Value Driven Data Science: Boost your impact. Earn what you’re worth. Rewrite your career algorithm.

[00:00:00] Dr Genevieve Hayes: Hello and welcome to Value Driven Data Science, the podcast that helps data scientists transform their technical expertise into tangible business value, career autonomy, and financial reward. I. I am Dr. Genevieve Hayes, and today I'm joined by David Cohen. David is a data and AI strategy consultant with a background in supporting the Fortune 500 clients of both Big four and boutique consulting firms.
[00:00:28] He's the founder of Super Position, a consulting firm that builds collaborative workshops focused on data and AI related use cases. In this episode, you'll discover how to flip the traditional data science process by using visual storytelling at the start of your projects, not just the end to increase stakeholder buy-in and deliver more successful outcomes.
[00:00:53] So get ready to boost your impact, earn what you're worth, and rewrite your career algorithm. David, welcome to the show.
[00:01:01] David Cohen: Genevieve, it's a pleasure to be here.
[00:01:03] Dr Genevieve Hayes: Regardless of where you learned your skills, most data scientists follow a fairly predictable process when it comes to delivering projects. It's usually something along the lines of gathering stakeholder requirements, collecting and exploring the data, developing models or analysis, and then only at the very end creating visualizations and presentations to communicate results.
[00:01:30] We all do it because it seems logical and few of us have ever been shown another way. But what if this traditional approach is actually working against us? What if by saving the visual storytelling for the end, we're missing out on opportunities to genuinely engage our stakeholders in our work, right from day one.
[00:01:50] It's questions like these that inspired the creation of super position, a data consultancy that aims to flip the script on data storytelling. To begin with David, can you tell us a bit more about superposition in particular, how you originally came up with this approach of flipping the script on data visualization?
[00:02:10] David Cohen: Yeah, absolutely. So like you mentioned, the inspiration for superposition and for the concept of what we do really came about from a reflection on my expertise in the consulting space and in general of a career in data consulting. That, as you said, from a messaging standpoint. We data people are traditionally taught a very, what I would say, bottoms up approach to communication.
[00:02:36] So we essentially always work off of the details first. So as you said, working on the minutia of putting together data messaging and then working our way backwards into essentially what the message was. The challenge with that primarily is that from an impact standpoint, we tend to. Not ever hit the mark because we don't actually understand what our users need.
[00:03:00] So from a, day-to-day standpoint, in my consulting career, that was always a challenge of delivering products and delivering data, tools and processes that work for our clients is that if we don't have the ability to understand what they need from the top level and then break that down into use cases and its requirements we don't have the ability to properly serve the people that we build data things for.
[00:03:23] So that's also true of anybody that's outside the consulting space as well. Effectively. Anybody that works in data always works bottom up to be as minutely detailed as possible, and to really understand all the building blocks that go into a hypothesis backed by data, and then the tools that get built from that.
[00:03:41] So superposition came from, a reflection on that reality where I realized that, affectionately us data nerds don't have the ability to effectively communicate the value that we bring to our clients especially when we work in services oriented businesses, consultancies, and companies like that.
[00:03:59] So I wanted to build a series of frameworks and structures to help people like myself navigate that challenge of communicating effectively as a data expert. To visualize and both structure the stories that they tell to their clients and navigate the challenges of ambiguity in our work. So especially whenever us consultants deliver work to complex clients like the ones that you mentioned in larger.
[00:04:26] Fortune 500 type companies. One of the biggest issues that we face is ambiguity, especially at the beginning of designing our projects and, our approach to solve the problems that they have. Primarily because there's large, complex dynamics that are largely human at play.
[00:04:43] And when we data people do our work, we tend to forget that our work is human work largely. So the impact of it is primarily to the way that people do their work on the day-to-day. So we need to be thinking about it in terms of behavior change versus technical capability instead.
[00:05:01] Dr Genevieve Hayes: What interests me about superposition though, is everything you've just said I'm aware of from conversations I've had with a lot of other. Senior data scientists and executives. But what strikes me as unique is your concept of moving the data visualization piece to the beginning.
[00:05:21] Where did that come from?
[00:05:23] David Cohen: So it came from a. Understanding of where the value is in actually reflecting on our client's problems, which is at the beginning. So when you think through the model of delivering quality data products, putting the focus on. The start of a design of a data product or tool is how we can effectively ask the questions that get to the why of why we need data in the first place.
[00:05:48] So what business problems are we solving? Whose lives are we making easier through data, products, tools or in the current world it's AI as well. So, the impetus for superposition was to gamify and define structures that would make that process of. Understanding the why of data easier for both consultants and then the people that end up being our end users or clients.
