Design for Action

Juhan Sonin at UXDX Community: Actionable Design, AI-Native Products. Video: https://youtu.be/3_MKvi3_1uY

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

Getting in the game

[00:00:08] Hello, my name is Juhan. I pretend to run a design business in the Republic of Boston called GoInvo, and I teach at a private equity firm in Cambridge. Because I'm talking about health care and software, which are all political, it's nice to know what my motivations are, or what they're not. So yes, all my health data, including my entire genome, has been open source. So feel free to plant it in a crime scene somewhere. If you want to get a hold of me, I will try to get back to you within a few business hours. Just email me, yell at me here or in public. I'm easy.

[00:00:47] Look, a lot of engineers, of designers, of clinicians are struggling here in the dis-United States. We've had this weird funding spigot with the Fed. It's been really strange. But I get this question a lot: I want to have an impact on healthcare. Okay, so how do we go about doing it? Where do I go? Now, there are a metric ass ton of options out there. These are just a handful of them. So that's also daunting.

[00:01:18] And at the same time, just on Sunday, I happened to be at the Camden Conference, and I heard this from a really good reporter who's covering the Middle East: "Hey, I'm a journalist. I don't have solutions." I've heard this too from the designer-engineer world, that says, "Hey, I'm a designer. I don't have a point of view." Well, okay. But part of this is, haven't you seen potential solutions, or have you reported on them? What does your research suggest that we do?

[00:01:52] But the idea is, okay, yes, we're probably overwhelmed a lot. But the gist has to be: let's get in the game. And by doing that, you can just start making stuff. That's what I tried, at least, and maybe screwed up on at the beginning. I'm like, "Okay, how do I see my own health?"

The health card and the projects it begat

[00:02:11] This is some time ago. This is how I got into the field, because I went to one of these little nursing stations at a company I was working for at the time and had my cholesterol checked, and the nurse looked at me bug-eyed after she saw my results, going, "How the hell are you a functioning human being?" Well, I'm like, "Well, maybe I'm not." But this got me started on what the hell is happening inside of me? Where's the dipstick? What's it saying? And this started the health card.

[00:02:40] This isn't anything magical or anything like this, but this is a few decades ago. And this is what I popped out after some trial and error. And it's funny: nobody asked for this thing. Nobody. Just me, my brain. Nobody in the department funded this. I was at a bigger company at the time. We just started to make it.

[00:03:08] Now, I started bringing it to all my clinical visits. I had it in my pocket all the time for years, actually. And what was interesting, part of this maybe for others, is it was open source. It drove inputs to some of the software we were about to make at the time, and we used it as a validation service for the health data that we were capturing, because we could render it. And ultimately it started other projects.

[00:03:42] Now, it's funny: I was walking around Quantified Self conference, the number one, number two, and someone recognized my mug from that, and they actually used it at WebMD to validate their patient views. But then we said, okay, well, how do we operationalize this? What kind of data inputs do we need? And so we actually made a prototype. This isn't glorious. This isn't remarkable. But we started: how do we make this thing real?

[00:04:07] And look, there were three projects that came around during that time. One was a US testing service, an entire nation testing service for electronic health records, called Laika. Another was hData, which eventually was folded into the small and inconsequential thing called FHIR, for the planet: a data API, or how to make data more accessible to everybody and systems. So these three little projects nudged a whole bunch of others. Not at a huge rate, but they had some tangible nudging for bigger national plays. So a lot of other projects also germinated from these three.

[00:04:51] Apparently I'm good at designing what goes with circles. But that's what happens. Projects can beget other projects, and even if the first one's not great, like my health card, it drives other potentials. So my hope to you, to all of us, is: hey, we make stuff, we ship stuff. We hopefully publicize or open source it, depending on what kind of license you want to use, and you're planting the seed for others. And ultimately this starts to drive your own agenda.

Stop seeking approval

[00:05:28] Health care is really risk averse. Hell, the city of Boston has become so tight around the health care space, because of all the money involved, it's gotten even more risk averse. The lawyers, their job is to say no. And look, healthcare is lives on the line. So it begins to say: this is hard by design.

[00:05:53] So the first thing, to break through some of this, is stop always seeking approval, because the first thing your colleague, your manager, the department head, the president's gonna say is no. But this somehow, for the species, occasionally evokes paralysis: "Oh, we can't do this." So for great ideas, for your interesting stuff, you have to see no as yes, because they're going to keep saying hell no. So you just got to start. You got to make stuff. And don't wait for consensus. It doesn't work. Design requires judgment and direction. You just got to go.

[00:06:33] And by the way, this is a little side note: the more ideas you have, even a ton of crappy ones, the potential for having better quality ideas is validated to be true. You can look at an MIT study here. I'll forward my slides at the end so you can see it. But the more ideas you have, the bigger the count of good ones in there.

