How To Identify Your Customer's Needs

Alan Klement at UXDX Europe. Video: https://youtu.be/s0TPFtvd-Zc

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Environments, constraints and needs

[00:00:01] Today I've been asked to come here and talk about customer needs, consumer needs, demand, all that good stuff. This is a topic I love talking about with others, because it involves so many other exciting fields of study that I also like to investigate: fields like psychology, cybernetics, complexity, systems and so on. All of these contribute, I believe, to a really important understanding of what motivates people to go out and adopt and use products.

[00:00:41] I've got really good news for everybody. The answer to what customers want, how we think about needs and how we communicate those needs is very simple, really. It comes down to two things. On the one hand, you've got the environments that people want to create and maintain for themselves. On the other hand, you've got constraints: things that get in the way and block people and make it hard for them to create and maintain the environments that they want to have for themselves. When those two collide and come together, it creates this emergent phenomenon of needs.

[00:01:27] To give you an idea, I'm going to visualize this; this is how I think about it in my mind. You've got the desired environment that we want to create for ourselves, and then our situated constraints, and they clash together, they smash, and the output is needs, and of course other behavior. Shopping: I'm going to look for something different. Maybe reinvention: OK, I need something, so I'm going to mess around with what I have and maybe try to tweak it a little bit. This is how I think about how needs are created, and how we should think about them and communicate them to others.

[00:02:10] For most people, the idea of situated constraints is pretty straightforward. Constraints are just things that get in our way. Where maybe people are not so familiar is the idea of environments. That's what I'm going to start off with today and spend most of my time talking about. What does it mean when consumers are aspiring to create and maintain certain environments for themselves? How is this different from outcomes? And what's the relationship between outcomes and environments?

Singer and the acorn

[00:02:47] All right, I'll begin. I believe the story of environments is very fascinating, actually. A place I like to begin is with Singer. Singer was a philosopher of science who lived in the late 19th century and early 20th century, and he did something quite revolutionary. He changed science forever. Up until this point, when he started writing in the 1890s, science was approached with what they call billiard ball cause and effect. Meaning, if I'm holding the pool stick and I hit the cue ball, it hits the three ball, which goes into the pocket. Very simple cause and effect. This was how science was done, or thought about, for hundreds and hundreds of years. But he proved that that's not how the world works.

[00:03:51] He did this through a long, complicated proof, which I'm not going to go through today. I doubt anyone here wants to hear me go through ten minutes of a mathematical proof. So I'll give you the illustrative version of what he showed. It goes like this. If I plant an acorn in the ground, is the outcome an oak? Some of you might be thinking there's probably a trick question in there, so I'll make it easier. If I plant an acorn in the ground, will that cause an oak? Or if I plant an acorn in the Sahara Desert, will that create an oak? How about if I do it at the bottom of the ocean, or on an iceberg? Will those actions create the outcome of an oak?

[00:04:50] I think intuitively we all know: of course not. Why not? Because planting an acorn in the ground is not enough. You actually need an environment of things working together in order to grow that oak and allow it to flourish. Things like nutrients in the soil, moisture content in the soil, humidity in the air, sunlight, not too much, not too little. All those things must work together to create that oak. That's the first thing Singer pointed out.

[00:05:28] The second thing he illustrated is that, now that we've moved away from cause and effect to environments producing outcomes, you can't really give credit or blame to any one particular contributor to those outcomes. For example, suppose I do plant an acorn in the ground and it becomes an oak. Can I take credit for that? Can I say, "I made that oak tree"? Of course not. But the inverse is true as well. If I plant an acorn and it doesn't create an oak, is it my fault? Those two things had a huge impact on how people thought about and understood the world. You need environments of things all working together to produce outcomes, and if you do or don't get a certain outcome, you can't really blame or give credit to one particular factor.

Shewhart, Deming and the control chart

[00:06:38] This idea of environments caught on, and the next person to introduce this thinking was Walter Shewhart. He was a mathematician working at Bell Labs, and he thought, "OK, environments, this is interesting," and he brought that thinking to manufacturing. He also revolutionized manufacturing. Talk to anyone who's familiar with quality control or running plants, and they may not remember Shewhart, but they'll say, "Oh yeah, I was supposed to study that guy at some point in school."

