Episodes

Rob Haisfield on Roam

“Roam really makes it powerful to connect the dots between ideas and between projects”

My guest today is Rob Haisfield. Rob is a behavioral product strategy and gamification consultant. He’s also an expert user of Roam, “a note-taking tool for networked thought.” In this conversation, we talk about Roam — what it is and how it can help you think more effectively.

Listen to the full conversation

Download episode 43

Show notes

Read the full transcript

Jorge: So, Rob, welcome to the show.

Rob: Thanks for having me, Jorge.

Jorge: It’s a pleasure having you here. For folks who might not know who you are, can you please introduce yourself?

About Rob

Rob: Yeah. So, I’m Rob Haisfield. I am a behavioral product strategy and gamification consultant. I run a consultancy independently around those two ideas, behavioral product strategy being, how can we build our products better? So that way they influence users to use the product better. You know, because ultimately, product can only do so much for a person’s life. Like it can give you access to value, but people still of need to use the product in certain sorts of ways in order to gain value.

I also do gamification, which to me is essentially just taking inspiration from behavioral science and from game design and applying it to software products to encourage voluntary behavior. That means I end up working on projects, like how to increase adoption for an app, improving onboarding, or coming up with a strategy around retention for users and through the act of influencing user behavior.

And I also work as a behavioral product strategist for a startup studio called Spark Wave, which is a startup that starts startups, for lack of a better phrase. We come up with business ideas, build the first version of them, and then we find outside entrepreneurs to take over and run the company with our assistance and our support. I do the exact same thing in both gigs, it’s just with the startup studio, it’s with portfolio companies, of course.

Jorge: And what is your background? How did you come to behavioral product strategy?

Rob: Yeah. I studied behavioral economics in college. I was absolutely passionate about it, but when I was studying it… My school, we didn’t have a default major for it; they actually had me create the major. And so that meant I had to combine courses from a lot of like different domains that attacked behavioral economics from like a sideways perspective, you know? And then I kind of had to put the pieces together from there.

I also worked for a little while as a research assistant at the Center for Advanced Hindsight in Durham. And I’ve been working with startups for pretty much, as long as I’ve been working. Every internship that I had was with a startup; I had my own startup. And as I was studying, behavioral economics, and as I was working on startups, I was really just realizing the cross — the intersection — that’s really there. Because ultimately, we’re trying to influence people’s behavior in order to use products better. Most product problems just come down to, why aren’t our users doing this, you know?

I’ve just been combining behavioral science and products for a while. When I graduated college, I had to make a decision for myself. Do I want pursue academia, PhD, or do I want to go straight into consulting for startups? And what I ended up coming to was that consulting would be a better path for me to learn and would be a better path for me to make an impact, because I can still read all the papers that I want on my own. In fact, I’m reading more papers than I ever was before, because now it’s driven by curiosity as opposed to a professor giving me a syllabus and telling me this now. And I’m learning also more because I’m having to apply my ideas to the different companies and see, okay, does this actually work? Am I willing to put enough credence into this idea to bet my relationship with a client on it? Over the course of the last couple years, I’ve worked with over a dozen startups. Well, really just over a dozen companies. Not all of them have been startups and, I’ve learned so much, and I’ve been able to make an impact through their products.

So that’s really where a lot of my background comes from, started in behavioral economics, branched out into, I’m just going to read whatever I’m curious about within behavioral science and ended up in now I’m seeing those ideas in application.

About Roam

Jorge: That’s such a rich area for conversation, and I wish that we could go down that path. This idea of modifying user behavior towards particular goals… there’s a lot there to explore, not the least of which is the ethics of all of that. But instead, I want to focus on something else here, which is related to what you said at the end there, where you said, “I’ve learned so much.” I think it’s fair to say that you’re an expert user of a tool called Roam, which as I understand it, is where you are capturing a lot of this learning. And I was hoping that you would tell us more about Roam, how you use it, what it’s for, et cetera.

Rob: Yeah, so at a very high level, it’s for tracking my thoughts through time, and acting on them. So, it’s like one of my worst-case scenarios that I could imagine for my career is, like I’m really passionate about what I do. I think it’s one of the richest areas of study out there. We’re never going to get fully at the bottom of why people do what we do. There’s always going to be more to learn. And I can see myself still wanting to be applying behavioral science to products 10 years down the line. What I don’t want to happen is, I don’t want it to be the case that 10 years down the line, that I’m working primarily based on recent memory, that my thoughts, my ideas, my questions are primarily recent memory. And that’s what happens when you have a note taking system that you never reference or when you never take notes, right?

And Roam is incredibly powerful at disconnecting your thoughts from time. By really prioritizing the connections between notes and between ideas, then it makes it so I have faith that 10 years down the line, I’m still going to be leveraging what I’ve learned in my first two years of consulting because I’ve worked with over a dozen companies. I’ve worked with 13 products in the last two years, and Roam really makes it powerful to connect the dots between ideas and between projects. I’m able to learn interdisciplinarily within my own work.

