Today I am bringing you a recording of a live interview I did at the TDWI Munich conference for data leaders, and this episode is a bit unique as I’m in the “guest” seat being interviewed by the VP of TDWI Europe, Christoph Kreutz.

Christoph wanted me to explain the new workshop I was giving later that day, which focuses on helping leaders increase user adoption of data products through design. In our chat, I explained the three main areas I pulled out of my full 4-week seminar to create this new ½-day workshop as well as the hands-on practice that participants would be engaging in. The three focal points for the workshop were: measuring usability via usability studies, identifying the unarticulated needs of stakeholders and users, and sketching in low fidelity to avoid over committing to solutions that users won’t value. 

Christoph also asks about the format of the workshop, and I explain how I believe data leaders will best learn design by doing it. As such, the new workshop was designed to use small group activities, role-playing scenarios, peer review…and minimal lecture! After discussing the differences between the abbreviated workshop and my full 4-week seminar, we talk about my consulting and training business “Designing for Analytics,” and conclude with a fun conversation about music and my other career as a professional musician. 

In a hurry? Skip to: 

• I summarize the new workshop version of “Designing Human-Centered Data Products” I was premiering at TDWI (4:18)
• We talk about the format of my workshop (7:32)
• Christoph and I discuss future opportunities for people to participate in this workshop (9:37)
• I explain the format of the main 8-week seminar versus the new half-day workshop  (10:14)
• We talk about one on one coaching (12:22)
• I discuss my background, including my formal music training and my other career as a professional musician (14:03)

Quotes from Today’s Episode

• “We spend a lot of time building outputs and infrastructure and pipelines and data engineering and generating stuff, but not always generating outcomes. Users only care about how does this make my life better, my job better, my job easier? How do I look better? How do I get a promotion? How do I make the company more money? Whatever those goals are. And there’s a gap there sometimes, between the things that we ship and delivering these outcomes.” (4:36)
• “In order to run a usability study on a data product, you have to come up with some type of learning goals and some kind of scenarios that you’re going to give to a user and ask them to go show me how you would do x using the data thing that we built for you.” (5:54)
• “The reality is most data users and stakeholders aren’t designers and they’re not thinking about the user’s workflow and how a solution fits into their job. They don’t have that context. So, how do we get the really important requirements out of a user or stakeholder’s head? I teach techniques from qualitative UX interviewing, sales, and even hostage negotiation to get unarticulated needs out of people’s head.” (6:41)
• “How do we work in low fidelity to get data leaders on the same page with a stakeholder or a user? How do we design with users instead of for them? Because most of the time, when we communicate visually, it starts to click (or you’ll know it’s not clicking!)” (7:05)
• “There’s no right or wrong [in the workshop]. [The workshop] is really about the practice of using these design methods and not the final output that comes out of the end of it.” (8:14)
• “You learn design by doing design so I really like to get data people going by trying it instead of talking about trying it. More design doing and less design thinking!” (8:40)
• “The tricky thing [for most of my training clients], [and perhaps this is true with any type of adult education] is, ‘Yeah, I get the concept of what Brian’s talking about, but, how do I apply these design techniques to my situation? I work in this really weird domain, or on this particularly hard data space.’ Working on an exercise or real project, together, in small groups, is how I like start to make the conceptual idea of design into a tangible tool for data leaders..” (12:26)  

Resources and Links:

• Brian’s training seminar

Transcript

Brian: Welcome back to Experiencing Data. This is Brian T. O’Neill, and currently, I’m in Europe and not in the studio as you can probably tell from the background noise. Last week, I was invited to the TDWI Munich Conference for data leaders to give a new workshop on using human-centered design to increase adoption of data products. And as part of that, the director of TDWI, his name’s Christoph Kreutz, asked me to do a brief interview about this training that I was going to be premiering later that afternoon.

We recorded this so that their at-home remote audience could get an idea of what some of the in-person events were like. So, a lot of the talks and stuff were also streamed for people who weren’t ready to attend in person. So, our chat here was taped in the TDWI booth, which was in a large conference exhibit hall with all the different vendor booths, and people mingling, so it’s definitely going to sound a little bit noisy. But hopefully, my Zoom recording here sounds good. You’ll hear Christoph initially addressing the at-home audience in German, and then he’ll be switching to English for our short interview segment.

So, in the end, I had a blast giving this new workshop format training in person for the first time, and I’ll probably be offering this as a new service for data teams who are looking for a shorter and less intense version of my main training seminar. So, if you want to get on my mailing list, get all my writings and insights and hear about that, just head over to designingforanalytics.com/podcast. And without further ado, here is my brief interview with the VP of TDWI Munich Christoph Kreutz.