[00:07:20] How did he revolutionize manufacturing? He did it by introducing this: the control chart. What this is describing is the environment of the manufacturing process. It's sampling important variables of that manufacturing process to try to figure out which things are out of scope and which things are in scope, but also things like trends that are telling us maybe something is wrong, something's off.

[00:07:54] Because Shewhart, from Singer, recognized that focusing on the outputs and outcomes of our manufacturing processes is too late. You've already made the product, so if it's no good, it's too late. It's difficult to figure out what the problem is, because then you have to work backwards and figure out why this lot of telephones was no good. That's maybe a little hard to figure out. So instead, if you want to get the outcomes and outputs that you want, you have to create and facilitate the right environment and monitor it continuously. So that's Shewhart.

[00:08:35] Next thing: Deming. Deming took Shewhart's ideas and Singer's ideas. Deming was a protégé, I guess you would say, of Shewhart; they worked together. Deming applied this idea of environments to management, and I think this quote illustrates Deming quite well: we should work on our processes, not the outcomes of our processes. Same thing. Focus on the environment and the processes in that environment, and if you take care of that, you'll get the outcomes you want.

What this has to do with customer needs

[00:09:16] I've been going on for ten minutes. Some of you might be thinking, thanks for the history lesson, all very fascinating, but what's that got to do with customer needs and design? How is it relevant to us, to my work? Well, I'm going to argue that this is immeasurably relevant and important to your work, and actually I have been talking about needs all this time.

[00:09:45] I'll bring it back to this. Have any of you seen something like this before? OK, some hands going up. Actually, let me change the question: you have seen this before. Does anyone know what it is? Can you describe it? Oh, sorry, I'll go back. I put up too much. Customer journey map.

[00:10:18] With the control chart, what we're describing is a manufacturing environment, sampling it over time and testing for certain variables: are they in scope or out of scope? Same thing for a customer journey map. You're describing an environment in which a consumer is interacting with a brand or product, continuously sampling that experience and those important variables, and seeing whether they're in scope or out of scope. This is very helpful for figuring out where they're running into problems, where they're unhappy, where there are maybe huge swings in the experience.

[00:10:56] And it's the same thing, instead of just waiting for the customer to... The reason the control chart was so useful to manufacturers is that they could anticipate problems. They could say, OK, here on August 18th, or in this month, between April and May, we had this time where the manufacturing process was acting really odd. They could anticipate that the outputs, whatever widgets they're making, were going to have defects. Same thing here. You can get ahead of the customer: "OK, they're having a bad experience here," even before they leave your product or start complaining. You can anticipate that they're probably going to complain or be unhappy about it.

Outcome goals versus environment goals: Brexit

[00:11:47] So that's the idea of environments at a high level. Now I want to go a little deeper into it and talk about environments versus outcomes. I'm going to use the Brexit example to describe outcomes and compare them with environments. I don't need an American to come here and explain Brexit to you all, but if I could just reduce it down: you've got the referendum and the voters. You can almost think of the voters as the users and the referendum as the product, or the brand. There's an interaction: they came together on a certain day and cast their vote, and then you had two possible outcomes of that. Brexit passes, Brexit fails. And of course, we all know what the outcome was.

[00:12:48] I want to describe some important characteristics of outcomes. For a group of people, Brexit passing was an outcome goal. Why are you showing up to the poll and casting your vote? Well, I want Brexit to pass, for example. Important characteristics of that, I believe: notice it's an output of a specific interaction. There's a referendum and users, and they come together at a certain time and date. That's the interaction, and then there's a particular outcome of it. Also notice how there's a clear start and end to that interaction. Brexit passing mattered only for that one particular day. You could argue people were anticipating it and it was important before then, but it really mattered on that day. Afterwards, it didn't really matter so much to that group of people whether one vote went one way or the other, because it had already happened.