That’s pretty much what I use Roam for. I like that it’s built around making connections. And I like that it’s built around creating a lasting knowledge base for yourself, and really turning it into a conversational partner. Because once you get how Roam knows how blocks relate to each other — like you indent a block under another block in order to say, this idea builds on that idea or this idea breaks down this idea, and you can also put multiple page references within the same block to say, mental models and onboarding, and a feedback loop in game designs are all related to each other, through this series of sentences that I’ve written.

And then you can query Roam. Like you can actually… like an Excel database, you can query it. And so, I can really talk with it. I can ask it questions and because I work with the way that Roam knows how blocks relate to each other, then that means Roam knows how to answer my questions. So that means if I am writing a note about you, Jorge Arango, if I’m writing a note about you, and in that note, I also relate that to information architecture and I relate that to podcasts, then that means later I’m going to be able to find the note where all of those ideas are connected. I’m going to be able to find the note where Jorge and information architecture are related. I’ll be able to find the meeting notes from the time that I was talking with a client of mine about a certain subject.

And so, I can ask Roam essentially, what have I written about this? What have I written where this connects to that? Or doesn’t connect to something else? And so that really allows me to do some powerful things, like I’m taking in so much information from so many different sources, that’s all segmented, right? Because I’m working with multiple clients at once, I might have meeting notes with one client, I might read a paper about a subject that’s relevant to a project that I’m working on. And I’m able to just through queries, compile the information that I need in order to support my present work. It means that my past thoughts are connected to my present work.

Jorge: I’m going to take a step back here because folks listening to this might not have seen Roam. Let me try to describe it… This is one of those “face made for radio things,” right? And I have to say, I have not used Roam; my understanding of the platform is based mostly on things like YouTube videos and the site itself. Note-taking is a kind of linear activity, right? I have a thought and I put down that thought and then I have another thought. And by the nature of the medium in which I’m taking notes, which is usually either paper or something like OneNote, those sentences get stacked in the order that I wrote them. And Roam presents an interface that is primarily based around the idea of an outline, where each one of those items is — I think you use the word “block” — but it’s like its own object. And those objects can have connections to other objects that you’ve written previously. And those connections establish backlinks behind the scenes. So that enables the surfacing of relationships that might not be obvious at first… during your first pass, when you’re capturing the notes. Is that fair?

Rob: Yes, absolutely. A concrete example is, I might ask myself, how does behavioral product strategy relate to gamification? And I can see every instance in my notes where those ideas have been connected. Because it’s all based on these blocks as opposed to full pages, it lets me really get granular. It means that I’m not constricted to this linear order that you described, because I don’t know about you, but if I’m writing for 20 minutes and just getting all of my thoughts down, I’m not going to be on one subject for the entire time, right? And Roam really lets me get granular in talking about this relates to that, and this is what I’m looking for in particular.

Public repositories

Jorge: And I think that we’re talking about it in the context of taking notes for ourselves, but I know that you’ve also made part of your Roam repository public, is that right?

Rob: Yeah. Yeah. My public Roam repository… it was an experiment, right? I wanted to see if I could write nonlinearly in a way that other people could follow because I’m really fascinated by information search behavior. Like, how is it that people find what they’re looking for? And one of the challenges that I’ve always had when I’ve gone to blogs before, is I don’t always know what to look for when I get to the search bar. I can either scroll down linearly, which takes a pretty long time to find an idea, or I can go into the search bar and search for something specific. But if I don’t know, what’s in that person’s blog, how am I going to know what to even put into the search bar?

With my public Roam, what I wanted to do was give people a starting page. And then from there, everything should just be connected in the right ways, so that way people can find what they’re looking for, even if they don’t know what they’re looking for. I want people to come onto my public Roam and find exactly the thoughts that are most relevant to them. And what was really interesting to me is that it seems like this is working to a degree. Like someone messaged me once, she’s another behavioral science practitioner, and she messaged me saying, “Hey, I found your page on why you chose to go into consulting versus pursuing a PhD, and I’m also just now considering going into consulting.” Right? And then there was another person who came onto it, wanting to know how to integrate with Roam for their product. And they ended up coming to my page on how Roam knows how blocks relate to each other. Someone in the spaced repetition community got to my page pretty quickly through a few clicks on the rogue-like genre of gamification as applied to flashcards. So, yeah, it seems like it’s working.

That to me, I think, illustrates a lot of the power of Roam very well. Like, my public Roam is of course, much neater and much more intentionally designed than my private Roam is. My private room is an absolute mess. It’s chaos that I’ve wrangled to a degree, right? That’s what my private Roam is. But really, what Roam lets you do and part of what differentiates it from every other knowledge management app that I’ve found is that it really designs itself in a way that you’re going to find what you’re looking for, even if you don’t know what you’re looking for.