[00:06:58] One dumb example of that is when we were doing hData, this data standard: how do you put your information in code? That led to the standard health record. Well, we were thinking about cancer, and like, okay, this is the second biggest killer in the United States, on the planet, along with heart disease. And we're like, what can we do in a week? What concept can we start to generate in a week to say, "Okay, this is something worthwhile for our R&D tax dollars"? So we just put this together. It's not great. In fact, the graphs drive me bananas because I know they're fake. But we put a couple of ideas together of what we could do to prove that having structured data for cancer, and for patients and clinicians, could actually do good.

[00:07:49] And so what happened is we got this together, and all of a sudden it begat a whole bunch of small projects that turned into big multi-million dollar projects over the course of the next year. So a week or two seeded much bigger projects. I can go into these at some other point. If we have questions, just let me know. But the idea here is you have to fight for good ideas. You have to start fast. Make something that could be usable. Be willing to change your tactics, because these things don't stay the same when you're making them. You have to get feedback and move them. You know this too as a designer or engineer.

[00:08:26] Ultimately, those little screenshots I showed were completely different than the thing we built a couple of times over for several cancer institutes here in the US. And then you have to never give up, really, hopefully, and keep pushing through, and make sure your ethic is there.

Caring enough to fight: food stamps and eligibility

[00:08:43] Now, many of you probably have heard of Gropius, the architect. After he moved from Germany to here in Massachusetts, he taught at Harvard, and this was one of his graduate student buildings that he made. And there was this classic line when he was interviewed about it in the Harvard Gazette by one of the reporters. They asked him, "So why did you bolt down the furniture in all the dorms?" And he looked quizzically at the reporter and says, "Well, why on earth would you want to change it?"

[00:09:18] That's insane. But also there's some greatness there. But totally insane. But the idea that, okay, sometimes his ego got a hold of him. Sometimes it happens to me. But if you don't give a crap, that's what really kills projects. So okay, ego sometimes gets in the way, but if you don't care, what the hell? So why am I mentioning this? You have to start caring and fighting through the BS, especially in industries that are regulated, like healthcare.

[00:09:49] So in this slog to create beautiful things here, you also need to think about your products, who's going to use them, your own team, and your city, your state, your nation. And one project that we happened to work on that involved all three slices of that, from local towns to the state to the nation, is we worked on food stamps. This idea where if you earn a certain level, or at a poverty level, you actually get some reimbursement for your food. Sounds pretty basic. Most civilized countries on this planet have something like this. But we made a service and shipped it for Massachusetts.

[00:10:34] And we had to fight with the lawyers, fight with the policy wonks, to drop how many questions you asked. It was a long, arduous task. But after we shipped it, at the beginning only 7% of any Massachusettsian, someone who lives in Massachusetts who qualifies, did it online. But after only a few years, now 70% do it online, because they can do it from basically a dumb phone and get it shipped to their house, or walk in and get their SNAP card. That's a pretty remarkable service. That is an example of how we work with our towns and cities to make something a little more elegant. I'm not going to show the design, because the design is pretty straightforward.

[00:11:24] But what happens too is you can think about more than just your initial problem set, like food stamps, and think about: well, how do I then think about applying for all the services that I qualify for? Why don't they just come to me? If we're in the country of Estonia, this already happens, where if you sign up for being a resident there, all of a sudden, automatically, it shows up on your phone what you qualify for, what you can get. But here in Massachusetts, that's not how it works. We had to fight for this over time. We had to say, well, look, all these services should be connected, and you should automatically get all these things coming to your phone. It took years for us to fight for this. And it's not just us fighting for this. It's other organizations who are doing the bulk of this, advocacy groups. But we had to start somewhere. So we started making it.

Having a policy agenda

[00:12:17] Now, when you're dealing with national or state services, you also have to think about: well, how do we change the law? What laws are there? What should we be doing? What's your point of view on the law? And so here are mine. You're not going to agree with all these, by the way, but it allows me and the design studio that I work at to say, okay, what do we pick and choose to work on? We all have our points of view. Not all of us agree, but we can agree on the things that we want to jam on in an open source fashion. So these are a few of them. But I think this is a really good exercise for you and for all of us, to say: what's our agenda? What's the law we want passed? Easier said than done. But you have to show up. You have to have this kind of agenda.

[00:13:14] And we publish all this too as a studio. We say, "Okay, we need to keep ourselves on the line." So we publish it online. We say, "This is what we're going to do." Sometimes we break it and we have to change plans. Okay, it happens. Having a national federal health agency about literacy: maybe it's fantasy at the moment, but we'll still fight a little bit for it.

[00:13:40] So we publish and we work on this. We say, "Hey, look, if we're really interested in owning your health data," and I'm wearing one of my health data shirts, "Own it," what do you start with? Well, do a little research. So we did. We wrote a position paper, actually two of them. We got it published. We wrote an open source graphic novella. You can download it at ownyouhealthdata.org [?]. And we published it, and we seeded a small piece of legislation for Massachusetts.