[00:06:13] Dr Genevieve Hayes: And pictures make that easier.
[00:06:15] David Cohen: Yeah, right. So the thing with data work and the way that we structure it, especially in a consultative setting, is that it tends to be very verbal, we go places and we tend to largely lecture our clients on how to do the things that they need to do. So the journey of bringing them along so that they can see and buy into the work that we're doing needs to be more visual than the traditional aspect of consultative work.
[00:06:40] Being able to see it and experience the approach that we as experts have to build effective tools. Makes it so much more real for you as a user, as an end goal. So you can understand how we got to the answer versus just getting there and trusting us that we got to there the right way, .
[00:06:58] So.
[00:06:58] Dr Genevieve Hayes: So allowing people to see what's going on rather than just. Telling them. So it's show don't tell
[00:07:04] David Cohen: exactly, and, obviously the concept of it comes inspired largely by my background in data visualization, as you mentioned in that the world of database is largely concerned with making complicated things simple from a visual standpoint, and distilling very complex messages.
[00:07:24] About things that data tells us by creating very simple visuals that define how that story comes alive. And so the same is true for my work. It's effectively distilling data strategy into a gamified framework that can then help people like myself deliver their work.
[00:07:45] Dr Genevieve Hayes: Okay, so to make this a bit more tangible, can you walk us through an example of what this looks like when you are doing a project with clients.
[00:07:55] David Cohen: Yeah, of course. So imagine for instance, if you were a consultancy that works in the data space and has a client that is. Either poorly organized or does not understand how to navigate the challenges of implementing a new data infrastructure or just data capacity end to end. The challenge of enabling modernized data capabilities for them is that you have to first understand what their issues are from a data perspective
[00:08:23] quality, infrastructure, human problems, et cetera, et cetera. And you need to be able to showcase to them how your approach specifically is going to solve those problems for them. So the way that I structure superpositions workshops is to essentially set activities up that are contained, visual games that enable you to solve specific problems across that entire value cycle.
[00:08:50] So it's basically answering anything from the very first question, which is always what problem are we solving through data? And then creating a narrative thread that goes to. What opportunities do we have to capture value out of those problems? How do we capture those problems tactically, who needs to be involved in making those happen and so forth, so that you go from, I have a problem to, there's a potential solution that, the company that I work with is helping me with.
[00:09:18] And then here's the tactical plan to make that happen in reality such that the output of the workshops and the games is essentially a plan to make. The work real to make the data work meaningful and actionable.
[00:09:31] Dr Genevieve Hayes: Okay, I'm getting it. So we're talking about. Video game type games, so highly visual games, and rather than me sitting down with a stakeholder and asking them questions like, what are your business problems? Why do you need this? What led to you speaking to me the client or stakeholder would actually work through these games and that would allow me to gather the requirements that I need.
[00:10:01] In order to do my job.
[00:10:02] David Cohen: Yeah, and it's essentially adapting , a dynamic of workshops that exist in other contexts. So there's, you know, ux UI design product design, more generally , and other disciplines that use workshops as a framework for getting answers out of people's heads, essentially. So a challenge that happens when you work with large clients that have distinct perspectives on how their organization functions.
[00:10:27] Is if you are doing it the traditional way, getting answers from everybody individually, is a very time consuming, difficult and often frustrating task. So , the workshop constructs that I build at superposition are meant to bring everybody together. Get them to work with each other and then be able to get answers from all those people in one setting so that you as the expert, can then showcase what you're good at, showcase how you're gonna help your client solve that, and then build a plan to make it real.
[00:10:58] Dr Genevieve Hayes: Okay. Can you give us an example of one of the sorts of games that you would use in order to gather these requirements?
[00:11:04] David Cohen: There's a variety of them. I think that there's a very simple one. We call it how might we it's inspired by design thinking frameworks. I like to call it now how do we, and that it implies a degree of certainty in it, but essentially it's a very simple conversation based.
[00:11:23] Activity where we think through how we might serve somebody as a momentum builder for understanding what we're trying to achieve through data work. So for instance, if we are trying to build a modernized data infrastructure or platform for a company, it might be how do we achieve better.
[00:11:42] Metrics available for the CEO of our company so that they can better understand how to grow the company and rearrange their resources or whatever. So it's framing the outcomes that we're trying to achieve in the project, in the form of a question so that the user is able to think through both.
