences? What are the constraints? because sometimes, there are financial constraints, there are cultural constraints.

There’s sometimes a lot of politics, internal politics, that play into how you are going to be doing research and use the research, and the type of style of decision-making in-house. So, you have to become a little bit of a diplomat internally and negotiate, where you’re going to place the research in decision-making. I have had clients that really like research, but at the end, they bounce between the gut feeling and the data, you have to understand that balance, too, and what is the style of decision making, it is one individual making the decision, or there is a committee making a decision? All of that impacts how the research is going to be used and plays in that context.

Once you start doing the problem audit and be very clear about what the decisions are, that’s how you kind of flip it around.

Why Benefits Win Over Product Features

00:26:50

I recently had a meeting with a client, again, they came with the app, we need an app, we need to, we’re going to be testing this. They had like 60 screenshots, you’re gonna kill the user here. And what is it you’re trying to really do? Oh, we just need to know if people are going to be using the app. Well, do you know what the benefit of the app is? But we have all these interactions, but that’s not the benefit.

The benefit is what they get out of it. What is the problem you’re trying to solve? So, I kept moving them back and back and back and back, and what was initially thought of as a usability test ended up being a more high-level concept testing, trying to figure out what is the main benefit, what is the positioning, really. Maybe from there, there’s maybe a couple of features that will support that claim.

I always recommend UXers to go in and watch advertising for hair products from Procter and Gamble. If you go and look at any commercial for that category, you have, Pantene, Pro V Pantene, you’ll see that it’s only one main message. I had an example from this British singer Ellie Goulding, I think is her name and she’s talking about strong hair, that’s the main benefit. I want strong hair. That’s it. And how do you get strong hair? Oh, by the way, we have these three features, antioxidants, and this compound and this, these three things. That’s the claim justification in the why you can say, Yeah, I’m gonna get strong hair.

That’s the same thing. If you’re going to tell me that this app is going to solve my problem, make my life easier, because you know, in this area in this particular area, it is because you have these three things, it’s going to make it easier for me, that’s what people care about, then, of course, the next level is, okay, how easy it is to use, how much of hassle is this. If the interaction is not good, if you’re going to have troubles and errors, and I just recently in I was trying to make a purchase on a website, it’s a pharmacy chain here in the US. I just want to make one thing, one buy, one thing, I kept getting so many errors in there on the website, I was just totally the experience was terrible. I tried to call customer service, they dropped my call. And then they send me a little survey, will you use it again? And I said no.

I mean, the proposition, the value proposition is great, you’re gonna make my life easier, I can make it purchase easy, faster, may not easy, but faster is available. It’s convenient, in a way I don’t have to go to the store anything, then the experience was terrible. Right? You have to understand at what level you are doing the research once the need is established, the bigger need, the bigger ask from the user, the biggest benefit, then you go into the deeper interactions, the barriers and the friction points that people have in that experience, I mean, a bad experience is going to kill you if you don’t do it right, but you have to start with the initial need.

Another example I had was a while ago, a company came and say we, you know, it’s an insurance company, they want to do this app for car insurance, we want to put all these 30 features in the app and let people to, you know, they’re going compile all the information about the car and the insurance and it’s going to be great for them. And I kept asking, do people really need this? Yeah, they should, and nobody has done any research to figure out that people need this. I said, well, let’s just let’s do that first initial step of exploring the bigger need. And you can look at how people were solving this, nobody was willing to pay for that. They already had their own habits of capturing that information, their system. And at the end, they could do it, of course, a lot of barriers that you needed to break down. And then the business has to consider how much is it going to cost me. How long does it take? How much effort just to do something that very few are going to use? Right? And so that’s the piece where you have to go back to the constraints, the business goals, the information that you need at different levels to decide on, should we go with this or not?

So many times, my research is coming back and giving the bad news like yeah, I don’t think you should be doing this. And sometimes they just look at the data and say, Yeah, it’s pretty clear, not at this time, maybe in the future, when needs change, when other things happened in the market that may make people ready for this solution, but right now might not be the right time to do it.

