Continuous Discovery Habits, with Teresa Torres

Ep 27

May 01, 2024 • 55 min
Teresa Torres, author of Continuous Discovery Habits, joins to share lessons from her popular and acclaimed book. We talk about the benefits of constantly talking to customers, why it can be worth seeking delight rather than strictly building products in terms of problems, how to avoid choosing perverse outcomes and metrics, and how to get the truth from customers – despite their aspirational overconfidence.

Allen: Welcome to It Shipped That Way, where we talk to product leaders about the lessons they’ve learned helping build great products and teams. I’m Allen Pike and today’s interview is with Teresa Torres. Teresa is an internationally acclaimed author, speaker, and coach. She’s best known for her book Continuous Discovery Habits, which in only three years has already sold 100,000 copies and counting. She has also taught over 12,000 product people discovery skills through her business, Product Talk Academy. Welcome Teresa.

Teresa: Thanks, Allen. I’m excited to be here.

Allen: I am very excited for you to be here. I read your book a couple of years back and it’s changed a bunch of things about how I do product discovery, talk to customers, and I often tell little stories from the book to people when I’m talking to them one-on-one about this stuff. I’m excited to give you some chances to tell a couple of those stories and then go a little bit beyond some of the stuff that’s in there.

Teresa: Sounds great.

Allen: Before we get into those tactics, how do we do product discovery? What are some of the things that you’ve learned and some of the things that you’ve learned since publishing the book? I’d like to give you a little chance to recap the Teresa story so far. Just give people a little bit more context than my official bio read there. How do you like to describe the context of who you are and how you came to this topic?

Teresa: Yeah, today I work as a product discovery coach. That means I help product cross-functional teams. Typically, product managers, designers, software engineers, anybody else who’s part of the building team make better decisions about what to build. That sounds so simple, but there’s some key components to it. First of all, there’s this implication that the team that is building the product is the team that should be deciding what to build, which already that doesn’t happen everywhere. The second implication behind that is that it should be cross-functional. That doesn’t happen everywhere. We actually should be collaborating when making these decisions. Then how did I get here? I spent most of my full-time employee experience spanning roles. My very first job, I was a both front-end engineer and a designer. My second job, I think I started as the same thing, both a front-end engineer and a designer and then moved into a hybrid design product management role. From there I moved into a leadership role where I was managing a product and design team. At that company, I actually was there for five years and in a five-year time period, I went from their director of product and design to the CEO and managed that startup through the 08 economic downturn. That was one of the hardest things I’ve ever done, but I learned a ton. Then the last full-time job I had, I ran a product and design team at yet another startup. All the companies that I worked with were small. I was very much a zero to one product manager. What I saw everywhere I worked, and especially as I moved into leadership roles and worked with other executives, most product teams were just not spending enough time with customers. They weren’t building in feedback loops. A lot of executive teams were sitting in a room by themselves putting together the roadmap and giving it to product teams. I feel like this is a really hard way to be successful. We’re just guessing. It’s like throwing spaghetti at the wall. Maybe this will work. I think one of the advantages of digital products is we can actually measure impact. We can see what’s working and what’s not working. That means that we can start to look at how do we make better decisions about what to build and we can measure the impact of those decisions and then iterate. Eventually I got to the point where I realized if I was trying to do this one small startup at a time, I was going to burn myself out. I decided to leave the company that I was at and to zoom out and to look at the market and say, “How do we get more product teams working this way?” More teams building feedback loops on their decisions. That’s really where I’ve been focused for the last 13 years now.

Allen: I love it. You had this background in the zero to one spaces, which is also where I’ve spent a lot of my career. I’m curious, especially now as you’ve gone in, you wrote the book and obviously that perspective informed it a lot. Now having coached 13,000 product people, I assume you’ve also been working with a lot of people at larger companies. There’s been a bunch of different perspectives coming into the folks that you’ve been doing workshops with and I’m sure talking to you about the book. I’m curious, and we’ll get into some of the details in the stuff in the book but just context setting, have you found any particular type of organization, either large ones or small ones or rapidly growing ones or maybe ones that are feeling a little stagnant or tech software or non-software organizations where these approaches and the mindsets that you’ve been writing about and advocating have been particularly impactful?

Teresa: Yeah, one thing that I’ll highlight, a lot of the advice we read about in the tech industry is based on what the FAANG companies do, and that’s not the process I follow. I don’t want to disparage the FAANG companies. They’re obviously very wildly successful. But I think the challenge with basing our methods off of what those companies do is it really brings up survivor bias, right? We don’t know if those companies were successful because of the way that they did those things or which of the things they did led to those success or if they were successful despite those things. I feel like it’s really easy to fall in love with like, “Oh, Spotify is a hot company, let’s mimic what Spotify does.” Every organization is really unique and I think that what we need to be doing and especially as leaders, we need to be looking at given our goals and our context and our culture, what’s the right approach? When I started to think about how do I help teams adopt some of these practices? Build in some of these feedback loops, I really wanted to base my work on what we know from research, what’s evidence based. A lot of my work is grounded in decision-making research, problem-solving research. How do we collaborate well? Critical thinking research. When people ask me like, “Is your book good for startups or is it good for enterprises?” I’ve actually found it’s really broadly applicable, because I think that research is very broadly applicable. When we’re deciding what to build, it’s fundamentally decision-making. If we base our methods and tactics on what do we know about, what leads to good decisions. It doesn’t really matter if you work in healthcare or you work in banking or you are a direct-to-consumer retail company, there’s going to be differences, right? If you work in healthcare and banking, you have regulations you have to follow. Your organizational context is still unique, but we can ground our practices in these evidence-based tactics that I think are pretty universal.