[00:14:00] So that's outcome goals. Now, does this really describe what these people wanted? You could argue, yeah, they wanted the referendum to pass and they wanted the UK to leave the EU. That's what they wanted. But is that really what they wanted? Luckily, people have gone out and asked them, "Why did you want the referendum to pass?" Two of the big reasons they said were that decisions about the UK should be taken in the UK, and that the UK has control over immigration at its own borders.

[00:14:42] These are a little different, I think. Are these the outputs of a specific interaction? Is there really a clearly defined start and end? Does immigration control matter on Monday, but not on Tuesday, and then maybe it matters again on Friday? I don't think so. I consider these environment goals.

[00:15:14] Using control over immigration as an example, let's unpack some of the characteristics of an environment goal. It's describing an attribute of a state. Meaning that as long as it's important for me to have a particular vision of how I think the UK should be, control over immigration is going to be relevant to me. It's describing the state, and as long as that state is important to me, that goal is also going to be important, on Saturday as well. The other thing is there's no defined start and end. As I mentioned before, it's not like control over immigration matters on Monday and Tuesday but no other day of the week.

[00:16:09] I like to visualize it this way in my mind. Think about this barbell for outcome goals: here's where it matters, here's when it matters, here's where it doesn't matter. Whereas an environment goal is a process. It's always going; it's always relevant to me.

[00:16:24] I'm going to make the case that consumers really want the environments and not the outcomes. I'm speaking generally. I'm not going to say outcomes are never important and so on, but they're intermediary goals, and what people are really aspiring to is certain environments.

Arlo: protect and defend, care and aware

[00:16:44] I'm going to prove that. This was a research project I did a couple of years ago with Arlo. The story here is that Arlo had established itself as the number one brand and product for the consumer home security market. In fact, they probably created that market. And as often happens when someone creates a market that is very lucrative and grows, it attracts competition. So they reached out to us and said, "We have to grow our product. We have to get more people to buy it. We also want to stimulate innovation: how can we find more growth opportunities? And how can we fight against the competition? Help us stay ahead of our competitors, who we know are going to come in and do their best to steal our customers."

[00:17:44] That's what we did. My team and I went in and started talking to people who had recently purchased these Arlo cameras. At first, the Arlo team was a little confused by what we were doing, because we weren't asking about the product at all. A little bit, but most of the conversation was about other stuff. The Arlo team was saying, "How easy was it to set up? Ask them what it was like to set it up." Or, "The picture quality: ask them what they want from the picture quality, what it's supposed to show and not supposed to show." Things of that nature.

[00:18:27] But that's not what we talked to them about. We started asking them questions like, "Before you were using the Arlo camera, what else were you using? What did the Arlo camera replace for you?" Or, and this is the magic question, "Now that you have the Arlo camera, what can you do now that you couldn't do before? How has your environment changed?" Another great question to ask your customers to really get at what they're trying to get to is: "Compare you with the Arlo camera today and you three months ago when you didn't have it. Let's compare and contrast those."

[00:19:13] As we asked these questions more and more, the R&D team started to get it. They started understanding that what we were driving at was the environment that these consumers were trying to create and maintain for themselves. We came across multiple environments, but for this talk I'll keep it to two. One environment we call protect and defend, and the other we call care and aware.

[00:19:42] Some attributes of protect and defend: "I can work with my neighbors to fight crime." "I know who is in and around my house." "Criminals don't see my house as a target." Things of that nature describe the environment, what it feels like, the experience of protecting and defending my home.

[00:20:07] What about care and aware? We heard things like, "Now that I have this camera, I can keep an eye on my kids while I'm away. I know that they're OK." Some people talked about being really excited about knowing what kind of wildlife was on their property. A lot of the people we talked with lived in the Midwest of the United States, and they loved seeing turkeys and deer and other little wildlife creatures roaming around their property. Another was, "I want to always know about gas or heat or water problems going on. If there's a gas leak, I want to know about it right away. If someone left a window open and that's causing the air conditioner to work overtime and use up lots of energy, or there's a temperature drop problem, I want to know about that." And another one which I really liked was, "I want my home to welcome me when I arrive." These were the characteristics of the environment that they were hoping to experience and create and maintain for themselves.