[00:12:00] What the outcome that they're trying to achieve is, but also who it's for. So like who it's affecting and then what outcome and what purpose it enables in reality. So it's basically structuring what typically is very difficult to conceptualize, which is all the parts of what a wishlist item from a data project perspective would accomplish.
[00:12:19] And then structuring it into three parts, like I said, what the wishlist item is, who it's for, and what purpose it does.
[00:12:27] Dr Genevieve Hayes: This is a game. How do you win it?
[00:12:29] David Cohen: So typically there's no winner per se, like the competition aspect is just to like. Drive people's engagement in it. I've built some workshops in the past where we do build winners and losers, so to speak, but it's really meant to drive effective work at the end of the day.
[00:12:44] So it's more bringing out the competitive nature and people, and then bringing out people's capability to work with each other rather than somebody winning and somebody losing per se. That's not the focus necessarily, but in most cases I actually don't build the activities to be very competitive.
[00:13:01] They're actually more collaborative.
[00:13:03] Dr Genevieve Hayes: Okay, so you're not having chocolate bar prizes or anything like that for this.
[00:13:07] David Cohen: No, I've met other people in the space that do build some of that in, and I think that's definitely a part of it too, is that to solve conflict sometimes it requires to have winners and losers, in that decisions. Prioritize somebody's opinions versus others. And sometimes that does require, a zero sum outcome to happen.
[00:13:28] Fortunately in my world, we can always make it more collaboration. So the activities and games tend to be more cooperative than competitive, so they're essentially building off each other to build a roadmap versus competing, and then there being a winner. Typically because the organizations that I work with , don't want to have somebody's opinion being valued more than somebody else,
[00:13:49] Dr Genevieve Hayes: so we're all winners.
[00:13:51] David Cohen: yeah, we're all winners.
[00:13:52] Dr Genevieve Hayes: I could imagine a low tech version of this involving butcher's paper and post-it notes. But from looking at your website, it sounds like you've gone, the high tech version you've got this. Visual aesthetic of your work. That is very much a mid 1990s video game aesthetic.
[00:14:14] It reminds me of Super Mario Brothers or the Legend of Zelda, which I wasted way too much time playing when I was a child. I love it because it reminds me of my childhood, but it strikes me as being a rather unusual choice. It's essentially the opposite of the business-like aesthetic that most data scientists typically strive for and are encouraged to use.
[00:14:40] What led you to make that design choice, and how do you tie it into the rest of your work? I.
[00:14:46] David Cohen: Yeah, it's obviously a very intentional choice in representing the work and myself from a brand perspective. Coming from the data consulting industry, what you realize is that, as you said, from a presentation standpoint, the work is traditionally very buttoned up, very stale, sort of.
[00:15:04] By the book. So from a branding standpoint, I wanted super position to be something that stood out both in its presentation, but also in what it represents about the work. So, as you said, the brand is meant to be very colorful. The visual aesthetic is meant to be very eye-catching and very obviously referential to the video game world.
[00:15:24] And obviously the gamification aspect of it, from a concept standpoint. It's also meant to showcase that our work in data is and can be more fun and can be engaging and can be collaborative. Because traditionally our space is very lonely, to be honest. It's something that most of us end up working.
[00:15:44] Not collaboratively, and we end up achieving outcomes that are not optimal because we don't have the ability to bring other people in. And so the reflection on that was that the brand superposition and that the work that I do should reflect all that. It should reflect the ability to collaborate and to build fun, engaging active both workshops and strategies that enable us to do our work as data people, but also.
[00:16:09] Not take ourselves too seriously doing so, we don't have to be overly formal in everything that we do. And informal does not mean unprofessional to me at least. And so that's a reflection of how the brand comes across is that it is the combination of that fun aspect with the expertise that I bring to the table.
[00:16:28] Dr Genevieve Hayes: How do clients and prospective clients typically react when they first see your visual aesthetic? I.
[00:16:33] David Cohen: It's a difference maker. It catches your attention, right? Because it's different. Either you love it or you don't get it, and it takes you a second to land on why it hits. So once you see the full picture of it, typically it makes a lot of sense when people understand that there is an intention behind it, beyond just it's a, QC video game aesthetic, then that.
[00:16:57] Clicks in people's heads because they understand the connection between color to eye grabbing visuals to the gamified aspect of my work and to the outcomes that the games provide from a structure perspective , as a consultant. So thing is that it just grabs your attention, right?
[00:17:14] And then you ask the question naturally that you've asked, which is, how does that tie into what you do? And so I get to tell you what I do and I get to explain the connection, and then it clicks in your head.
