23 MIN READ
Interview with Nir Eyal and Bill Cushard: Changing Behavior Through Software Training
We're pleased to share an interview with Nir Eyal, author of the book Hooked: How to Build Habit-Forming Products and ServiceRocket's Head of Training Bill Cushard, moderated by Sarah E. Brown. Nir and Bill discussed how behavior psychology relates to customer training and customer success and how companies can apply psychology to enterprise software training.
Sarah: I'm here with Bill Cushard, head of training at ServiceRocket, and Nir Eyal, author of the book Hooked: How to Build Habit-Forming Products. Eyal blogs frequently at Nir and Far and lectures at the Stanford Graduate School of Business Design School. Nir and Bill, it is so great to have you here with us.
Nir: Thank you. Pleasure to be here.
Bill: Yes, it's great to be here.
Sarah: Today we're going to talk about how behavior psychology relates to customer training and customer success and how companies can apply psychology to enterprise software training. Onto our first question. Nir, we're big fans of your book, Hooked, and would love it if you could tell us a little bit about your book and what inspired you to research and tackle this topic.
Nir: Sure. The book was really something that came out of my own frustration. I spent many years in the gaming and advertising industries. And before that I was at another tech startup where we were constantly building new products, new ideas to market, and we constantly came up against this problem of not knowing exactly what to build next. And, of course, there's a large element of risk in terms of knowing what to build next, but this was something that we constantly struggled with.
"What do we build? Do we build what the investors say we should build? Do we build with the loudest customer says we should build? Do we build with the highest paid person in the room says we should build?" And so I was frustrated with that process and I didn't feel like it was very evidence-based. It was a lot of throwing spaghetti against the wall and trying to see what sticks. And some of that is part of the innovation process that we have to try new ideas, test them, measure them, and learn from them as the lead start of methodology tells us.
But I was hoping for something a little deeper. What I wanted to do was to find a guide book, so to speak, on how customer brains work. How does our deeper psychology affect the decisions we make and the products we buy? And I didn't really see that book on the shelf. I saw a lot of books around behavioral economics or consumer psychology. Some of them were academic and some of them weren't very practical. What I wanted to do as someone who had been on the front lines building product, I wanted to read a book that could help me build better products, that could help me spend less time, money and effort building the wrong stuff so I could spend more time building what customers really wanted.
Sarah: Excellent. That's a great overview. In software training, we're offering training that drives this changing behavior to help customers be successful. What are some of the key lessons in behavior psychology that can be applied to software training?
Nir: I think that training and learning, in general, should be an enjoyable process. We see across the education field in general that companies have a really tough time keeping students engaged for a couple of reasons. One is that much of the content is just mind-numbingly boring. That's one problem and the other is that it's just so easy to get sucked into these other tools that we use all day, these other products like Facebook and Twitter and Instagram and our e-mail that are designed to be so habit-forming. A lot of what I talk about in my book is why are these products so engaging? How did these products create habits in our day-to-day lives? One of the first lessons that I would utilize is that people become much more likely to do a behavior when it's something they want to do as opposed to something they have to do.
There's a phenomenon in psychology called reactance. And reactance tells us that when people's autonomy is threatened, when they're told what to do, they rebel. And so this is a very important lesson to us that we can expect to get much worse performance when we make people use our product and our product is not enjoyable to use, when it's difficult to use, when it's not something that's fun. Whereas if you think about a product like Facebook or Instagram or WhatsApp, these products are not products that people were forced to use. These were products that people wanted to use. And so there's all kinds of things that we can do to try and make these experiences. It may require effort, they're not effortless behaviors, but things we can do in the design of these products to make them something that people actually want to engage with.
Bill: I have a follow-up to that. Because I'm in the business of training people on software. And many times the people in the room are not the people who bought the software. Their bosses have told them, "We bought this software, now go into that training course and learn it." And even sometimes the trainer isn't aware of all this. The trainer has been sent to do this thing and the students have been sent to learn this thing and there are times when people are wondering what the heck they're doing there because their old software was just fine, thank you very much. And so I wonder, how you get to the want to?
Nir: I hate to tell you this. I think you're in the middle here. The trainer is in a bad position because frankly for far too long, we have accepted the fact that enterprise software can suck and most of it is awful. It is so bad. And so the good news is that these companies are finally being taken to task. What we're seeing is that with the consumerization of IT, having crappy software that's difficult to use, that's hard to learn, that's abysmally difficult to try and understand the steps to utilize the product, that age I think is coming to an end because what we're seeing is that the consumer side has shown us what good software should look like, that it's should be easy to use, that it should be habit-forming to follow these four steps of the hook that I describe in my book.