Why The Fast Horse Metaphor Is a Lazy Excuse

00:31:27

I get always this pushback about the Fast Horse metaphor with Ford when, you know, if we asked people if they need a car, they answered they needed fast horses. And that’s a narrow and lazy definition of users’ needs. Of course, people sometimes cannot see in front of themselves how they could, what is it they need. That’s okay. People react to things. They cannot necessarily generate things because they are not really involved, that’s not the job, they are not in that category. Depending on the level of involvement, they might give you more or fewer ideas, but that’s not, the job of the user is not to come up with new ideas for you.

The researcher needs to find ways of understanding the needs and then derive what are the potential solutions to meet that need, but don’t, don’t say that we don’t need research because if people we ask them, they will never know. It’s never about asking directly. What do you think? Tell me. You have to pay them, if they’re going to do all that thinking for you, to come up with new ideas.

It is your role as a researcher, to really extract as much information as possible with different methods, not always direct questioning, and just come up with solutions, and then put them in front of them and see how people react to it. But that’s, that’s a simplistic way of looking at research when it comes to new product ideas. There’s a lot of hard thinking and work that has to go into that process to be able to extract things from the research, that’s insights, essentially, it’s not data, is the insights that you can derive from the data. And that’s one of the problems too with the trend now, with a lot of do-it-yourself research, faster, cheaper. There is really not a lot of thinking time. People don’t have budgets and enough resources to allocate for thinking time and derive the insights.

There is a lot of focus on data collection, and that seems to be easier, faster, cheaper, but there is the back end. That’s where the value is for the research. Even in the market research industry, the big focus now is on what’s called the ResTech. Of course, now everything has the tech in the name. We’re going from X to the Tech or at the same time and trying to let the machines do everything with artificial intelligence and all that and automation. But we are forgetting the insights really come from humans, still. And we need time thinking time to be able to analyze that and come up with ways of extracting those insights. And that’s the piece that sometimes is missed.

UX is very much into do-it-yourself, which I have another point of that is, it can be isolating and narrow, and it doesn’t allow really researchers to grow, and too much focus on things that not necessarily add a lot of value. So, there are a lot of new tech tools now that allow you to collect data, all that can be cheaper, faster, that’s fine. Don’t spend too much time on it. But then use all that time to really do the actual analysis and synthesis and reporting for learning, future learnings too.

Somebody was saying the other day on LinkedIn that they didn’t do reports, because who cares? I mean, who’s going to have time for that? And nobody’s going to read in the future, like you are losing a lot of knowledge in that process. If you do a lot of research, you don’t report it, you don’t document it, nothing gets into, there’s no memory of what you did. So, every time, you have to do it all over again. Again, reinvention, starting things from scratch. It’s just because that takes time. That takes resources. And you have to put, define where you find the value.

There’s always an opportunity cost to do-it-yourself research. It can be cheaper and faster on the face, but there is an opportunity cost long term where you don’t really do more strategic type of research that can advance the business, advance you as a researcher, many times.

How To Use Primary and Secondary Research

00:35:28

Tina: This is where we don’t want to talk about the methods but I will still like to have your perspective on it. Because one of the very important thing is reinventing the methods all over again. And there are big names in the business world, which tell you like don’t put demographics, or don’t use personas or don’t put demographics into personas or jobs to be done is the only method because it’s looking not at the people. And I know you’ve been vocal about this, and I would like to look into this one because I think there is a strong message from your side.

Michaela: so once you define the problem, I would say that sometimes to define the problem you have to always start with secondary research. I always recommend that. Secondary research is probably the cheapest, the fastest way to do, to explore a problem, if you don’t have enough information, or if you have already defined the problem, to find at least initial nuggets to, to answer some of the questions you have.

Secondary research is done for purposes other than your own. So that’s where you go to industry reports or category, product category reports or data collected by something else, even internal, collected data, transactional data, sales data, web analytics, those are secondary research data because it was not really collected for the primary purpose of your research, it’s just there for other purposes, but it might have some nuggets of information that might be useful for you. The problem with secondary research is that it’s usually not specific enough for your needs, you still need to follow up with something else, you need to keep doing more research. And that’s when primary research comes in. Right? 