Allen: I like that mindset. One thing that I noticed in the book that is a little bit, not to say the book isn’t broadly applicable as you say, but I notice that more so than some product books, you take an actual perspective on the fact that there is a new organization here and it may or may not fully agree with everything in the book when you start out. Sometimes books can a little bit fall into the trap of… Here’s some great stuff. If you just completely transform your entire organization to all buy into this simultaneously, then you can work it this totally different way.

Teresa: This is actually a huge narrative right now because Marty Cagan went on Lenny Rachitsky’s podcast and he called out individual contributors and said, “You’re responsible for changing your organization, but Marty didn’t really say that. What Marty said was, “If you want to be successful in the market moving forward, you need to start learning these new ways of working.” And I actually agree 100% with Marty. What I don’t think he said, but people thought he said was you’re responsible for transforming your organization. In fact, he said, “You can’t transform an organization without a CEO on board driving the transformation and an individual contributor is not likely to convince that CEO unless you’re at a small company.” What does that look like? Okay, I work at an organization where they don’t work this way at all. The CEO is not interested in transformation. How am I supposed to follow Marty’s advice of I need to learn these skills to be individually successful in my career moving forward? Well, one of the things I tried to do with the book is to say, “Look, these habits are easier to adopt in an organization where your leadership thinks this way, but even if your leadership doesn’t think this way and your organization doesn’t work this way, you individually have a lot of agency around the way that you work and you can start to adopt these habits on your own. And when you do that, even in an organization that does not care about these habits at all, you will get better at your job.” I think everybody, regardless of where they work, if they get better at identifying their own assumptions, even if they can’t test them, they’re going to get better at the way that they work. If they can find a way to get some exposure to customers, they’re going to be better at their work. If they can start to think about what outcomes matter to their organization, even if their leaders aren’t thinking and outcomes, they’re going to get better at the way that they work. What Marty was trying to communicate is you have more agency than you think. You can start to adopt these habits. It may never change your organization, but you can still change the way that you work.

Allen: When the worst-case scenario is you learn some useful skills and have more impact, which is then, of course, how you’re going to get your next role when people see like…

Teresa: Absolutely.

Allen: This person is having a big impact and they’re building stuff the customers actually want and maybe the rest of the organization is a challenge. I like that mindset.

Teresa: I’ll share that all of my full-time employee experience was that organizations, they were founder-led startups. If you’ve worked in a founder-led startup, this is going to sound really familiar, the founders weren’t terribly interested in talking to customers. They had a vision and they were way overcommitted to their vision of what the world should be and what the product should be. And I sat in a lot of meetings with executives that just made things up off the top of their heads and said, “This is what we should build.” I will tell you in every single one of those organizations, I still talk to customers. I still tested assumptions. I still worked to understand the business outcomes. I still worked the habits, even though nobody in the organization was telling me to do so. Nobody in the organization was defining outcomes for me. Nobody was saying, “Who did you talk to this week?” But I knew that if I wanted to do well at my job, I needed to have feedback loops on the decisions that I was making.

Allen: Yeah, I love it. That’s one of the things I look for in interviewing leaders on the show and who I want to learn from is try to find other leaders that do actually want to talk to customers and have their assumptions invalidated. Obviously, this sort of things that lead us to found companies are sometimes the things that lead us to hubris a little bit, so we try to have folks like yourself help us see the way. Early in the book, obviously you define continuous product discovery because that’s what the book is about. You talk about the importance of, as you’ve been talking about here, the people building the product, talking to customers at least once a week, conducting small research activities in the pursuit of a desired outcome and the default structure that you talk about in the book, which, I think, you’ve helped popularize this idea and certainly in labeling of it, of a product trio where you have a product, you have an engineering, you have a design leader, and they work together. A lot of orgs don’t have those, or at least not formally. Increasingly it’s more and more common, especially in tech, but especially really small startups. Traditional orgs don’t have that. What do you see as the minimum viable piece of the set team? You don’t want to be going this entirely alone. Obviously, as you’re saying, the habits can help even if you are… What do you see, especially as orgs? You’ve been talking to a lot of orgs that have been, I’m sure, experimenting with different ways of setting up… Some orgs are like, “We don’t have product managers anymore. Everything is this that.” People makes a blog post and everyone’s mad about it or happy about it. But what do you see as the minimum viable core of a product team to be able to benefit the most from this loop?

Teresa: I think the key is that we want a cross-functional approach, and regardless of what people you have available to you, you need people on the team who can think about feasibility. You need people with engineering skills. That almost always exists because we always have engineers. We need people that are thinking about usability, accessibility. Oftentimes that’s a UX person. I’ll say that I’ve worked at startups that didn’t have UX people. I played that role. You need somebody who’s thinking about viability. Is it good for the business? You need everybody thinking about desirability. Does a customer even want this? You need people thinking about the ethical implications of what you’re building. You honestly could have a product trio of one person if that person is well-rounded enough to think across all those different lenses. For most of us, that’s not the case, right? I’ve rarely met an engineer who’s good at also thinking about viability and usability and ethical. Now they exist, right? There are some engineers that are really phenomenal and love this part of the work. But like I said, I always spanned roles. I don’t want to discourage anybody from spanning roles. I actually think if more of us spanned roles, we would actually collaborate better together. A product trio is not about three people. It’s not about these exact roles. It’s not about excluding some people and including other people. We have these five areas we need to cover when we’re building a product, desirability, viability, feasibility, usability and ethical. We want to make sure that our products meet all of those five needs. How do we make sure that as we’re deciding what to build, those needs are being represented? And oftentimes that means a product manager is involved looking at viability. A design person is involved looking at usability. An engineer is involved at feasibility. Maybe they’re working together on desirability and ethical. Now here’s the thing, it’s easy to take away, “Okay, that means the engineer makes all the feasibility decisions, and that means the designer makes all the usability decisions.” I don’t advocate for that either, because what that leads to is a big battle between I want to do it this way because this is the most feasible and I want to do it this way because this is the most usable. Whereas really, we want everybody responsible for all five of those areas so that we’re truly collaborating on the most usable, most feasible, most desirable, most viable, most ethical solution.