Christoph: [foreign 00:02:07] yeah, now I’m already starting switching to English because I have a very special guest with me. It’s Brian O’Neill. And welcome Brian.

Brian: Yeah, it’s great to be here, thank you for having me.

Christoph: How are you?

Brian: I’m doing great. Yeah. It’s nice to be back in person at the conference. It’s been, I don’t know, three years for me since I’ve seen anyone in the flesh. [laugh]. So, it’s really nice to be here.

Christoph: And that’s so special for us. So, we’re happy to be back again. And yeah, you took all the way from the US?

Brian: All the way from Boston.

Christoph: From Boston? All right?

Brian: Technically Cambridge, Massachusetts. The People’s Republic of Cambridge.

Christoph: [laugh].

Brian: [crosstalk 00:02:56] I guess.

Christoph: Okay, good. And how did you get to TDWI Munich?

Brian: Yeah, so I think our mutual friend Mark Madsen, who’s a fellow at Teradata and he was a speaker here—

Christoph: Yeah, another time he [crosstalk 00:03:08].

Brian: Yeah, he told me about this. And we cross some of our work processes over. He’s very interested in design and product and how important that is to make sure that the solutions we make actually get used in the last mile. So, I’ll never forget Mark’s presentation when I—I think it was like [Strata 00:03:23] in London or something, any—he had a dashboard slide about how users don’t like dashboards. So, it was a dashboard about dashboards. Like, “I got to know that guy,” like.

Christoph: [laugh].

Brian: So, now we, like, always go hang out at these conferences and stuff. So, Mark’s a great guy, and unfortunately, I was on a plane when he did his talk. But.

Christoph: So, he recommended to submit [crosstalk 00:03:43].

Brian: Yeah. So, he was like, “Oh yeah, contact to TDWI, and, like, maybe they’re looking for, you know, speakers or training or a workshop or something.” So yeah, so I suggested that you guys, take a look at some of my work and see if it would be a good fit for your audience. And here I am.

Christoph: Yeah. So, Brian submitted at our call for contributions, and we selected, or the technical chairs selected the talk. And yeah, we will have it today—

Brian: Today—

Christoph: —in the afternoon.

Brian: —yeah.

Christoph: So, maybe just for the viewers—

Brian: Sure.

Christoph: Unfortunately, it will not be in the stream, but what do they miss?

Brian: Sure. Yeah, so the workshop is called “Designing Human-Centered Data Products.” And so, the main thing that I’m trying to do with the work that I do in my podcasts, and my writing, and speaking, and all this stuff is solving this problem of low adoption, or what I call technically right, effectively wrong solutions, of which there are many of these, particularly in the enterprise. So, we spend a lot of time building outputs and infrastructure and pipelines and data engineering and generating stuff, but not always generating outcomes. And so, users only care about how does this make my life better, my job better, my job easier? How do I look better? How do I get a promotion? How do I make the company more money? Whatever those goals are.

And there’s a gap there sometimes, between the things that we ship and delivering these outcomes. So, why is that and how do we deal with that? And the world that I come from, which is more on the vendor or software product design side, there’s a lot of techniques for this that are quite mature that can be applied to analytics and data science. So really, I’m trying to give non-designers some of these design tools that they can use. And so, the workshop is really a very condensed version of this eight-week seminar that I typically give online for teams and individuals.

And so, we’re just going to focus on three specific things in that. The first one is if you’re going to design something that is supposed to be useful and usable and deliver value, how would you test or know—before you ship that [and 00:05:40] is too late, whether it’s good? Like someone will actually use it and it’s useful? How do we get everyone in the team to agree on if we were to test that, what that will look like?

So, we use a method called usability studies to do this. But   And then we observe them doing that. And then we decide, like, did they pass or fail? And if they fail, we can take some learnings from the failure, fix it before it’s too late, and then that de-risks the whole project, right?

So, we’re going to learn how to do that. The second thing I learned how to do is how do I get unarticulated needs out of the heads of stakeholders and users? So, there tends to be a lot of discussion is still about requirements documents, and you know, we get this long list of features, and just give me this data or just give me this, you know, CSV file, and that’s all I need; don’t ask any questions. Well, the reality is most users and stakeholders aren’t designers and they’re not thinking about someone’s workflow and how this fits into the job of say, the CFO, or some salesperson. They don’t have that context. So, how do we get the really important stuff out of their head? So, we’re going to talk about using qualitative interviews as a method to get these unarticulated needs out of people’s head.