Primary research can be qualitative, quantitative, is about the goal is research designed for your particular objective to solve the problem at hand. And you can start with qualitative methods, you can do, if you want to explore needs, again, depending on at what level we’re going, you know, high-level market needs or interaction product-level needs. Just might start with qualitative research, focus groups, in-depth interviews, ethnographic observation, like digital diaries. You need to have a sense for many different approaches. There are tons of that in market research. By the way market researchers do qualitative research a lot, tend to specialize in one or the other based on preferences and personalities. I do both. But that’s kind of rare many times. It requires different types of skills, if you are going to go deep in quant and also deep in quality. It’s the switch in the brain. Doing both on the same day is almost impossible. It’s too exhausting, but it can be done. And so you might, you might start with that with a qual and then you still should be doing some validation with quantitative research, and quantitative research could be surveys, could be passive data collection, can be some experimental design. 

Why We Need Validation of Qualitative Insights

00:38:34

I know many times qualitative researchers don’t like to hear about validation because it feels like, are you saying that what I am doing is not valid. No, when I’m talking about validation is, qualitative research generates a lot of great data, I love it, is rich in data, rich in nuances, we just still need to understand the magnitude of your findings.

When I’m talking about validation, is it how big is it, this problem or need that is worth investing in? So, you can find tons of stuff in qual, but it might be just things for, you know, small segments of that population universe, because remember, the businesses have to make an investment, and they have to get a sense for how many people are going to be reached with this product or service. And the qual, just because the way the sample is designed, it tends to be very narrow, it is not representative enough, you’re not really necessarily capturing all potential segments in your user population.

So, you might happen to have five people, you know, come from a particular segment, and you only will be reflecting results from that segment. Or you might have people from many different segments, but then you still don’t know how big they are to be able to say, Okay, should we focus on this or that if you have a lot of variability in the data, so qualitative research is, the main goal is exploratory. We want to surface some hypotheses, what the needs are, what the problems might be, then we still need to go and validate.

Many times we do, depending on time and budget, we go the other way around. We start with quant. We might do what is called, there is exploratory quantitative research too. When you do a market segmentation, there is what’s called empirical segmentation. You really don’t know where you’re going, if you don’t know where you’re going to find. You just try to ask as many questions as possible in different dimensions, psychographics, demographics, behavior, all type of variables, just to see if you see a pattern, is there some groupings, so you have to do statistics to be able to find those groups. And then sometimes you can follow up with qual, and you go deeper, because any quality quantitative research is probabilistic. There is a trend. Nobody is 100%, one thing or the other, we have just so have tendencies to do and behave and need certain things. Of course, there are situational factors. And sometimes they are used as invalidating criteria by people that don’t really have an understanding of statistics, they say, well, that person is not 100%, it doesn’t meet all the criteria in this segment, so that segmentation is wrong. It’s not.  It’s just you have to know what are the other variables, you know, increasing the probability that someone is going to be behaving in a particular way or selecting, having certain type of preferences, So, what you want is to go into the qual and get a richer profile sometimes.

Why Personas Often Die Quickly

00:41:37

When it comes to personas, in particular, because it’s connected to segmentation, personas is just another name for segment profiling. If you profile a particular group, in, say, call it a persona, that’s a persona, you better have data to back up that that persona exists in real life in a bigger in, not necessarily in one individual, but it’s a slice of the population, of your market that is stable enough, so you can really talk about them in general terms. And so, you make it look like one person because you’ve maybe picked some attributes to make it alive and relatable to stakeholders.

And that’s a trend that started actually marketing many years ago, that’s not any UX thing, but it started marketing, because when we do market segmentation, those they tend to be huge studies like big samples, 1000, 2000 people, because you need to have enough sample to be able to slice and dice to find trends and patterns in the data. If you just present percentages, it gets a little dry, hard to communicate, it’s too much data, people get really overwhelmed by that. We have to, when we communicate segmentation studies, probably like you have to do like a workshop, half day, a whole day, just to be able to digest all that because there’s tons of data.