Allen: I love it. We have this product trio in whatever form it is, if you’re super lucky and/or unlucky, it’s all in your head or you ideally have some collaborators that you’re working with doing this assessment loop, building these habits, which by the way, I always love anything that talks in terms of habits. It’s like very much plugged into my brain. Honestly the whole title, Continuous Discovery Habits like, “Hey, I love continuous processes, I love habits.” Discovery is like, “Well, what I’m all about. This is like catnip for me.” A big theme in your book is when we’re doing this thinking in terms of what you call opportunities rather than the language you hear more often on teams is they talk about problems like, well-functioning teams often will say, “We got to think about the customer. What are the problems the customer are really facing? What are we solving a problem here? You advocate actually stepping back one level to opportunities. What motivates that framing, and then what counts as an opportunity in your nursing?

Teresa: Yeah, I don’t know that I can claim responsibility for this. I think this opportunity language predated me. I can tell you why I chose to adopt it. It’s really easy to think about we’re going to solve problems, but the design world, the UX world started to talk about this idea of like, “We shouldn’t just solve problems. We all should delight our customers.” Right? I started to think about that in the product sense, and there’s a lot of products that are very successful that don’t solve a problem. In fact, we could argue they maybe even create problems. A few that come to mind, sports betting. I’m a big sports fan. I hate sports betting. I’m not sure that it’s helping sports. We just saw an NBA player get banned from the league for betting on basketball. That’s a product that people clearly love and are very addicted to. It is creating a ton of viability. It’s creating a ton of business value. I think it raises some ethical questions, right? It definitely doesn’t solve problems. It’s not solving a problem. Maybe you could argue it entertains me and that’s a problem, but I’m not sure that entertainment is solving a problem. I think what it’s doing is it’s satisfying a desire. And we have lots of products that satisfy desires, ice cream, Disneyland, most travel, luxury resorts. Lots of things, fast cars. These are desires. There’s lots and lots of products that satisfy desires. I wanted to move away from this idea of when we talk about solving problems we hyper-focus on removing friction and doing things that create utility. All of that matters, but it’s not the whole story. We also create things that are designed for pleasure and that entertain us and that really create quality of life that we’re not going to get from just solving problems. This is not a new idea, and I did not invent this idea. Around the time that I started coaching, there was already language circulating, especially around the UX world about opportunities, so I adopted that. This is second reason why I really like it. One of my goals with my opportunity solution tree, which is just a visual that helps teams chart their discovery work as they go. One of my goals with that visual, and this is actually the subtitle on my book, is to help product teams align customer value with business value. At a lot of companies, it’s at odds. Do we do this to support the business or do we do this to support the customer? What I want to see is teams do things that create customer value in a way that creates business value. When we talk about opportunities and not problems to solve, there are opportunities for the business. It allows us to create business value, but those opportunities actually represent customer value. I also felt the opportunity framing, there’re opportunities to create value for the customer in a way that drives business value. I felt that language represented that much better.

Allen: I zeroed in on this thing for the reason your point one, which is I have a background as a product person where I’ve often been the person pushing for more delight and thinking like, “This isn’t just about mechanically solving the problem. It’s possible for software also to be desirable and that people enjoy using it and makes them want to use it more.” Often that kind of work does drive business value and thinking in terms of opportunity, obviously, you can’t just focus on it 100% unless you’re making ice cream. If you’re making a SaaS B2B tool, you probably also need to solve a problem as well. But I like that opening up broader.

Teresa: This shows up in the design space too. I wish I could remember what company this was. Maybe you’ll know it, but I remember reading about a company that on their login screen, they had a little animation of a bear or some kind of animal. And as you typed your username, it moved its head as you were typing, and then as you moved to the password, it covered its eyes.

Allen: Oh, that’s cute.

Teresa: There’s no utility to that, right? But I would argue there is utility to that, because here’s what it does. It seems like it’s superfluous like, “Why would you spend time building that?” Well, this was a startup. I’m pretty sure it was a new product. I don’t know this brand at all. I don’t know if I can trust them. And from the very first experience of creating my account, they’re making me smile. That’s brand equity, right? There is utility in brand equity, and I think it’s so easy if we frame things as we’re just solving a problem. “Oh, we just got to get you to create an account.” That’s just utility. Well, they added a layer of delight, and I would argue that delight is creating business value.

Allen: Yeah, and I think there’s been a trend over the last few years of people making better and clearer arguments why in 2024 and beyond, just having a product that has these features is less and less enough to compete because the playbook of how to build a SaaS business, I’m assuming we’re in SaaS business. Not everyone is in SaaS, but the playbook of how to build software is becoming better and better known. Just the fact that you have a thing that does this thing, if you’re the only one who has that, yeah, maybe you have a competitive advantage for a few months until someone’s copying it. But in the medium and long term, execution is going to be of how you differentiate yourself. Those little things that even just demonstrating that you care and making people build that trust of why are we going to use this thing instead of this other thing? At the end of the day, at least some percentage of that is going to be like, “I liked it.”