"The consumer side has shown us what good software should look like." -Nir Eyal
So I think those days are changing and we're already seeing that products like Slack, best growing enterprise product in history, GitHub, SalesForce, Stack Overflow, Minerva, all of these products are coming to market by showing that they can be adopted from the ground up. It's no longer the case that enterprise products have to be bought from the top down, that many of these products are bought from the bottom up because they're just easy to use, because they're something that people want to use to get their job done.
Sarah: Following that, how can trainers apply your research?
Nir: The focus of my research is on software, so that's the first place to start. You have a much more difficult time if what you're trying to shove down people's throats is very difficult to learn and unintuitive and poorly made. But that being said, what makes a great teacher, I think is the same no matter what they're teaching. And the core of great teaching is great storytelling, that being able to use what's called a variable reward, which happens to the third step of the hook.
A variable reward comes out of the work of B. F. Skinner, the father of operant conditioning. Many of you probably might remember B.F. Skinner from your Psyche 101 class. He was this guy who took these pigeons. He put them in a little box and gave these pigeons a disk to peck at. And every time the pigeons would peck at this disk, they would receive a reward. And at first they would receive a reward on a predictable schedule. Peck at the disc and receive the little food pellet and he could very quickly train these pigeons to peck at the disk whenever they were hungry.
But then Skinner did something a little bit different. He introduced what's called a variable reward. Sometimes the pigeons would peck at this disk and nothing would happen. No reward would come out. The next time the pigeons would peck at the disk, they would receive a reward. And what Skinner observed was that the rate of response and number of times these pigeons pecked at the disk increased and occurred more frequently when the reward was given on a variable schedule of reinforcement. And so we find these variable rewards in all sorts of things.
When you think about what makes news interesting, the first three letters of news is new. It's what you don't expect to happen. We don't have news telling us, "The same thing that happened yesterday is what happens today." There's a reason that newspapers are worthless the next day, even though much of the content takes the same format. Nobody wants yesterday's newspaper. They want today's newspaper.
We see this variable reward in all sorts of things. It's what makes games fun. It's what makes movies interesting. It's what makes books interesting to read. Any of these things that keep us engaged use these variable rewards. One of the best tricks that we can use as teachers, as educators is to tell these stories with what's called a mystery gap, an information gap, something that leaves the listener guessing about what might happen next.
Bill: That's very good because it brings me back to a question about your definition of trigger. And there's a phrase in there about doing something with enough frequency to get to that habit. And so I wonder if there's the element of information gap and the mystery. Is there a frequency level that has to occur? And I'm saying in a training course, but I suppose in a product too. But what's that frequency level like and what does that mean? How often?
Nir: Frequency is a very big deal because it's almost impossible to form a habit around the behavior that does not occur with sufficient frequency. There's all these urban legends going around about this magic number. It turns out that that's all scientific funk, that there's no magic number. What we do know is that the more frequently the behavior occurs, the more likely the habit is to form. So, in the training context, the more repetition around that behavior, especially in the early beginning learning phases of learning a new behavior, becomes incredibly important. And we also know that there's a precipitous drop-off in the likelihood of retaining this habit if the behavior does not occur within a week's time or less. So that seems to be the only magic number that we can find.
Bill: So I guess, in my world, that would be something called a forgetting curve. It's a similar concept, do you think?
Nir: Similar concept? I think so. I'm not exactly familiar with that term, but I think that would be right. After a week or so, that's really when you start forgetting the stuff. Is that right?
Bill: Yes. Or a short period of time, yes. And trainers want to intervene in that forgetting phase so that it doesn't drop off too far.
Nir: Makes sense. Exactly. So we've got to intervene, we have got to get them to do that key behavior to reinforce that well, ideally, every day. If you think about some of the most habit-forming technologies that we have around today and the companies I profiled in my book, Facebook, and Twitter and Instagram and WhatsApp and Snapchat, all of these products, these things are not just used every week. They're not just used every day. They're used multiple times a day. They're intra-day behaviors. There's some stat that show that the average smartphone user checks their phone 150 times a day. Well, the habit-forming potential of these products is immense because of how frequently these behaviors occur.