And so a way of, when storytelling came into, in vogue, in marketing said we need to communicate data through storytelling. Then these personas, came scenarios came, like oh, let’s, let’s bring that segment to life. And this is Tina. She lives here. She has an average age of this, and she does that and that and that to give you something to connect, some heuristics to talk about that segment. And they call them sometimes even names like this is Tina segment. This is my Michaela segment. In that, over time, got confused with the fact that we can now come up with personas with no data. It seems to be the jump. So, you have internal teams, many times salespeople, because they’re interacting with customers, they see some patterns in their interactions, and they think they know what this, you know, the users are, the buyers are and so we have these personas. But what you can see in many companies is, over time, those personas don’t leave long, they die priest very quickly, because there is no data behind them. They are not stable enough to survive changes and more in-depth analysis.

Even market segmentation, depending on the category, their the shelf lives, as we call it, if it is a category that changes a lot, there is a new competition, new technologies, the market is changing quickly, they might have, the might last for two years. And then you have to renew, refresh, find new segments because things are changing. So, it’s not like you do one segmentation and that’s for life. In the product categories, the segmentations are focused on product categories, many times, unless you do are doing some type of, you know, life stage type of segmentation, more in the sociology side of things, when it comes to commercial segmentations, that has to usually connect it to what people are doing, and needing, and  behaving, and preferences in a particular category, and that changes.

I did a segmentation for a client 10 years ago where we found segments, this was for a retailer in the automotive category. That was before Amazon was selling a lot of auto parts. And so we have, people who are behaving in different ways before YouTube came and you know, you can do anything, and learn anything by watching YouTube, right? And so before that, people were behaving in certain ways, buying certain things, doing, having certain preferences. When that market shift came and other things became available, people started behaving in different ways, having other needs. So now 10 years later, they wanted to reproduce the same thing. I said, Well, I don’t think it’s going to be valid. We ran it again because they insisted, so I’m going to prove it to you, and everybody was in one segment. Before we had four, now those differences disappeared, because people are now behaving in a different way. And so, we had to go back and try to, we ran a qualitative study trying to figure out the customer journey, and what people did and needed, preferred in each of the stages up till they came, because they were, just the retailer, they were actually the last step in the journey. We have to understand the full journey from when the problem for the user started until they got to the retailer where they’re going to buy the product to solve that problem. But in the middle there were a lot of stops and decisions that people have to make and where the client can go in and help them, so they can guide them to their own store, so we have to understand the full journey, and things change in the market.

When you come and talk about personas, and there is no data, there is no actual segmentation behind it, I guarantee you those personas are going to die pretty soon, they’re not going to have a lot of evidence to justify them, and you’re going to be questioned in the C suite because now you are telling the decision-makers, oh, use these groups to make a strategy for the company, the company is going to need data. So that’s why you need to have an arsenal of different tools, understanding both qual and quant, and decide what is that we need to do that. Of course, timing and budget always are factors there, so, you have to design research considering all those factors, and I come back to the issue of defining the problem first before you start deciding on methods.

Why Demographic Information Is Needed for Inclusive UX Research

00:47:18

Tina: We could go further into different methods and looking at because it’s beautiful how you were working on segmentation and explore the journey and came to different touch points but if I go a step back to the personas, I saw you in some discussion about having demographics and not having demographics in the context of diversity and inclusion and many people consider now personas to not be inclusive, and you are more on the side it actually makes, it might make you more inclusive, which is what I found interesting

Michaela: Yeah I see, I have seen those discussions on LinkedIn, and it’s really, it shows you the lack of understanding of sociological variables in the user experience.

People are more than a user of a product. They live in the context of whatever they are. We all as human beings belong to different groups and at different levels. Our close group is our family then come our friends, then is our community, there’s a city, and there’s the country, so there are a lot of layers of identity that is part of who we are, and demographic variables, by themselves, don’t mean a lot but they mean always in the context of the cultural values associated with them, right?

Just as simple as gender, what does it mean to be a woman and a man in different cultures, in different countries with all that connotation when you say male and female, and that’s why we have all these cultural wars about sexual orientation and transgenders. If that would not be an issue, if that demographic piece of information was not important, we wouldn’t have these types of discussions about, you know, gender differences and the me-too movement and all that. It has sociological cultural connotations.