Teresa: Yeah, sentiment definitely matters. What I love about this example is I often say people ask me, “Do I have to do all the habits for everything that I build? That doesn’t seem possible.” It’s not possible. What I say is you want to do more discovery for the opportunities that really matter for your business and you can do less discovery for the opportunities that matter less to your business. I often use the example of if you’re building a forgot password flow or a create an account flow, that’s probably not where you need to innovate. But it turns out if you’re a brand new company and you need to build trust, that might be where you need to innovate. I love that this company did this because it’s such a… You really, when you’re unknown, when you’re an unknown entity, you do have to do everything you can to get people through the hurdle of creating an account.

Allen: Yeah, I love it. One of the problems, we were talking about business impact and you’re saying trying to deliver this and align your way not to like, “Okay, we solved a problem.” But we’re actually helping the business, which is one of those obvious things that, of course, we all know, but then you often see teams just not do it. A problem I’ve seen in a lot of product teams is that they’re encouraged to have a business impact, and so maybe they’re tasked with some high level goal like reduce churn or if you’re really unlucky increase margins and they’re like, “Okay, we’re going to go increase margins for the whole company.” Which is, I guess, better than not having anything measurable, but it’s not very useful because they are going to be a small contribution to it. How do you think about metrics and what we measure and what we focus on as product teams and choosing the right things to focus on measurement wise and the right metrics for a given product team?

Teresa: I think the challenge with things like increased margin is not that we can only have a small impact. I think product teams could have a huge impact on margin. I think it’s that we don’t directly impact it. Let me talk through this. In the book I introduced this, again, not a new idea, but I think it’s an idea that helps people distinguish between types of outcomes. I distinguish between business outcomes versus product outcomes. A business outcome measures the health of the business. It’s those financial metrics. Every for-profit company is trying to grow a profit. The inputs into that are to increase revenue or to reduce costs. We can look at how our product makes money and we can generate a formula for how we would calculate our revenue. If we’re a subscription business, it would be number of customers times how much they spend each month times how much they stick around, right? Obviously it can get more complicated than that. That tells me these are the outcomes the business cares about. The problem is product people work in the product. What are they influencing? They’re influencing customer behaviors that happen in the product. Those are our product outcomes. The key is to create a clear connection between product outcomes and business outcomes. If I was tasked with increased margin. Okay, how do I increase margin? I need to increase revenue while reducing costs. Now I need to make a judgment on what side of that equation should I focus on? Maybe I want to focus on increasing revenue. Okay, now I need to understand how do we make revenue? We can acquire more customers. We can increase their average monthly spend. We can increase how long they stick around. Where do I think I can have the most impact? I’m going to focus on increasing how long they stick around. Okay, what’s a behavior in the product that I think is a leading indicator of increasing retention? And that’s my product outcome. I am focusing on increasing margin, eventually, but I’ve deconstructed it in a way where I can look at what’s the behavior or sentiment in the product that I can directly influence that is a leading indicator of those lagging business outcomes.

Allen: Right. Then that’s how we get to, like in the Netflix example, maybe the CFO cares about retention. The episode the product team is talking about, “Okay, well, we want to increase how many viewers love the shows.” Then you also make in the book the distinction between a product outcome. Something like, “Okay, we want people to have a better impression of this thing or this thing.” Whatever, and then the traction metrics, which might be something that we hypothesize is correlated with that like, “How many minutes of Netflix did people watch.” You see teams sometimes end up getting… In the case of Netflix, I think, they’ve maybe built an empire, and this is maybe a bit of your say confirmation bias of a FAANG company that Netflix focuses obsessively about minutes watched, and then now they’re so successful, so maybe we should also alter obsess over some traction metric. But you make a good case in the book, I think, for trying to also think beyond those micro metrics that maybe they’re the thing that drives the business value, but they’re going to have limits, they’re going to have blind spots and having things are a little bit more product oriented when you can.

Teresa: Yeah, let’s talk through this distinction between a product outcome and a traction metric. Typically, a product outcome is measuring a behavior in the product or a sentiment about the product. We can look at engagement as a product outcome. NPS is a product outcome. That’s a sentiment about the product. The key thing is for it to be a good discovery outcome, we want it to span features or if we’re on a feature team like the search team, it can be focused on optimizing a feature. What we don’t want to do is to pick a measure that’s measuring adoption of a feature. That’s what a traction metric is, and here’s why. Let’s just talk through this a little bit because I think this is often misunderstood. Let’s say that my business outcome is to increase retention and I work at Netflix. I probably do care about engagement and maybe the way that we measure engagement is by viewing minutes, average viewing minutes per week. That’s probably a pretty reasonable engagement metric. One thing I’ll say about engagement is that if you focus at engagement at all costs, we tend to get ourselves into trouble. See, Facebook, Instagram, we don’t like being addicted to products. Whenever I set an engagement metric, I like to pair it with a sentiment metric. I want to increase engagement without negatively harming NPS or CSAT, right? That’s one thing to keep in mind. Okay, let’s say our theory of engagement at Netflix is we’re going to increase your average viewing minutes. That probably has a ceiling, so we’re going to only increase your average viewing minutes. Maybe it’s we’re going to increase your sessions. There’s lots of ways. Number of sessions in a week. Maybe we’re going to increase the number of sessions in a week where you give us a positive rating on what you watched, right? There’s lots of ways to construct this to keep those in balance. Now let’s say we have 12 teams focusing on engagement. I want each team to have their own outcome. We need to deconstruct engagement, and one way to do that is on the viewing journey. Maybe we want to reduce the time it takes for you to find something to watch. That’s a great metric that’s going to feed into that engagement metric. If I can get you watching something faster, you’re going to watch more minutes. Maybe we’re going to get you to something that you like better. Now, I might have multiple teams focusing on reducing the time it takes to watch. I might have a search team, I might have a browse team, a recommendations team. It’s okay for me to set a metric around a single feature, like reduce the time it finds something to watch from a search start to watching. We’re making search better. What I don’t want to do, this is a traction metric, I don’t want to give the search team increase the number of users who search.