Bill: Interesting. Just the other day, I got an e-mail from a software vendor. It's an analytics software that I use, except that I don't use it every day and they sent me an e-mail, it must have been automated. It said, "Hey, you should use this every day," and they gave me three reasons why. I think they were tracking me. I think that's what they're trying to do.
Nir: There you go. They might have read my book.
Sarah: They'd be smart if they did. What are some of the biggest mistakes that companies tend to make in trying to hook their customers in regards to their software?
Nir: Right. I think the first mistake is thinking that it's just good luck, that these incredibly powerful products that I mentioned earlier like Facebook and Snapchat and Instagram and all these technologies that we can't stop using, that these companies got lucky. These companies did not just get lucky. These products are engaging by design, that these companies' business models depend upon habits. And so the biggest mistake that a company can make is not recognizing that in fact our habits are hackable, that there's a science to these habits, that the psychology of designing for behavior gives us some very clear rules that we can follow to make it more likely to form these habits in our users' day-to-day lives.
That's the biggest mistake is not take the time to understand your customer psychology. Because what typically happens is that we make a product and we believe it's a generally good product, whether that product is a course, whether that product is an app or any type of product or service that we're selling to our users, and we think it's fantastic. And when users don't use that product, we think, "Well, they're just dummies. They just don't understand how good this product is. You know what I'll do? I'll talk louder and I'll convince them more and I'll make them watch videos about why my product is good and I'll send them literature, telling them, convincing them."
But it turns out there's this whole other deeper psychology that drives our behavior in ways we can't articulate. People can tell us they'll behave a certain way, and with all good intentions, they want to behave that certain way, but if you've ever gone on a diet before, you know that your good intentions don't always last, that many times it's the fact that we don't understand our own psychology that comes back and bites us. So as a producer of a product or service, the critical mistake that many companies make is that they don't truly understand on a psychological level why their customer needs the product they're making.
Sarah: That's really interesting. And I loved the piece that the wrote about helping your father in terms of weight loss and the psychology of software being something that we have to almost be smarter than ourselves for, especially when we're trying to increase adoption. Along those lines of psychological benefits for software, is there any psychological benefit to receiving training before you buy the product, for instance, free sales of educating or harnessing psychology even before a sale is made?
Nir: We have to be careful about where we position anything that's effortful. In the four steps of my book model, the four steps are trigger, action, reward, and investment. Anything that requires the user to put efforts into the product, to invest in a product for a future reward, that qualifies as what I call this investment phase of the hook. And a lot of companies put that investment phase too early. They ask the user to do a lot of effort before getting the reward. And what happens with these habit-forming products is that they ask the users to do very little work to get the reward. But then after the reward is attained, that's when we ask the user to do a little bit of work.
So I think a big part of the future of education is experiential. It's the kind of products that you don't actually need a manual for, that you can just starting using, and in the course of using these products, you discover more and more about how they work, right? So you learn by doing.
I think that's a very powerful model to explore. That being said I think there is a potential hook around learning itself. And what we're finding is that we're drowning in information. We're drowning in content today. There's no scarcity of information out there. But what we're craving is to find is curation, is figuring out how to use this content in a way that provides a lot of value.
So I could see using education as the reward itself, offering free training, offering courses, and supplementary material to get people into your door in the first place. I think that could possibly be a very interesting hook to explore as long as it's done in these very simple to consume, easy bite size portions.
Bill: So I wonder if you can comment on this then because, on that issue of having too much effort upfront and the stakes, I think I'm seeing the light in training now, thanks to you. In training we're trying to do so much teaching or learning upfront that people can't handle at all and it looks like work and it seems overwhelming and it's no wonder training can fail.
Maybe we should only focus on one or two really important things at first and then have people learn things later and get the reward out of the one or two main things that help them get their jobs done with this new thing?
Nir: Right. I think that would be a great place to look, to think about the few things that I can do frequently. Typically, when it comes to these things, there's an 80-20 rule. That 20% of the features are used 80% of the time. And so if we can focus on the most frequent behaviors first and foremost, the things that for each person's job they would do frequently enough to form these habits, then we have a foundation, then we have a base. We can use this process of progressively revealing more once the user has these core behaviors. So stuffing a bunch of information down someone's throat and calling your job over might backfire.
Bill: Not that I've ever done that.
Nir: No, none of us have.
Sarah: Can you talk a little bit more about the motivation and ability trigger relationship in enterprise products?