Emergent structure

Jorge: That’s fascinating, especially in the context of this idea of designing for encouraging or discouraging particular behaviors, right? Which is something that in my mind, at least calls for very careful consideration of structures. I recently revisited the book Nudge, and I think the example that that book opens with has to do with the way that foods are laid out in a cafeteria and how the order in which food items are presented will influence the diets of the people using the cafeteria, right? And with a system like Roam, it seems to me that it is much less about creating such intentional structures and much more about enabling the structures to emerge organically from the act of building the corpus. Is that fair?

Rob: Yes. Because just through the way that the bi-directional links are structured, and if you understand how the block hierarchy works to relate blocks together, then you end up with this structure that works out really well.

One of the things that I’ve been working on and thinking about a lot lately, is that when people think about gamification, they tend to just think about points, badges, and leaderboards, that’s their mental model of it. And so, I’m just going to go ahead and say, what if we just call that the four squares genre of gamification. It’s a generally common set of design principles that work together on a generally common set of problems. Then what happens if we start coming up with more genres of gamification, you know? Like, what are these other repeatable structures that can be altered a little bit — within boundaries, of course — and what would those look like?

It’s like the four-square genre, the problem with it is, it’s a homogenous set of solutions that’s applied to a heterogeneous set of problems. And with this idea of genres of gamification, what I really want to do is over the course of all of the different projects that I work on, I want to be finding some of these common structures that I can use and that I can repeat and that I can start to understand what are the boundary conditions of when this is effective and when it isn’t effective, right? And then eventually I want to understand those boundary conditions so well and over the course of the career have established so many genres that I can start breaking genres and combining them together. Because those are really the best games. Like the first-person shooter RPG or the rogue-like Metroid Mania, you know? Like those are a lot of fun.

And what Roam lets me do is through all the ways that I just draw, these connections, I’m really able to start seeing these structures emerge, and I’m able to abstract out a little bit and say, okay, I’m working on this specific problem. Is there a little bit more general version of this that I can talk about that isn’t specific to my problem? And then, what other projects can I apply that to? This is this idea that I call like lenses. Like where I’ll just come up with a series of questions that I’ll ask myself. I’ll be like, ” what’s the lens of progress monitoring?” Okay. “What’s the person’s goal? How are they being given feedback about where they are in relation to their goal?” Stuff like that. Those are a common set of questions that I can apply multiple different projects. And Roam allows me to do that and see how all those connections work so I can refine those lenses better and refine these genres of gamification better. It’s a really powerful tool for connecting thought.

Growing usefulness over time

Jorge: I would expect that it’s a tool that only grows more powerful the more you use it over time, just because of what you’re describing. I’m going to call them serendipitous connections, although they might not be entirely serendipitous in that you have a framing question, for example. But the answers that those questions will yield will depend on the size of the corpus that you’re dealing with… the amount of stuff that you’ve put into the system, right? And with something like Roam, my impression is that that would take some time to develop a robust set of notes. What has your experience been in that regard?

Rob: My experience is that at first, I came into Roam with that idea of, okay, it’s only going to get more valuable, the more that I put into it. But now I don’t actually think that’s necessarily true. I think it just takes much longer on Roam than it takes on other products, like much, much longer. But once you start dealing with a really high amount of quantity, then you still need to have practices in place. This is a great example of what sorts of behaviors should a user be doing in order to gain the most value out of the product. I think that allowing structure to expand and branch out more and more and more is really powerful, but you also need to be doing compression behaviors.

You need to be consolidating your thoughts because, I mean, I just exported my backup of my private database the other day. It was like over 130,000 words over the course of my last, I guess, eight months now using Roam. But yeah, you really just need to compress that a bit. Like, if I have 20 different pages and a hundred different blocks that are all circling around the same idea, then I should just make a page for that idea. And know that anytime I’m referencing that, I’m referencing all of the ideas that support it as well. So I can go onto that page and I can put some block references and some page references to tell me like, okay, these are the sorts of things that lead into this idea, but that allows me to both attack problems from a high level, but also zoom in and get way more detail when I need more detail.

Jorge: Hearing you describe that, it makes me think that for the system to work as well as it is working for you, you have to put work into it, right? You have to take time out to reflect on what you’ve written, what you’ve put into it, spot patterns, consolidate, like you were saying, right? In the past couple of years, I’ve started moving more and more of my note taking into DEVONthink, which is another one of these tools. And as I’ve started consolidating my various information piles into this thing, I’ve discovered that I have a lot of work to do in consolidating things like tag lists, right? And doing this sort of work that you’re saying where it’s synthesizing certain things so that they’re not dispersed all over the thing, and it takes time. And it seems to me that the effectiveness of these systems depends not just on the tool, but also on the practices around the tool. I’m wondering if you could tell us about what other practices you have. You’ve already talked about this idea of synthesizing, are there other practices that you do that help you keep this on track and useful?