Designing a constellation of products

[00:21:25] Armed with this new thinking, they did make some improvements to the camera, but they were now able to abstract and think about opportunities outside of just their camera. For example, they started creating multi-sensors that you could just pop around your house to check for moisture, water leaks, gas, carbon monoxide, temperature, light, motion, things of that nature.

[00:22:00] They also thought about how they could make changes to the app. One really important change they made: in the old way, people would always complain about getting all these notifications. Every time the camera saw motion, it would pop up: "Motion detected. What do you want to do about it?" Nothing, apparently. People would complain, "There's too many notifications," and the team thought, "OK, we'll add a sensitivity setting." That's how they controlled the amount of notifications. But once they recognized that it was about a particular environment, they could create more meaningful push notifications.

[00:22:54] They did something really cool, which was to match it with facial recognition, for example. Instead of saying "motion at the front door", and you think, "What is that? Oh, not important," they would say, "The mailman just came." Or, "Michael just got home," or just left, like your son. Or things like, "An unrecognized person has been detected." So now you're creating more meaningful push notifications for people, and not just directly responding to their complaint about too many notifications. That was one really big change with the Arlo app.

[00:23:31] Another thing is the Arlo remote, which I thought was great. Again, tying it back to the experience of "whenever I come home, I want my home to welcome me," they came up with this: you can pre-program certain welcome home sequences. "I'd like to have the driveway light up, and certain things, the front door, and so on." But we also heard stories like, "When I come home, if there are any bad guys in there, I want them to be frightened away." So they have security options here too. It's dual purpose: it welcomes me home and scares away anyone inside, but if that person is still inside, I can hit the warning button and call the police.

[00:24:19] Another thing here is the Arlo siren and strobe light. If it's an unrecognized face, a bad guy is seen, shine a spotlight on them to make sure they know they've been seen, but also a red strobe light to notify the neighbors. So this also touches on that goal of "help me work with my neighbors to fight crime." It's one way of notifying the neighbors: "Hey, there's something going on here. Please respond."

[00:24:51] The way I like to think about this, as an experience designer, is to think not just about your product itself, but how you might create a constellation of products and services that work together as a system to create and maintain the environments that consumers are hoping to have.

Dos and don'ts

[00:25:19] As I wind down, here are some dos and don'ts that I want to leave you all with. Don't think of outcomes as what customers want. I'm speaking generally; I'm not saying outcomes are not important, and I probably should say "only". Don't think of outcomes as the only thing customers want. Rather, think about environments. That's really what they want to get to, and outcomes are really a means to help create and maintain those environments.

[00:25:51] Another big don't, I believe, is to study consumers in their current environments only. Instead, I recommend studying the environments that consumers aspire to create and maintain for themselves. I call it environment experience, EX. Maybe it'll catch on, who knows. I'm not a huge fan of the ethnography methodology and contextual, embedded design or research. I think that's good for usability and for finding issues with your current product. But if you really want to figure out what kind of environments people aspire to, you have to talk to them about that. It's a different type of research.

[00:26:44] And don't think of yourself as a problem solver only. Think of yourself as a gardener. Think back to what Shewhart was doing with his control chart, and the same thing for the customer journey map. You're not really being a problem solver in the sense of, "We've identified a problem, now let's go find the cause of it and solve it." Again, that's the billiard ball cause-and-effect model, and that's going to be very difficult to figure out.

[00:27:16] Instead, what does a gardener do? Is a gardener really a problem solver? Or is she someone who has an idea in her mind of the kind of garden she wants to design for herself, and figures out what she needs to do to make sure the garden is thriving? What soil content needs to be just right, sunlight, and so on. She doesn't wait until the tomato plant dies before recognizing it needs more sunlight. She's looking at the sunlight, figuring out how much more or less she needs to add, and then adjusting. All right, that's all I have. Thank you.