Allen: Yeah, horrible things can happen if you do that.

Teresa: Here’s why I don’t want to do that. What if I’m a perfectly happy customer that’s super engaged and I’ve never searched for anything because your recommendations are spot on and I always find something to watch. We don’t want to incentivize the search team to try to change my behavior. I’m happy. It’s very insidious. It’s a very subtle framing. We really want to make sure we’re not setting outcomes that incentivize our teams to encourage our customers to do things they don’t want to do.

Allen: We’ve all been thrashed in products with that pop-ups, trying to get all these different teams competing for your attention. I will be thinking about that next time I see one of those pop-ups. You’ve written this book, it sold 100,000 copies already. It was only in 2021 that you published it and it’s going strong. You’ve given workshops, you’ve coached more than 10,000 people on this stuff. I’d love to hear a bit about the parts of the book that you’ve gained new perspective on in the years since. Whether it’s things that you are seeing people misunderstand. Things that you wish you had gone into more detail on. Things that you would maybe walk back a little bit and qualify people who are overapplying certain parts. What comes to mind in terms of things you’ve learned since?

Teresa: First of all, I want to clarify some of the numbers because I don’t want to misrepresent anything. I have probably coached in the lower hundreds of teams. I’ve coached for years. I’ve worked with a lot of teams, but it’s in the hundreds. The thousands number, that 13,000, I think it’s actually now closer to 15,000, is the number of people who have gone through our courses at the Product Talk Academy. Courses are a little bit different from coaching. I don’t want to over-represent that, but all of our courses are very hands-on. There’s lots of interaction. I do feel like I have worked with thousands of people, but coaching is a more high-touch activity. I just want to distinguish that.

Allen: Sure.

Teresa: Things in the book that if I was writing the version two that I would focus on, some people took the product trio idea to literally, so a lot of people are upset with me that I left out their roles. I probably would clarify that. I have done a lot of clarification of this in the way that I talk about it. I’ve written clarification blog posts. I wish that made it into the book. I actually tested the book content a ton and I’m surprised this didn’t come out in testing, but too many people think the product trio idea is rigid and it’s not. The underlying principle is cross-functional collaboration. You can make up a trio, a quad, quint. Once you’re getting more than five people. It’s really hard to collaborate, but generally use your judgment on the best rules. It does say this in the book, but I get a lot of flag for this, so I probably would’ve done a better job on that.

Allen: Yeah, I did just reskin that part before and then I still did ask you the question in this interview. It isn’t strict, right? And then I contribute to that. Sorry.

Teresa: I think this is maybe an area where the book could have been stronger. I think when you define something as these three people and you call it a product trio, people are going to take you literally, and I did add a caveat of you can adapt this and customize it, but I think what they’re going to remember is how you defined it. Where I’ll defend myself is I do think in an overwhelming majority of companies, your product trio is going to look like a product manager or designer and a software engineer, but I want everybody to have license to adapt it to their own context. If you’re working on a data product and you have a data scientist full-time on your team, you probably have a quad. If your company is embedding user researchers on your team and you have a full-time user researcher working with you, they’re probably on that squad with you. And there’s a lot of variation in flavors of this. I think I would emphasize that a little bit more in the book. I tried really hard, especially in that last getting started chapter where I give advice on how to adopt the habits to highlight that you don’t have to adopt the whole framework to get value from the habits that you really can work on one at a time. I don’t know if this is something I would change in the book or maybe there’s a need for a workbook as a follow-on, but I see way too many teams trying to make wholesale changes and then they fail and that’s really hard. I really want to encourage people to work one habit at a time. Big changes, hard. Small changes, much easier. That’s another area where I see people struggle. I think as part of that, just emphasizing more, you don’t have to do every habit on every single thing that you build, that these are tools in your toolbox. You can use judgment.

Allen: It’s interesting that you say that, and maybe this comes in teams that have spent less time on organizational change and organizational design, but one of the things I appreciated in the book was, your intent came through to me at least that’s like, “Here’s a bunch of tools and then here’s a toolbox, and if you apply some of these tools, each tool will probably be helpful to you and you don’t need to do it wholesale for it to work.” That’s, I think, part has been the success of the book in contrast to sometimes you’ll have a book that’s a iconoclastic and say, “Everything you know is wrong. Start from scratch with this totally new thing that has nothing in common with what you’re doing, and if you don’t do it this way, you’re doing it wrong.” Which is not the way that the book is framed. It allows for you to build these skills, because there’re skills. There’s a whole bunch of skills in the book. There’s a whole bunch of habits. You’re not going to form all of them at maximum effectiveness the first time you try it. I’ve been pulling them piecemeal. Anyway, that’s interesting to hear that.

Teresa: Let’s contrast it with Sprint. God, there’s two authors and I’m only going to remember Jake Knapp. Are there two authors? I can’t remember. I think their next book has two authors. I think Jake Knapp wrote Sprint. If there’s a second author on Sprint, I apologize. I’m not trying to leave somebody out. That’s always a bummer when you’re the second name on the book. Shout out to Josh Sidon on that one. Here’s what I love about Sprint and why I think it did extremely well. It gives you a very explicit process to follow. It’s constrained to one week. Literally anybody can buy that book and they can do it. It doesn’t require a lot of organizational change. It requires that you can get everybody in a room for a week, which isn’t always easy, but you can do it and it literally gives you a recipe that you can follow the exact recipe. That’s phenomenal. People want that. But I think there’s a problem, and I think even Jake would agree with this, you’re not going to run Sprint 50 weeks of the year. What are you doing the other weeks of the year? And that’s the book I tried to write is what are you doing the rest of the time? The problem with that, there is not a set process. You have to adapt it to your organizational context. It is a collection of tools. It is a collection of habits, but I do think there is a non-trivial subset of people that really want a process that they can just blindly follow.