Nir: Sure. Of those four phases of the hook, the trigger, action, reward, investment, the second step of the hook is the action phase. The action phase is where the behavior itself occurs. And this is this behavior that occurs with little or no conscious thought, something as simple as opening an app or logging into a website or scrolling through a page, all examples of these things that we do habitually, you do them with little or no conscious thought. And, of course, we build these behaviors on themselves, almost like the layers of a pearl. A pearl has deposited layer upon layer upon layer and this is how the brain learns complex behaviors.
We form the ABC so that we can decode words, so that we can understand sentences, so that we can enjoy novels. And so that process of building these complex behaviors is done one layer on top of another until these very complex behaviors that we do day to day become something that we do with almost no conscious thought, like driving a car today is something that we can do out of habit. We can talk on the phone. We can hopefully not look at our phone, but we can do all of these other behaviors. We can listen to the radio or a podcast while we're doing this behavior that is rather complex. It takes a lot of time learn. We can do it automatically. And so the way these behaviors are built is over time, through successive cycles through the hook in the phase you asked about, the action phase, behavior, anything you want someone to do, anything you want to do yourself, your customers, your spouse, your children, anything with behavior, comes down to three basic elements, sufficient motivation, how easy or difficult something is, and a trigger must be present. For any human behavior, we always have to have these three things, a trigger, sufficient motivation, and sufficient ability.
Now where most people focus, as I alluded to before, they focus on motivation. If the student isn't picking this up, it's because they're not motivated enough. If the software is too hard to use, it's because the user doesn't want to use it enough, right? They're just not motivated enough. Motivation can't overcome ability, but it's much easier for us to move ability before motivation, that many times just increasing ability, making the intended behavior easier to do, and by doing that make the behavior much more likely to occur. But, of course, that becomes a product design decision. That becomes a course design decision that when we train somebody, when we try and change someone's behavior, it's easy to blame their motivation. But really we should look at ourselves and figure out, "Are we making the intended behavior easy enough to do?"
Sarah: That's incredible. It's really helpful and I think it's also very important to be, like you said, self-reflexive on how we can actually fix our own piece of that, not just put it on our software learners. In terms of trends for software behavior. What are some that we should be looking out for as enterprise software companies are building for the future? What do we need to actually incorporate in order to address every aspect of your hooked model?
Nir: I'll be publishing about this in the next few weeks. I'm interested in companies that are designing with distraction in mind. When we're in a classroom setting, the reason that the classroom setting is much more effective than typically online learning is because in a classroom, there's a lot of things that make it difficult to leave, right? And a lot of things that make it difficult to not to pay attention, which is the social pressure of your peers and the room and knowing that to get out I have to cross by these people's desk. But there's these simple things that are designed into the environment that make it difficult to leave. If you look at why online education typically has awful completion rates, if you look at MOOCs and different online learning platforms, we were talking oftentimes single digit completion rates. Really, really poor completion rates. Part of the reason, not completely the reason, but part of the reason is that it's just so easy to leave. When I have one tab open, listening to some professor drone on and on in a little letterbox screen with a camera angle that never changes, that's just boring. And so that compared to Facebook, which is just a click away, or my e-mail is just a click away or Twitter is just a click away, it's just so easy for my to stop paying attention and do something else. So back to this ability over motivation.
What if I really want to take this class? The environment, the context of the way we make people learn online is really not conducive in its current format to completing these courses. So what I'm looking forward to and we're already starting to see some rumblings of this is companies that are designing for a distraction, as I mentioned earlier. These companies that understand how hard it is to keep people's attention online because it's so easy to bounce, to leave and do something else, so they've changed the core experience into something that requires... Again they've used the psychology of requiring paying attention to be something that you just can't not do. You have to pay attention. So one company I've been watching eagerly is this company Minerva, which is an accredited university. It's an undergraduate school where their online learning has no lectures, zero lectures. You will never watch a lecture on Minerva. Instead what they tell people to do is, "Go out. You Google the information. You go do whatever you need to do. Go to the library. Google whatever it is that you need to learn. There's no 101 classes here."
The online learning occurs in an interactive format, so it almost looks like a game show. Their interface almost looks like online jeopardy because you've got this one teacher who's almost like a television host and he asks his group of students, "Now, I want you for five minutes to do this problem and I'm going to check on you. You're going to share with the group. And I want this team over here, your job is to do something else and then you're going to share. And together we're going to answer this question. We're going to debate about it." And that professor can see online who's participating, who's interacting, who's not interacting, and use the Socratic Method in a pointed fashion to make sure that everybody is still engaged. And I think that's going to be the future of distributed education.