[00:14:08] And what do you do after you seed it? Well, you send it to some representatives, which I have done. I've sat down with my local representatives: how do we get this on the docket for Massachusetts? That's a longer play. But it also doesn't just start there. Usually you want to team up and figure out what other design or engineering or policy firms or healthcare firms are doing this. And so we worked with several organizations to come up with, hey, what's a draft patient data use agreement? It's like a contract between you and the clinician or the clinical teams that identifies who owns the data. We came up with some prototypes, and in parallel did our position papers and graphic novel.

[00:14:54] So this is part of how you think about it, or part of what we've done in this particular case on policy. And then you always want to start with the end in mind. So if you want this idea of, hey, you step on US soil, you step on X country's soil, you get healthcare. Okay, so how do you get there? What are the steps to get there? Because here in the United States it doesn't just happen. There are these incremental components. So maybe you think about what are these ways to get there, so that you have a path to predition [?].

[00:15:31] This happens all the time: "Well, I don't want to get into politics and all this kind of crap." Yeah, but the problem is, what happens if you don't? Because the CEOs of these companies are going to take advantage of you, basically, in egregious ways. We have seen it over time in memorial [?]. It's become more hyperfocused now, because everyone's like, "Ah, they won't even bother fighting us." "Oh, we'll tweet something," or Twitter's dying or whatever it is, "but we'll put up some banner somewhere. Oh, that's our way to fight." Okay, it's a start. But big companies don't see it that way. So for me, it's like, hey, you have to have an agenda and do it multiple different ways.

Pace layers and your own policy list

[00:16:20] And one way is to think: let's pretend for a second we're not at the Harvard Business School, where finance and money is the center of the earth, but actually more like the pace layering that Stewart Brand has. What I'm trying to show here is that things at the very periphery of this move very quickly, and as you dive deeper and deeper and deeper, it gets harder and harder to have influence on it. So in the design world, in fashion, some of our work moves fast and changes a lot. But as you go to commerce, as you go to infrastructure or culture, these things take a lot longer to move the general public on.

[00:17:04] And so this is the way I like to think about why you should get into the policy game, or at least think about it: because it's really hard, but it has enormous impact once you pass something, hopefully for good. And that's one additional way, other than just designing and shipping things, that you can expand your impact.

[00:17:27] Now, you also have to think about what's the system of this. And so yes, I put together some of my position pieces for the US. Again, you're not going to agree with these. In fact, I have a whole bunch of new ones, especially over the last six months, with things going down strangely here. But I published these as well, so there's nothing I'm hiding here. And not all of these are right, but the idea here is: what's your agenda? You need your own policy list. So, hey, this is what we're fighting for. This is what we need as a city or state.

[00:18:02] Ultimately, actually, we need a policy Mad Lib service, but that's a separate little problem. And I talked briefly about an eligibility engine: how does a state or country automatically take and give you benefits that you qualify for? Well, we've worked quite a bit on this. And even though the state at the time, Massachusetts, wasn't into this as much, we came up with a position piece. What needs to happen? Why does it need to happen? What would it act like? What would the policies need to be? We even came up with what are the common data elements for being a resident of your country or your state. This is not an easy problem. And so we came up with a list.

[00:18:45] And finally, I think Massachusetts now is prototyping and has a draft of this, working with Medicaid, MassHealth, and food stamps, which is fantastic. Not that we forced them into this, but at least we had a position, and we published it and open sourced it. My hope is that other states follow.

Signals, top problems and wrapping up

[00:19:07] Now, again, because I live in the US, this is actually pretty great: there was a non-binding vote, you have to start somewhere, on whether single-payer healthcare, universal health care, should exist here in Massachusetts and in the country. And Massachusetts had this on several ballots across the state a year ago, and it was unequivocal that people wanted this. Republicans, Democrats, independents: 65% said yes. And usually it's 51, 52, 53%, but you have 10 points, 12 points higher than that. Clear signal.

[00:19:53] And when I talk about owning your health data, it's nice to see other people getting involved. MITRE is a nonprofit company here in the US that actually put this in their strategy for the company. That's pretty rad. It's not exactly a right or left-leaning organization that's particularly aggressive. So they put it in there. That's pretty great.

[00:20:21] But now you think about, well, is owning your health data a top 10 healthcare problem? Well, probably not. Maybe it's in the top 20. Yes, we've made a list of what we think the top 20 or 25 problems are here in the US, based on what the impact is to patients and humans and how much money is at play. You can see that here at US healthcare problems. But do you have your own? What's your top 20 list for your town, for your state, for your nation? I don't know. We probably need that. And guess what? We're working on a prototype, another open source project, that begins to look at: well, what are these things? How do you know people agree with you? And then how do you begin to synthesize, or know, that people are going to agree with this over time? And so we have a prototype here. It's on GitHub, by the way. It's just evolving. If you're interested, you can give me an email and see the early concept here that we're just brain farting at the moment.