And then the third thing is sketching, how do we work in low fidelity to actually get people to get on the same page with a stakeholder or a user and design with them instead of for them and get visual early? Because most of the time, when things are visual people start—it starts to click, like, “No, no, no, no, no, that’s not what I’m talking about.” Or, “Oh, I see where this is going. And what if we did this?” And it’s a very different way of working. So, those are kind of the three learning outcomes we’re going to focus on today.

Christoph: Okay, that sounds totally interesting. And the thing is, the format is not a three-hour talk—

Brian: Correct. No, no, no.

Christoph: —but it’s a workshop. So, what is your approach? How—

Brian: Yeah.

Christoph: Are you going to do it? Is it some role modeling, a roleplay, or—

Brian: It’s funny you mention that. One of the things I had to try, that’s difficult about teaching this, you know, in the seminar format, I really tried to have—when I have eight weeks, I want people applying this to real work because that removes the need for people to roleplay, like, “Pretend you’re a CFO in a company,” you know, it’s very hard. And then, “Interview the CFO,” like, Christoph the CFO, well that’s not your job, so it’s really hard for you to know what it’s like. So, I actually came up with some scenarios that everybody can relate to. So, we’re going to actually use some realistic scenarios.

And there’s no right or wrong here. It’s really about the practice of using the methods and not the final output that comes out of the end of it. But we will primarily be working in small group format. So, it’s very light on lecture, no one wants to listen to me talk any more than this interview. So, short interview—I’ll just do a short talk about kind of the concepts about what we’re going to do, and then we’re really going to jump into, like, hands-on format and working together with a little bit of lecture, just 5, 10 minutes here and there to kind of get the concepts across.

But you learn design by doing design so I really like to get people going, trying it, you know, not talking about it. This is why design thinking if you ever hear about this, I like design doing and not design thinking because it’s really easy to get really conceptual about it. And let’s actually do it instead of thinking about it. [laugh].

Christoph: I like this approach because, of course, if you start doing, that’s actually with our TDWI [young guns 00:09:03], the young people in the TDWI Association or in the TDWI community in Germany, Austria, and Switzerland, we had one plan, so, start doing, and so then you get your experiences. And so, that design, I really love ‘design doing.’ So, as we heard, this is the session, the workshop today, and for example, the viewers at home, not right here in Munich, do they have the chance to get this experience? I’ve heard about that you’re doing remote seminars and stuff.

Brian: Correct, yes. So, this Designing Human-Centered Data Products, this workshop we’re doing this is again, a condensed half-day version of what I typically do in an eight-week format. So, I teach that publicly twice a year. The next one is, I think, September 26 is the first date. So, I teach that twice a year that is open to the public. Anyone come to that. And then I teach it privately all year round with teams. Typically those are enterprise data science, AI, analytics teams will come in and we’ll just arrange an eight-week period that works for them. So yeah, there’s an opportunity in the fall within the next time to take it.

Christoph: Okay.

Brian: Yeah.

Christoph: Eight weeks is quite a while. So, what is the format there, and what’s the difference?

Brian: Yeah, the reason it’s eight weeks is because I don’t believe, like, you take this one workshop on it and then go away and do it and that’s the end of it. There’s no accountability there. There’s no feeling about, “I’m going through this because Christoph, who’s also going—you know, he works at a consulting firm, and I work at a bank and someone else works at a pharma company, but we’re all trying to build decision support applications with different contexts.” And sometimes it’s knowing what it’s like in a different place is where you get ideas for things. And this is actually a very well-known concept in design where we want wide collaboration, especially at the beginning of projects during ideation phase because sometimes it’s not so much that the banking way was the right solution for pharma, but listening to someone talk about banking and some analogy got me thinking about a third thing, which I never would have thought of have I not heard this person in banking talking about whatever.

So, that’s the benefit of the seminar is to have that experience through our office hours and review time and all of that. So, the exercises themselves are, you watch the videos offline; there’s eight modules, each module is, like, 15 minutes of that, and then there’s, like, an execution plan. Like, “How do I go and take what I saw in the video and apply it?” Then really what we’re doing on the live calls each week for an hour or two is talking about what was it like when you went out and try this? Or you didn’t try it? Why didn’t you and what’s the internal resistance that you’re having, or the organizational resistance? And we try to work through some of that.

And sometimes we have to talk about politics or, like, the internal office situation, or dealing with a stakeholder who doesn’t—“I don’t have time for that. Just give me what I asked for.” How do I negotiate those situations? That’s really what we’re doing. And so, the eight-week time period is really it’s a time to practice this and have support from me and the rest of the cohort so it sticks and it doesn’t just get flushed down the toilet when you leave and then you don’t ever use it again. Like that doesn’t help anybody. So.