Age is the same. Different age groups, depending on the times that they live, the events they’re exposed to, they’re gonna have different views of life, and we have done research where you can see how different groups see the same thing from different perspectives.

Race is another variable that really has a lot of impact. It will vary depending on the country where you are, but here in the U.S., given the history of the country, there’s a lot of discrimination against blacks and non-white groups, so if you don’t consider those variables, you’re going to be discriminating people because you are not necessarily understanding the things, the values attributed to sometimes features that are important to them. I’m going to give you a simple example that was brought to me by talking to a group of black researchers. Black consumers in the U.S. are trendsetters here. They are the ones who define many times, although they are appropriated by other groups, trends in fashion, and products in many product categories. The icons to represent females, if you look at how the hair is designed, it indicates, it signals straight hair, just that simple, while we know that hair, natural hair in many non-whites and black African American is just curly hair, different types, they’re like nine categories of them, but they’re essentially curly many times. And just looking at that piece, they immediately don’t feel represented.

I have been, I work particularly in that category, I do a lot of work in the hair category, and I can see when opportunities are lost, and there’s a lot of misunderstanding when you have a whole team of researchers that are just white, white women trying to sell and produce products for black consumers, black women with curly hair. They don’t see it because that’s not how they relate to, that’s not their point of reference, that’s not their experience, their individual experience, so that’s why diversity is important and representation in the research groups, not only the research participants but also the research teams because we all bring own biases, our own culture, our own experiences, and we need to have others to keep us in check. We’re always going to have biases, that’s part of how our brain works, and so you need to have diversity, and when you do research, it’s not necessarily, how you actually do the nitty-gritty of the research, but in the design of the sample, you can do perfect methodology, and it might be biased in the sample, might be biased in the interpretation.

Many times there’s no question about the methodology but then the interpretation of the results is where the bias comes in, and if you don’t consider gender, age, race, religion, whatever the variable is, that is relevant to your product category, and some of them are really relevant and others are less relevant, so you have to always get that in context.

I always say well, demographics may or may not be important for this but if you don’t know capture that information and then analyze it later and see okay, does it make a difference or not, but not just come a priori under the assumption without any data that says we don’t care about that, we should not be looking at that, because you are going to be missing, you’re going to be probably be discriminating.

You know there’s a lot of discussion in the U.S. coming about disabilities and if you don’t ask about that, you would never know if you are excluding people because they might have any type of disability.

It’s the same thing if you are developing an app, for example in some categories, like dating. I have also done work in that area. There are a lot of niches. People, when they date, there’s some criteria they consider, and for example, religion and political affiliation, and race, of course, age and gender, they’re kind of a given, but people because of cultural factors might prefer to just date within their own group, so if you don’t consider that, you’re going to be missing big time if you are developing an app for everybody, and so for each category, we know the demographics, if you have been doing market research long time, you realize that in some categories demographics is really really important and others are less but they are, we call it covariates, they kind of mediate some of their approaches to products and services. They might not differentiate segments up front, like maybe not necessarily, you might find something, so it’s not different by gender or not different by age, but you might find big differences by race, and so you shouldn’t a priori assume that they don’t matter, they always give you context.

Demographics give you context about the world they’re living in, and right now with so many cultural wars in place which usually hint at religion and gender and race, it’s very hard to argue for not including them. You should always include them and then see if they make a difference or not. I’m not saying that they do. I’m saying as a researcher you should capture the information and analyze it.

What Should Be An Integral Part of A Researcher’s Work

00:54:44

Tina: Michaela maybe the last question to summarize, what you are trying to tell the listeners or not you, but what we were trying to give away as a message, what should be in your opinion an integral part of every researcher’s work?

Michaela: I would say that in terms of the process first, conduct a problem audit, identify what the problem is, what the decision is at play, what course of action should be taken, and translate that into the research problem. I would direct you to just look at the longer blog on problem audit. I think that’s so important, it is so underrated and totally ignored many times because people are in a rush to do things but many, many, many bad decisions are made when the problem is ill-defined, so start with a clear definition of the problem and the level of decision that is made.