Allen: Yeah, it’s interesting when you write a book that you have to make decisions like that. Then when the people are deciding which book to buy or read or focus on, people are often not self-aware about either what they’re looking for, or what their organization or the way their mind is set. I know my mind. I’ve learned about myself and now that I make career, that I do really well wit… If a book has a whole bunch of ideas and then I can pick and choose them and then you apply them to improve myself, then I do better with that. Then here’s the formula and print out this thing and follow it exactly like I have a little bit of a founder allergy to being told exactly what to do. Maybe that’s why this one clicked with me.

Teresa: I do too, and that’s why I didn’t write a book that said, “Here do this step by step.” Some people interpret my book that way. I get emails from people that are like, “Hey, I did all these steps for this really micro thing and I just replied like, “Why?” And they’re like, “Because your book told me to.” And I’m like, “It didn’t really tell you that.” Right? I think some of it is maturity like, “Have we developed our own practice and our own point of view?” And that’s a totally normal part of growth, so I don’t want to pooh-pooh that, right? Some of this is us just maturing our own toolbox and our own mindsets. But I also think if you really want to get good at discovery, you have to work on your ability to manage uncertainty and ambiguity. And a lot of that is not having a step-by-step process. It’s being willing to react to what you’re hearing back from that feedback loop and then matching the right tool based on what you’re hearing.

Allen: I love that observation, that being able to deal with uncertainty and ambiguity is one of the core things that really you see people build as they go in their product journey from someone who’s very junior, is just at a university. They’re almost always allergic to uncertainty and ambiguity, and they want to remove it as fast as possible so they can have a plan. Everyone’s uncomfortable with that most of the time when they start out. And then as people get better and better at that, they’re going to be able to do better at product. And, of course, a book that gives you a bunch of tools without explicitly saying you have to follow this plan, in their mind when they’re earlier on that path might be a little more likely to say, “Okay, no, this is the plan. I have to follow these exact steps in this order.” One of the things I wanted to do is just… I’ve actually been really enjoying doing this off-topic, or at least I hadn’t really planned to go in as much into the meta stuff of the book, which I do enjoy. But I want to dig into one specific thing that I found really useful and click with me and I often will tell people about this aspect or this bit of the book. You talk about the idea of confidence in how you have this really great quote from Daniel Kahneman, which is now one of my favorite quotes about how people are confident, not when they’re correct, but when they have a coherent story that they tell themselves.

Teresa: Yeah, I love that quote too.

Allen: As soon as I heard that, I was like, “Oh man, that’s totally true. And that’s why people lie by accident.”

Teresa: This quote… One of the most depressing things about the book Thinking, Fast and Slow, first of all, it’s a really long book and it’s not the most… I love Daniel Kahneman’s work. It took me work to get through the book. I love biases and learning about them, and still it was a slog. I get to the very end, and one of the things he says at the end is like, “Even knowing about these biases, doesn’t protect you from them.” And it’s such a like, “Oh, why did I just do all those thing?”

Allen: Oh, man.

Teresa: And it’s part of that same quote like, “Even being aware of these biases, this is not how you overcome them because your brain will still construct a coherent story and it’s going to feel like truth.” I think this is a big reason why we need to do discovery together because somebody else can see your bias way better than you can.

Allen: Sure. Yeah. That was brought home for me when I started to apply some of these interviewing things. You tell a story in that section about doing customer discovery at a workshop on someone who is buying jeans and how they make that. Can you tell that story? Indulge me.

Teresa: Yeah, this used to be my go-to trick in in-person workshops is I would ask someone in the audience to volunteer. I would ask for a woman to volunteer because this is more common for women than men, although I have seen the same effect with men in this topic. All men and women have the same bias, but in this particular example, it’s safer to choose a woman in the audience. I always ask, “Is there a woman in the audience who’s willing to volunteer?” And then someone volunteers and I ask them, “What’s your criteria for buying a new pair of jeans?” The reason why this is more relevant to women is almost every woman on the planet is going to say fit. It’s because with hips, it’s really hard to find a pair of pants that fit women well. Some women have big hips, some women’s are straight legged, almost more like men. It’s a thing, and it’s such a softball question, I know the answer is going to be fit and the vast majority of time it’s going to be fit. The story I shared in the book was a very specific instance. I asked a woman, “What’s your most important criteria for buying a pair of jeans?” And sure enough, she said, “Fit.” Then I said, “Tell me about the last time you bought a pair of jeans.” And she said, “I bought it on Amazon.” And I said, “How did you determine fit?” And she said, “I bought a brand that I’ve previously bought.” And I said, “Have you ever purchased the same brand and they fit differently?” And she said, “Yes.” And I said, “Okay, well, help me reconcile your answers. Why did you buy this pair of jeans with the risk that they might not fit?” And she said, “Well, they were on sale and I didn’t want to have to go to a store.” Okay, what came out of that last answer? She has other criteria besides fit. Convenience of online shopping, saving some money. That doesn’t come out in the direct question. This is straight out of Amos Tversky and Daniel Kahneman’s work. When I ask, “What’s your criteria for fit?” Your brain doesn’t want to do the work. I’m going to get a fast system 1 response. And this response is coherent. She does care about fit, and she believes that is her answer, but it’s not a complete answer. I need to get her to slow down and to do a more system 2 deliver a response. When I say, tell me about the last time you bought jeans. She has to stop and think about it and consider what actually happened. That’s getting me away from the cognitive biases that tend to come up in that system 1 response.

