This interview is part of a series featuring the presenters participating in this year's Core77 Conference, "The Third Wave", a one-day event that will explore the future of the design industry and the role designers will play in it.
To a designer with a more conventional career path, Dean Malmgren's entry into design sounds a bit backwards. In 2009, Malmgren co-founded the company Datascope, a data science consulting firm. After an acquisition of their company by IDEO, he now works directly in their Chicago offices alongside his original Datascope team as an Executive Portfolio Director with a passion for human-centered design. You may be asking yourself, how does an initial curiosity around data collection and algorithms lead to a focus on human-centered design? The way Malmgren sees it, utilizing data alongside user research in considered ways can lead to even more human-centric product solutions than ever imagined. Malmgren's work hopes to push the idea that teams can invigorate their design practice by, as he frames it, "using data as a design medium".
We recently spoke with Malmgren, who will be taking part in a panel with IBM's Joe Meersman and Marijke Jorritsma of NASA Jet Propulsion Laboratory at the 2019 Core77 Conference "The Third Wave", about his background and how to reframe your relationship with data as a working designer.
Dean: We started Datascope as we were wrapping up our PhDs at Northwestern. Mike and I were both in the Amaral lab at Northwestern and we were studying the space of complex systems, which in a nutshell, basically involved taking big datasets and telling stories about what we were observing through the lens of predictive models and lots of other things.
It was through that experience that we gained a lot of expertise and developing models that made sense and that you could interpret. And it was really kind of the inspiration behind [us] designing algorithms that were useful for people, which I would say is sort of the founding principle behind Datascope. As we got started, as you might imagine, we learned pretty quickly that that involved a lot more than just data and code. It was a time of designing not just those algorithms but actually the surrounding products and experiences that accompany them.
So we very quickly started to learn from the agile development community, the lean startup community, but especially the design community. And along the way, we'd met IDEO who was at first an informal mentor to us. We would just grab coffee with them every little bit and talk to a few of their partners about what the future of design looked like and why data science is relevant. [We would also] ask for advice on growing a consulting practice that ultimately turned into an externship, basically where I spent a couple of weeks at IDEO doing a project and we had a senior designer come and work with us on some projects. That all went well. And led us to sort of question, you know, what was the next thing?
What was a step that made sense? And realized it was designing intelligent products and services in an authentic way. We do our best work [when] data scientists are working side by side with designers; it's not just the practice of human-centered data science, it's actually just another discipline of design. And so that was the impetus for actually being acquired by IDEO, just to continue pursuing what that looks like and bring that all to life.
By training I've never taken a design class, but to be fair, I didn't take any data science classes either.
Yeah. So I studied math and chemical engineering as an undergrad. My PhD was in the chemical and biological engineering department, although I never took a biology class, like since high school. And I've also never taken a statistics class, so I kind of learned statistics on the streets as they say. Design is kind of the same thing I suppose.
Yeah. At Datascope especially, we would take tools and approaches from the human-centered design toolkit and apply them to a data science context. So one thing that would happen all the time with our clients is that they would come to us and say things like, "we have this big data, what can we do [with it?] How is this valuable? Or we just read about deep learning and we need to use it in our business." And while those like statements came from a good place, the reality is that they're not grounded in a business problem or context.
And so it involves sort of sketching out what that could possibly look like. And we did that through the lens of design and learning about people's needs and what drove them. And then using that to sketch interfaces or services or whatever, you get a better sense for the data that you want to collect and how that might be valuable.
So it never felt like a forced thing. It felt really natural, just to start using those design toolkits for our purposes. And we've also bent it the other way to bring some data science-y tools to how we think about design.
Malmgren (right) with Datascope co-founder Mike Stringer (left)
Well, that's a mouthful and I'll get into that next. I can highlight a couple of things that are different about our work here versus at Datascope. One thing that I don't think would be a surprise to anybody is that IDEO has an incredible portfolio of clients and partners that we work with. So the scale of problems that we're working on is quite a bit more expansive than what we were doing at Datascope.
I would also say that the degree of collaboration that we're having across disciplines today versus then is obviously different. But that's intentional and it's been incredibly fruitful. I share this with a bit of trepidation, but it's a fact that's worth sharing, which is that everybody that came from Datascope remains a part of IDEO today, which in the grand scheme of acquisitions by bigger companies is pretty incredible, two years in. So I feel like we all are pretty passionate about continuing to push this edge of not just what doing human-centered data science looks like, but also more importantly, pushing the edge of what it means to design intelligent products and services.
I mean, I would feel a little uncomfortable speaking to this about the field at large because I've practiced it in a very specific way that's always been kind of intentional about how we're designing algorithms to be useful for people. Like when we were first getting Datascope started, as I reflect, I think it was largely led by this concept of big data, how data's the new oil and you better suck everything up now so that you have something to use later. And I think organizations are getting wise to the fact that that hasn't born as much fruit as you might think.
I think that's how the early era was. Today, I feel like in general, the field of data science remains pretty technology focused, but perhaps with a bit more open eyes. And there's quite a bit more conversation today about the ethical considerations that go into building models and thinking about the data that you're collecting in the first place. That's not only forced through regulations like GDPR or the equivalent law in California whose name I can never remember... The sunshine law, that's a good name. I should've remembered that! But at any rate, while those legislations are important, the fact of the matter is that these things have been top of mind in the data science community and have brought to light a lot of thought leaders in the space.