Christoph: And that’s where we again, start doing.

Brian: Yeah, [laugh].

Christoph: Like, you’re—

Brian: Start doing.

Christoph: Yeah.

Brian: Yeah.

Christoph: Design doing and then start doing, and then you have kind of a coaching?

Brian: Yeah. Yes. Yes. Very much so.

Christoph: So, through the different steps of the project [crosstalk 00:12:22]. Yeah, that sounds really fun.

Brian: Because everyone needs help with, you know, the tricky thing, especially with any type of adult education is like, “Yeah, I get the concept of what Brian’s talking about, but, like, how do I apply that to my situation?” You know? And that’s usually where the trick is. It’s like, conceptually, I understand these things, but I work in this really weird domain, or this particularly hard thing.

And it’s that crossing that bridge between some conceptual idea and really making it hands-on, that’s where people need help. And some people don’t, so if they want to take the self-guided [cough]—I offer this as a self-guided version, as well, which is just the curriculum, but no Brian, no cohort, no support. And if you just want to do it on your own, you can do it that way, too. But most people sign up because they need the accountability and making time for it. Because we all say, you know, everybody’s busy. And so, this way, there’s a one or two hours every week dedicated to getting together and try to make that habit stick, you know?

Christoph: Yeah, very interesting. Because yeah, I think that’s a crucial part. You go fully motivated out of a workshop like that, and then you’re back in the reality.

Brian: Yeah. Exactly.

Christoph: Yeah, great. And if I’m interested, how can I apply?

Brian: Oh, yeah. So, designingforanalytics.com. That’s my website. If you go to slash seminar, that has information about the upcoming seminars, and there’s a, like, a get notified list and an early registration list, they can hop on there. So.

Christoph: Okay, that sounds great. And then yeah, it was a long ride. And because we talked yesterday—

Brian: Yeah.

Christoph: And I like the stories because just for me, it’s not only the workshop, but it’s the person behind it. And so, maybe tell us more. What’s your background? What are you doing? [crosstalk 00:14:03]—

Brian: [crosstalk 00:14:03] so Christoph, if you guys haven’t listened, he’s a trumpet player you guys . So, we nerded out about some music stuff. But yeah, my formal training is in music, so I [foreign 00:14:13]. So, I have a degree in percussion studies and performance. And so, I have another career, I run my consulting practice and training that I do with Designing for Analytics, and then I’m also a professional musician. So, I play with theater, Broadway shows, and orchestras, and klezmer, and Irish music and all different kinds of things, and I have my own band, and stuff like that. So yeah.

Christoph: So yeah, for everyone that’s interesting to talk with Brian, not only from the technical side but also about other topics. And I’m always dreaming about a TDWI All-Stars band. And I don’t—

Brian: Oh wow.

Christoph: —I don’t talk about the Data Bros.

Brian: Okay. [laugh].

Christoph: I don’t talk about the Data Bros. That’s an inside joke.

Brian: No one wants to go there.

Christoph: [laugh]. But I talk about the TDWI All-Stars with because we have so many musicians—

Brian: Oh, really? Okay.

Christoph: —out here. Carsten Felden our Chairman, he is playing the drums as well.

Brian: Excellent.

Christoph: I have to start exercising again, start doing the trumpet again, and I’m pretty sure we will find some more.

Brian: [crosstalk 00:15:12] okay.

Christoph: So Brian, time is over and we jump to our other guests.

Brian: Great.

Christoph: It was so great to have you on. It is great to have you and I’m really looking forward to seeing your session.

Brian: I’m looking forward to this, too. Yeah. It’s going to be really fun.

Christoph: I’m really thankful.

Brian: Yeah. Thanks for having me.

Christoph: And so, have a great day, have a great workshop—

Brian: Excellent.

Christoph: And have fun at the conference.

Brian: Yeah. Good luck to everyone on the livestream, and the activities, and speakers and all that. So.

Christoph: Thank you very much.

Brian: Cool. Take care.

Brian: All right, folks, I hope you enjoyed my interview there—reverse interview, right? I guess I was the guest—with Christoph Kreutz from TDWI Munich in Germany. I had a great time at the conference and I really enjoyed working with the participants, mostly from different countries around Europe. If you’d like to hear more about that workshop when it rolls out again, just head on over to designingforanalytics.com/podcast and you can hop on my mailing list from that page and stay tuned for more information on that, as well as getting all the updates on the show here including the one page-summaries I sent out of all the podcast episodes, and my writing and future conference speaking and training. So, take care, hang in there. I hope you guys are having a great summer of 2022 right now, and stay healthy.