Allen: Yeah, it’s both a more thinking, but then also it removes the story the person tells themselves. Even in the case of jeans, it’s not as emotionally loaded, but just a few days after reading that passage, I was doing some customer interviewing on some leaders about how they manage their time and attention. And if you talk to a senior leader, they’re going to have this mental model that they’re really good at managing their time and attention. “Oh, I have the system.” Okay, tell me about your system? “Well, every Friday I go through and I check everything off and I make a plan for next week. I do this routine where I have my five priorities for next week. I’m like, “Okay, great.” And then I was thinking about maybe making software to help people do this routine. Then I started after reading this thing, I said, “Oh, well, tell me about last Friday. Where did the plan?” “Oh, I didn’t do it last Friday.” “Oh, well, what’s your to-do list right now?” “Oh, it’s a garbage fire.” Like, “Oh, interesting.” The thing they tell themselves, it’s consistent. Had not really that much to do with reality.

Teresa: We tend to think of ourselves through this like really rosy aspirational lens. And if we interview people asking where our questions don’t get around that filter, we’re going to build products that people aspire to use but don’t use. In fact, gyms, their whole business model is based on this. They designed a business model that gets you to buy based on your aspirational self, and they’re okay with that because they don’t want you to use the gym. The fewer people that use the gym, the more gym subscriptions they can sell, the more money they make, right? There are products that take advantage of this, but most of us want to build products that our customers use and actually get value from. We need to get around this rosy filter. And I do want to highlight again before people bash me. That example that I gave is not specific to women. In fact, now that skinny jeans are a trend for men, I suspect more men are going to have the same experience. But I can give other examples. I can ask you if you’re a really physically active person, I can ask you how often do you typically work out? You’re going to give me an answer. Then I can ask you what happened last week? And it’s almost always going to deviate. The one exception is I do know there are some runners that literally run every day and they’re not the best examples to show this effect. They’ll be like, “No, I ran every day.”

Allen: Every day is one of the few habits that actually tends to be pretty…` For a little while I was exercising every day, not because I need to or that’s the best routine, but because it’s a habit I didn’t think about and I was less likely to fall off from.

Teresa: I exercise every day for exactly this reason. This was a huge aha for me. I used to try to do strength training x days a week and cardio x days a week, and I’m a really active person. A lot of my hobbies are activity, but I was struggling to get resistance training in on a regular basis. The way that I fixed it is I do resistance training every single day, but my rest days are by movement. Instead of doing three full body workouts in a week, I take those same movements and I spread them out across seven days. And it’s had a huge impact on consistency.

Allen: Habits. They’re amazing when you can form them. We know we want to make sure that we’re asking questions that make it hard for people to lie to us. What are some of your favorite hacks. Many call them hacks, but approaches to cut through people both their cognitive biases and their willingness to just rapidly give us an overconfident answer?

Teresa: Yeah, I want to be clear. I don’t think people are lying to us. This example of the women with the jeans, her answer wasn’t wrong. She does care about fit. It wasn’t complete, right?

Allen: My interviewee was lying about their system.

Teresa: Well, he wasn’t lying.

Allen: Not the way we mean lie.

Teresa: Lie. I think lying requires an intent to deceit, to be deceptive, right?

Allen: Yes, absolutely.

Teresa: I think the challenge is we’re aspiring to live to a standard. We literally wouldn’t hold another human being to, right? We have to understand that, and then we have to ask questions that counteract that. I believe that product management, and I don’t mean product managers, I mean building products, so all the rules involved is fundamentally about behavior change. You’re trying to build products in a way that influence your customer’s behavior. And in order to do that, we need to understand past behavior. When I’m interviewing, I’m structuring my interviews to understand about past behavior. My recommendation is really simple. Ask the person to tell you a story about a specific past instance. Tell me about the last time you watched Netflix. Tell me about the last run you went on. Tell me about last week. Tell me about your last meal. Now this sounds so simple, but collecting a story is a skill that requires practice, right? To get all the detail to help the person remember. In fact, I get a lot of people say like, “This is not an effective technique because people don’t remember their stories.” Well, it turns out there’s a lot of research behind this. As the interviewer, if you understand how memory works, you can help the person remember and remember reliably. The interviewing techniques are really important that we invest a little bit in our skill. I think one of the easiest things to teach people is how to collect a specific story and then it does take practice.

Allen: Yeah. Is there a quick tip people can use? They’re going into their interviews, they’re bought into this idea, I’m going to ask people about a certain time, last time or whatever to try and help coax people towards actually remembering rather than just being like, “I don’t know.”

Teresa: Yeah, the main thing with helping people trigger memory is we store memories in two ways. There’s the visual component of our memory, and then there’s the conceptual component of our memory. If I ask you what criteria do you use to buy a pair of jeans? I’m tapping into that concept, that conceptual memory. If I ask you to tell me about the last time you bought a pair of jeans and I help you situate you in that context, where were you? Visualize the scene? What were you doing? Now we’re tapping into that visual memory. A lot of collecting a story is helping that person go back to that moment. I’ll tell you, we have a course called Continuous Interviewing. It’s a practice oriented course. We teach you how to collect a good story, and I do a lot of observing of people collecting a story. I was observing an interview. Students are just interviewing each other as practice. They were asking about, tell me about the last time you used Instagram. Now here’s the challenge with the product like Instagram, you do it all the time.

Allen: Yeah, without thinking about it even.