Design absolutely influences the data that you collect, which is actually directly related to how data science makes design better. I think in 2019 we live in an era where our watches are smart, our shoes are smart, we are using all these different tools and technologies that are connected. You have homes that are aware of when we're present and not. We have thermostats that are smart. You know, the list just goes on and on. And designing how those experiences unfold in an authentic way is, or prototyping what those experiences will look and feel like in an authentic way. It's not something you can just totally wizard of Oz.
Speaking of data science as a discipline in design impacts design in a really positive way by bringing these things to light and allowing us to experiment and learn from how these algorithms can be influenced by the experience and vice versa to make the holistic product or service a lot better. So we have some examples of that across our portfolio.
The economical project that comes to mind for me is this work that we were doing for a medical device company. And the way that this worked is that they had a new surgery, a new surgical tool and they wanted to help doctors facilitate conversations with patients to make them more aware of whether or not they were ready for surgery. And you know, some people when they go in for surgery, they're of the mindset that, oh, I'm just going to a body shop. You're going to swap out a knee and I'll be all set. And it's not that simple. Obviously, bodies react in weird ways to surgeries. And so there's a lot of preoperative and postoperative care that is super important. One of the big opportunity areas that came up was actually facilitating these conversations in a much more authentic way to allow patients to take ownership of their care and giving doctors the tools to sort of have that conversation.
And so we prototyped this interface that was driven on the back end by an algorithm that would basically connect different factors. I can't tell you the real [factors], but things like how frequently you exercise, your diet. These various factors that you can actually control as a patient and the outcome, the likely outcome, or how quickly you would get better after surgery from their data. And it was a huge hit that, frankly, I don't think would have been as impactful had the team not been working side by side along the way. And actually, the side story on all this, and the reason why I really like this story, is that the project team had come back from research and they had went down [for a visit with data scientists] for an afternoon. And the data scientists doing what data scientists do, threw together a halfway decent algorithm and a really bad interface to try to bring this concept to life.
Yeah. You see data impacting design in three different ways. I mean, data is often used to quantify design. You see that a lot in things like AB testing. We use data as a means to inspire design quite a bit. That can happen through exploratory data analysis to identify opportunities or more frequently, we find ourselves doing simulations of future states to illustrate how different designs could be experienced. In a way, that's sort of hard to do or hard to imagine as an interaction designer or an industrial designer. And then of course, data as a medium for design, where you actually are molding it or shaping it or deciding what data to collect in the first place, as you alluded to earlier. And that to me, that's what's most exciting honestly, is thinking about the data that you can and should collect and describing to users why it's relevant to them to sort of make that value exchange really authentic.
Well, that's how we like to think of it in the sense that at the end of the day, data algorithms serve people, not some robot overlord. And so it's really important to keep the people front and center when doing the work. And what that means in practice is that data scientists should, it turns out, talk to other people and learn about their needs firsthand. Data scientists don't talk to people in general, and that's something that I would very much like to change. I think it would generally improve the degree to which people adhere ethical standards and think about the impact that algorithms have on people in their lives.
Yeah. So we've actually done two of these exhibits. One was in Munich called hyper human and the other was called the discomfort zone, which was a exhibit that we did in Palo Alto.
So both of these we did as a means to push the edge on what we think that future could look like. So I mean, there's a lot of things that I liked about the tension that we tried to hold in those exhibits. It brings to light the tradeoffs between privacy and —well, frankly—convenience, which is often played out in a lot of these sagas in the news, but also showing how the future of work doesn't have to be scary actually. That there is lots of promise and reasons to be optimistic about having new skillsets and what that could mean for you. Also, giving people the sense of what it might feel like to be augmented by a machine rather than replaced by one. So there are a number of things that I thought were really cool. My favorite example from the Munich exhibit, I think it was called the "belief checkout". And so with belief checkout, the idea is that when you go to the grocery store and you purchase some products, you have to do a lot of research if you want to shop according to your values. And so the thought that you could bring all that to life in a really compelling and easy to access way that plays off the real things that people are worried about, like cost, convenience, materials it's made of, et cetera, et cetera. That's all really important. And it was fun to play around with that and see that come to life.
Well, that's a great question. I like to think about that a lot. The short answer is that, you know, these systemic challenges, whether it's education, poverty, climate change, equality, I mean, that list goes on and on. I think the challenge, in any sense, is a cultural one about agreeing on principles that we can all sort of align to. And from a data point of view, I would hope that means thinking of ways to make data accessible and transparent, but also respecting the value that it does bring. And so what that might mean is, you know, thinking about how we share wastewater data for example, or pollution data in a way that benefits everyone.
Speaking directly to the Industrial Design audience of Core77, one of the things that I've found really exciting is, you know, as we design artifacts and objects that are in fact intelligence, that's a two-way street. You know, designing the next thermostat, as an example, it brings together a lot of these different skill sets. And I think there's ample room for collaboration, particularly in an era of IOT and other things that we've really only started to scratch the surface on. And I'm pretty convinced by that, you know, designing objects that are aware of their surroundings and help us take meaningful actions and adapt to our ever-changing context. It's an area that's ripe for continued innovation for the years to come. So I'm pretty excited to be speaking with that group.
Hear Dean Malmgren and other design industry leaders speak at this years Core77 Conference, "The Third Wave"! Tickets are available now.
Don't have an account? Join Now
Create a Core77 Account
Already have an account? Sign In
Please enter your email and we will send an email to reset your password.