Rob: Yeah. I mean… One, I just want to say you hit the hammer on the nail in the sense that, like, what do you do with tools? It’s a tool for thought. What do you do with tools? You work with them, right? So, my job is that I think about things for a living. So, I need to track and develop my thoughts over time. I need systematic processes for myself to bring about creative insight and to consolidate all of the information I get from papers, from meetings, from lectures, all of that needs to be in one place. I will say that Roam makes it so you don’t need to do quite as much work as you would do on other apps. In fact, way, way, way less work, because the data architecture, as I mentioned before, with just knowing how blocks relate to each other, it makes writing in Roam into an extremely expressive thing. If you’re just operating intuitively under an understanding of how the data architecture works as you’re writing, then that means later you’ll be able to use queries and do a lot of this work in hindsight, pretty easily.

That being said, I do think certain practices like making index pages for yourself is useful. Just like, as you go, just add things to index pages, right? And then there’s also bringing things into outlines. Like, one thing that I do is, I actually have an article about this linked online. I have this like looking lens process where I’m like, okay, I’m working on this project, I need to come up with a set of answers on it. Like for example, I was working on the onboarding for GuidedTrack. How do we redesign the onboarding so that when people come into this app that is hard to use — or not actually hard to use, it’s actually really easy to use, it just seems hard to use because it’s based on a totally different UX than people are used to. Like people try to make surveys or experiments… they’re usually in a drag and drop thing. This is just an extremely simple coding language that people can use where they write “star question” and then the question, and a question appears for people. But it seems intimidating and people come in with a wide variety of use cases. How do we redesign the onboarding?

And so, I go into this page and I track open questions. Anytime I’m working on a project and I’m like, okay, here’s this thing that needs to be answered. I’ll tag that bullet with an open question and I’ll turn it into a to do. And that means I’m able to see what’s still unresolved and then from those open questions, I can turn that into a query. So, I might be like, “okay, how does search behavior relate to onboarding?” And I turned that into a course, and I pull in some past insight from my database where I’ve talked about those ideas in conjunction with each other. And then I’m bringing those together and just consolidating into new outlines and new understandings across a wide source of information. So yeah, I think just this idea of continuously consolidating… it doesn’t need to be that complicated. It can really just be, I have this page and I’m just going to drag in some block references for the most important things that I’ve already written about this so it’s just on the page. That can be a form of synthesis. It can just be, as you’re writing, if you think you’ve written something before, just write double parentheses and start typing in words and see if you can find the thing that you’ve already written about it, and then, bam! You have a connection right there to your past thoughts. So yeah, just tag everything, and consolidate every once in a while.

Closing

Jorge: Well, this is a tremendously exciting, and I wish that we had more time to get into more details. What I’m going to do is I’m going to include in the show notes — and I’m mentioning it here so that folks will check it out — but there’s a video on YouTube, I think it was an interview for the Keep Productive channel, where you actually show us your set up. And I’m just going to encourage folks who are intrigued by this to go check that out. Other than that, where can folks follow up with you?

Rob: People can follow up with me on Twitter, you can follow me @roberthaisfield. It’s just my name. You can also go to my consulting website, which is influenceinsights.io. You can go to my personal website, which is, RobHaisfield.com where it’s sort of like… It’s a digital garden, right? So, it’s a bunch of thoughts that are all very densely connected to each other, and you can just bounce around through there. It’s my ideas in motion. I’m not going to say everything you’re going to read in there is going to be complete, but next time you come back to it, you’re going to be taking an entirely different journey through it, and you’re going to be learning about different things, and maybe even some of the pages that you already read will look different. So, yeah, those are probably the best places to follow me. You can also add me on LinkedIn and email me at rob at influenceinsights dot io.

Jorge: Fantastic Rob! I’m going to include all those in the show notes. Thank you so much for being with us and sharing your knowledge of this amazing tool.

Rob: Thank you, Jorge for having me. I had a lot of fun. I love talking about Roam. Everything that I publish about Roam is 100% a hobby for me. I just think that it’s a cool tool, it’s made my work way more effective than it was before, it’s given me a knowledge base that I know I can trust 10 years later, it lets me build on ideas across projects and across time and I’d really just recommend anyone try it out. You know, like the way that I might be describing it, or the way that you might be seeing it online in YouTube videos or in articles, you might get intimidated, you might think, wow, this person’s doing something really advanced. We’ll just know, it’s like Excel. You don’t start out doing everything that you can do in Excel. The things you’re going to be doing six months later are going to be different than the things you’re doing one week in, and that’s okay. Your systems are going to evolve over time if you just play around with it. So, have fun!

Jorge: Thank you, Rob.