Teresa: The last time it’s sort of hard. The interviewer said, “Tell me about the last time you used Instagram?” And the participant said, “It was today at lunch.” And the interviewer said, “Well, what did you do?” And he goes, “I just scrolled through my feed.” And the interviewer said, “Well, tell me about it. What did you see?” He is like, “I don’t remember.” And they went back and forth. I jumped in and I said, “I’m going to interview him and model the difference.” Let’s say that you were the inner participant. I would say, “Allen, you are on Instagram today at lunch. What did you have for lunch today?” Right?

Allen: Right.

Teresa: That’s not relevant to the story I want to collect at all. What I’m doing with that question, I’m trying to situate you back in that moment. He had pork schnitzel. Okay, great. “How did you make the pork schnitzel?” “Oh, it was leftover. I took it out of the fridge, I put it in the microwave, I warmed it up.” I’m asking for this level of detail because I want to teach him, this is the level of detail that I want. It’s also helping to situate him in that physically embody that moment. “Okay, you’re standing in front of the microwave waiting for your food, what do you do?” “Oh, I got on my phone.”

Allen: Yeah.

Teresa: Okay, great. Now let’s get into what’s going on in your phone. And it’s such a small technique, but this is tapping into the visual memory of what happened earlier in the day. And we got to the point where he could tell me he could remember the exact posts he saw on that feed. And it’s easy to think like, “Oh, this isn’t a reliable interview. I need to have people open their feed and talk me through it.” Sure, I can do that too. But a lot of the skill of interviewing is learning how do we help people have reliable memories? There’s a lot of research behind this because who cares about reliable memory? Detectives and eyewitnesses, attorneys, right? Reliable testimony has high stakes impact, which means the criminal justice world has been studying this for a really long time.

Allen: Yeah, I love that. That’s like this thing of you go to your childhood home and then suddenly you’re like, “Right, I remember that.” And remember when we were here and the smells. Suddenly all these things come back. You’ve had success with this book, obviously, and you’ve been continuing to learn. You’re talking to all these people about this thing. Are you getting to the point yet where you’ve got the little tickle of like, “Maybe there’ll be a next book?” And if so, is there anything that’s catching your curiosity right now of things that you’re wanting to learn more about?

Teresa: Yeah, couple of things. I’m really focused on not just… Here’s a book, here’s some habits, but how do we get you to change your behavior and to adopt the habits? In addition to the book, we have a whole bunch of online courses that are all practice oriented. If you’re interested in those courses, you can learn about them at learn.producttalk.org. There’re all… Like our interviewing course. Our students interview each other. They learn techniques for collecting a story. They practice. One of my goals is to help create practice for people. We do that today through our courses. Our courses work for people in the US and Western Europe. They fit the professional development budget of people in Europe and people in the US. They don’t work so well if you’re in Brazil or India because they’re 800 bucks a pop, right? Most of us in the US, we go to a conference, we’re used to paying $800 for the workshop the day before the conference. Our professional development budget is generally like 1,000 to $2,000 a year. It’s in the realm of what we can pay. There’s a large part of the world where that’s not the case. One of my goals this year is to focus on increasing the accessibility of what we offer. I’m playing with two different ways to do that. One is we’re starting to explore, can we provide ways to practice through self-paced courses? This is a hard nut to track because most self-paced courses, what people don’t tell you is they sell well, but nobody completes them.

Allen: Yes.

Teresa: And I don’t want to take your money if I’m not going to actually help change your behavior. I make a good living. I’m going to be fine in retirement. I’m not just looking for more things to sell. I really want to find in an effective way to get you to change your behavior. One thing that we’re experimenting with is can we create self-paced courses that people complete, that has that learning outcome we’re looking for, which is you changed your behavior. We’re looking at that for all of the courses we currently offer. Can we create interviewing practice that’s at a lower price point for people that can’t afford the current course? Then the other thing I’m looking at is books to do the same thing. Some people are really good at reading a book and applying it. If they had a book that went even deeper on how to conduct a story-based interview, some subset of people will just go practice on their own and they’ll be fine. Some subset of people might need more guided practice, and that’s where this self-paced course might come from. And some people might need a more hand-holding, and that’s where this co-work-based course that we run would fit in. That’s what we’re looking at. I am not far enough along on anything to say this is the thing that’s coming next. The book was amazingly successful, and that’s really fun. And I did a ton to promote it, and it’s led to a lot of work. I think people underestimate. They think the book is done and now you’re done. But really it was like you put this big effort and then there’s a lot that follows it. I’m also just recognizing that I need to slow down a little bit and take care of myself in that process. I’m not really ready to commit to like, “I’m doing this thing next.” But those are some of the things that I’m exploring.

Allen: No pressure. Well, I appreciate the book, everything that you’ve been doing. You coming on the show. Thank you so much. Where can people go to learn more about you and the work that you’re doing?

Teresa: Yeah, I blog at producttalk.org. You can also find our online courses there. We have a community where the purpose of our community is to help people support each other as we’re putting the discovery habits into practice. Then I am on all the socials, like literally all of the socials. We’re on Instagram, TikTok, LinkedIn, Twitter. I just stopped posting on Threads. It’s just too many channels. Primarily, I’m on LinkedIn. I feel like with the changes on Twitter, the momentum right now is around LinkedIn. Yeah, I would say start at producttalk.org.

Allen: Awesome. Well, we’ll get all of those linked up in the show notes. Thank you so much for being on the show. It Shipped That Way is brought to you by Steamclock Software. If you’re a growing business and your customers need a really nice mobile app, get in touch with Steamclock. That’s it for today. You can give us feedback. Follow us on social media. Rate the show by going to itshipped.fm/contact. And until next time, keep shipping.

Read More