[00:21:26] So to wrap up here: when you're designing for action, you need to, one, start something, and make it, and ship something small, something that says here's proof of life. You have to live the problem. With my health card, I lived with it for a couple of years. It became something that many of us in the department started to use. And that means bringing it to your clinician, bringing it to the environment, the scenario that you're planning to design for, that you're engineering for.

[00:22:02] And then ultimately you need some stamina, because working in the civic space, working in the healthcare space, is not something where, just because the machine can poop out an app in an afternoon with you, that uptick in workflow is going to work for everybody. Sometimes that can take weeks, sometimes it's going to take months, sometimes years. So this requires a kind of "I'm going to fight for this over time." That's not sexy. It's not always fun. In fact, there's very little fun. It might be 3.1415% fun, but you got to fight. And this is over the long now.

[00:22:41] Look, it's complicated. Healthcare is messy. It's hard to make beautiful. But my hope is, like American democracy, healthcare is resilient and will fight back, bounce back. And what you make today is obviously, or maybe not obviously, going to look a lot different than what it manifests in a year from now, in five years from now. But you have to start. You just have to start.

[00:23:10] I've stolen from all of these people here, and they have given me critique over the years, and this is my email address and phone number. Feel free to publicly shame me at will, or just give me a call. You can download this talk too at designagenda.org. Thanks very much.

Q&A

[00:23:29] Rory: Excellent. Thank you very much, Juhan. I think it's the first time, on the closing of a talk, I've seen somebody put up their phone number. So we'll see whether you get any phone calls as a result. But yeah, thanks very much. I really enjoyed the framing, because I've worked a little bit in public sector, but not as much as you have. I've worked in some very slow corporates, and I know the same of fighting for what you believe in. My first question is really around this: I think it looked really impressive how you have all of your policy decisions, but where I was thinking when I was looking at that is, I don't even know where I would start on that. I think that it's an incredibly valuable thing to have, but there are so many unknown unknowns for me around the law, around the existing policies. How did you go about setting that up in the first place? Because I believe, as you were explaining it, that's your driver for everything from there on.

[00:24:39] Juhan: Well, it's one driver. And it wasn't a panacea. It didn't come to me in a fabulous dream after having a lot of mushrooms and OD-ing on cocaine, although it may feel like it. I think you have to grow into it. And so it started with me, when I started with the health data side: there is no health data standard for the United States, or country of choice. And I started with a couple, and that got me into it, and I go, "My God, why don't we have this? Why don't we have this? Or what's happening with this? Or this is totally fubar." God. And then there's another. Right after I had the first three little policies, that helped our work too, because it was the culmination of those early prototypes in healthcare: my God, there's no health data standard. Okay, how do we get our health... there are some rules, like HIPAA here in the US.

[00:25:41] It started small. So that's my answer, Rory: it started small. It wasn't this huge, massive policy list at the beginning, but then it has evolved over years to say, okay, this, this, this. Okay, now I'm more interested in that. And now it's beginning to say, well, how do we do this for the whole US? And there are other services out there where you can say, oh, this can help me figure this out. But sometimes it's just things that affect you too. And you start to think about it. You have to start somewhere, and so start small. Start with things that impact you.

[00:26:22] Rory: Excellent. And I guess the other thing that people might be thinking of, because yes, I know bureaucracy, I know all of these problems that I'd love to solve: where do you carve out the time, and how do you make this a priority when there's a million other things going on with work, with life, etc.?

[00:26:44] Juhan: I think you just have to make time. How do you do it? We live in a fantastic time where time is becoming more of a good thing for people. We're not toiling in the fields. Most of us here in the digital space are not wiping ourselves out completely. And how much time do you spend doom scrolling? Not you, Rory. I'm guilty. Carving out 20 minutes of a day to say, I'm gonna focus on, hey, I'm gonna write these two paragraphs, I'm gonna draw this diagram, whatever. You can do it. It just requires a little stamina. Even 15 minutes a day. It's like if you're learning to play an instrument: 20 good minutes a day is way better than three hours once a week. So that's my answer, but it's up to you. How do I answer that? It's really how you adjudicate your own behavior.

[00:27:50] Rory: Brilliant. Well, that actually brings us to time. So I just want to thank you for sharing. It's really inspiring. It's great to see that there is progress happening, even if it feels like progress isn't happening in some of these spaces.

[00:28:02] Juhan: Oh, we're going backwards.

[00:28:04] Rory: We're going backwards. So it is great, though, to see that there are people out there pushing it, and thanks for sharing your advice with everybody.

[00:28:11] Juhan: Thank you.