Bill: Interesting. On a related topic here is that many open source software companies are electing from the start to offer live onsite multi-day training courses. And you would expect that companies like these would be engaging online training because the audience is usually early adopters, and yet that's not the case. Many of them are live, online multi-day, they want to capture them, they want to put them in a room and give them the logo and indoctrinate them and shut the blinds. It's interesting. It's related. I think that people are seeing that trusting that people are going to learn your software on an iPad at the dentist's waiting room is probably not going to work.
"The biggest mistake that a company can make is not recognizing that in fact our habits are hackable." -Nir Eyal
Nir: There's something to that. One phenomenon that I'm constantly fascinated by is live conferences, which is related to the live classes that you mentioned a second ago here. If you think about it, whenever you go to a conference these days, you know when the videos are going to be online. They're going to be on Youtube, they're going to be on Periscope, they're going to be on Meerkat. If you wanted to find those talks, you could find the talks.
And yet, people want to go to these live conferences, I think, for two reasons. One I think is, of course, the social element of meeting other people that you can't meet online. But two, and more to the point, there's something about being forced to listen that people will pay a premium for. Even though it's something that they could do with enough willpower in front of Youtube, but they know if they don't go to the conference and sit in the middle row where it's difficult for them to get out, then they won't repeat the lesson.
Bill: There's a good time.
Nir: Right. It's exactly right. I think that people do recognize this either consciously or subconsciously, but there's a lot of value to putting yourself in an environment where doing what you don't want to do becomes more difficult, where I really want to pay attention so I'm going to put myself in a context that makes not paying attention difficult.
Sarah: That makes a lot of sense in your topic of designing for distracted learner, the distracted software user. That's another level of building success around that.
Sarah: This has been such an excellent discussion and love to know are there any concluding thoughts either of you wish to offer on these topics?
Nir: I'll let you go first there, Bill.
Bill: Well, I actually find this quite interesting. I would like to post a closing question for you, Nir, is that you focus your research on product design, of course. And so oftentimes we have customers that don't think they need training for their customers because they're product-focused and they're human-centered and they're such good designers that our customers won't need training. And I wonder if you think there is a world where products become really so good. No one ever took a Facebook training class, really. So I wonder if there's a world in the enterprise where an enterprise of 10,000 people wouldn't need training or wouldn't need formalized learning?
Nir: That's a good question. I think it's a function of how complex the behaviors are and how frequently they occur and also how well the software is designed. If we can provide this progressive... Facebook is a great example. As you mentioned Facebook, nobody ever took a training course on how to start using Facebook. It's because Facebook only asks you to do one thing at a time. If you're totally unfamiliar with Facebook, all you have to do is scroll your feed. That's the key. Open the app and scroll the feed. That's it. Then the next time you use Facebook, maybe you'll like something, maybe you'll comment, maybe you'll wonder how to post your own photo or comment, right? They reveal more and more and more of the product over time. Now where you will always need training is when somebody needs to learn something and quick. That process to get us to understand Facebook, there was no rush.
Of course if you're training somebody to use sophisticated software, if they don't know how to use day one, you're losing money, then of course you'll always need somebody to help him get up to speed.
Bill: Very good. I think I agree.
Nir: I'd love to know more about where you think of the future that training is going and what has it become?
Bill: I post that question because we really do get people that say, "I never needed training and so my customers won't either." I like to use that metaphor of the iPhone. Remember the old iPhone 3 commercials? They were all instructional. They were all, "How do you order flowers?" Except it was in the context of your life. You're about to forget your anniversary, no problem. You swipe over here and 10 flowers and you're a genius. How do you make a two-way phone call on an iPhone with two taps? They were all instructional and yet they weren't teaching. It wasn't about the features. It was about how do I make my life a little bit better? How do I balance the books for my boss by Friday and run that report later? Not, how do I use the report's tab in the software? More like training on doing the job better than it is on the features and too much software training is on features and not how do I actually do my job. That's where I want to help take us.
Nir: That's fascinating. So it's giving people information what they needed for functional purposes as opposed to just thinking that they can store a library in their heads.
Bill: Yeah, it's certainly a huge part of it for sure.
Sarah: Thank you so much to Bill and to Nir for joining us. It's been a pleasure. And everyone who is listening, please be sure to check out Nir's website, Nirandfar.com and get your copy of Hooked. Thanks again.
Bill: Yes, thanks again.
Nir: Thank you very much. It's my pleasure.