The second is always of course connect that business outcomes, consider where the business is going. If you don’t know, ask, find the people who know. There’s always people who know and try to understand how your tactical project is connected to strategic goals. It will really elevate the function of research and your value as a researcher. Never forget you are the voice of the customer, of the user. That’s fine, but you still need to help the business to stay in business, at least if you think that making money is dirty.

And get a good foundation on principles of methodology, both qual and quant. I know some people prefer numbers, some people don’t like numbers. You don’t have to do it yourself. That means, collaborate. Considering the cost of do-it-yourself, the hidden cost, the opportunity cost, the things that you could be doing, instead of that, that would add more value.

Remember, no method is perfect. Not one method is gonna give you everything you need.

No user interviews, no usability testing, no focus group, no surveys are gonna give you all the answers. Ideally, you should be using mixed methods to give you different perspectives, but sometimes there is no time or money to do that, but just know the limitations, don’t present it as is “this is it.” This is all the answers. You never know. It’s just one perspective, and be big on caveats. Right? But you have to know the methodologies to be able to defend them and propose the right one, otherwise, you’re gonna be a prisoner of the law of the hammer. You know, when you have a hammer, everything is gonna look like a nail.

You don’t want that as a researcher, but it’s easy because some companies, they stay in that lane, you create expectations, always the same type of research, and so you don’t grow, the company doesn’t grow, you really don’t solve the problems that need to be solved. And that’s why I always recommend collaboration, not only internally with other groups, but also bring outside resources, bring research vendors, suppliers.

They’re gonna have the expertise that you, you won’t have or your team have. If you’re gonna have all the expertise, all the specialties needed in research, you’re gonna have to have a big research team, and then you are in the business of research, which is not your business, righ? Internally. You’re gonna have a hard time justifying that.

So the market research industry has gone through that phase already. It goes back and forth. Sometimes it’s all do-it-yourself. Sometimes it’s all vendors. Ideally, in my experience as a researcher on both sides, because I have worked on internal teams too, is just collaboration. It’s always a hybrid model where you can have a small team that brings experts as needed for different types of projects.

Not for all of them, but some pieces of the projects, on different types of projects, strategic and tactic, depending on your own skills and resources. So, collaboration, open your mind to that, and bring resources and you can justify having a budget for that And finally, be aware of your own biases, your cultural biases.

That’s why diversity is important because you are many times blind to your own biases. You [won’t] be able to catch them yourself, even if you are vigilant. Knowing what you don’t know is very hard. So you need a diverse team, you need another perspective, people who can point out they can complement. You need a diversity of methodologies, of people, of skills, of people with experience in different adjacent research fields and that will really complement and enrich your perspective and, and the value of the research.

But again, don’t stop at the research piece, you have to go to the next step, which is the harder one, which is the implementation of the insight. And that’s what we need to include the stakeholders. The stakeholders don’t have to do the research.

That’s another thing. There is now democratization of research. They expect everybody’s gonna do. Not all stakeholders are interested or [have] time. They’re not getting paid for, to do that work. So don’t expect them to do the research, but you have to find ways to involve them in the problem definition, and ideally observation of the process, if they can.

At the end, in the insights implementation, you have to be able to communicate the results, the insights, and help them make the decisions. That’s where the implementation comes. If you do all the research and then that goes to a shelf to collect dust, the value is lost. You have to be able to make it actionable.

That’s the piece that sometimes requires more work and thinking time. So you have to advocate for that. If you are a research manager, please, please find time to give your team time to think about the insights implementation. That’s where the value is. That’s where you can really bring research alive.

The actual doing of the research, which I know now is a big thing with research operations, that really, for the business, has less value than what you do with the research. Again, nobody cares about the features, how you do it, what they care is, what they get out of it, and the same thing applies to research. You have to make sure you do it right, that you have a team who can do it, but the business itself, the stakeholders, they don’t care in that way. They care is like you did it, right? I’m assuming that you do it right, you’re an expert, you know what you’re doing now, tell me, I can trust you, this is good research, now help us to figure out what to do with it, help us to figure out how, what decisions we should be going for. At the end, at the highest level, that’s, that’s our role.

 

To listen to this and more episodes of the UX Research Geeks podcast by UX Tweak, go to:

🎧 Apple

🎧 Spotify

🎧 Google