[00:17:23] Dr Genevieve Hayes: Okay, cool. And the clients you work with, are you actually helping data consultants to build these workshops themselves? Or are you using this for your own consulting work?
[00:17:34] David Cohen: The former, so I primarily work with other consultancies to help them navigate either internal or external challenges. So externally, obviously to collectively embed with them and help them build the workshops to deliver the early stages of. Consulting engagement. So effectively as the start point of either selling an engagement, if you can imagine convincing a potential client that a data project is worth doing or, actually structuring it once it's been sold and it's started so that the use cases, the requirements, the definition of structure for how we navigate the ambiguity of work as a consultant is.
[00:18:14] Facilitated through the workshops. Also, internally, I use the same workshops to help people that also do consulting work structure frameworks for delivering their work, for growing their company and just generally thinking about how they build their own consultancies as well.
[00:18:31] Dr Genevieve Hayes: The majority of our listeners. Aren't data and AI consultants, but rather employees working as data professionals within organizations, what aspects of your work could they potentially apply to their own work?
[00:18:45] David Cohen: The lesson that I would share is flipping the script that we talked about, kind of thinking about our work top down versus bottom up, in that when we think about data work, I. We are typically most effective when we tie our work to business problem solved. And asking the question why or what are we solving through?
[00:19:05] This is how US data professionals in all capacities are effectively able to affect change in the people that we serve. So for data scientists, for data people in general, consultants or otherwise, the outcome should always be, what are we doing this for? Why is the way that we're doing our work gonna contribute to an actionable result in people and
[00:19:29] at the end of the day, in behavior change, that's what US data people are trying to create in the work that we do. We're trying to enable somebody to make a decision. We're trying to make a organization able to adjust and adapt to something that it sees in the market. In many different ways. So the way that we approach problem solving is the best advice I can give,
[00:19:49] is that thinking about solving problems versus finding the data to solve those problems is the most appropriate and most effective thing we can do as data people.
[00:19:58] Dr Genevieve Hayes: For data scientists who wanna start implementing this approach, what's the. Single most important change they could start making tomorrow.
[00:20:06] David Cohen: So the biggest change that I made that made my work very different was honestly to start on pen and paper from a visual standpoint. Like write down the things that. You want to solve for and think critically and think structurally about the problems that you're trying to solve, and visually draw yourself a map of where you're trying to head.
[00:20:28] And that sounds super counterintuitive to start data work on paper, but having a map of why you're solving a problem that you're trying to solve. And for me it was in dashboards, because that's what I specialized in. So I built. Reporting and dashboards and stuff like that, it helped to think about the problem that I was trying to solve via communication by just making a wire frame and even more basic than that, by just making a bullet point list of the things and problems that I was trying to solve through the dashboard so that I could then use that as a design tool or as a facilitator tool.
[00:21:01] To say, am I actually doing what I'm supposed to be doing through this? And as a professional, as a data person, am I actually doing my job effectively in that sense? And it's surprisingly powerful. You'd be surprised how much it improves your work to actually just put down things on paper.
[00:21:16] Dr Genevieve Hayes: I probably understand. Yeah. It's the same process that I use. Everything starts with pen and paper 'cause I just can't think on computers.
[00:21:24] David Cohen: Yeah. The challenge is that especially in our field. The work that we do is a little ambiguous and ethereal, I suppose, once you start doing data work, unless you. Actually conceptualize it properly and put bounds around it. It just tends to grow into whatever the end users want it to grow into.
[00:21:41] Especially in many organizations that we tend to work in as data folks. We serve as support functions and we serve as amplifier functions for other parts of the organization. So the dynamic there is unfortunately that we are also a reactive function, so we don't proactively design and understand.
[00:21:59] How our work can evolve the business function, but rather reacting to whatever it needs. So in my experience, the first step to being able to become that proactive function is to think about it ahead of time and to actually put it down on paper so you can understand the whole vision of what data means to your team.
[00:22:18] Dr Genevieve Hayes: And then once you've got it on paper, the big advantage of paper is. Computer files can extend pretty much forever. You can have a million page document, theoretically, not that I've ever done it, but if you're gonna put things on paper, you're probably gonna limit yourself to less than a dozen pages. And that means that you can see it all at one time.
[00:22:41] David Cohen: Yeah. And I would say that one of the founding principles of superposition and the things that I most advocate for is that artificially forcing yourself to have a limit in the amount of. Information that you can work through is what forces you to think critically about what's gonna have an impact as a data person.
[00:23:00] So if you're building a dashboard, for instance, having a literal limited amount of paper that you can draw on in terms of information that's gonna go on it is what forces you to think, oh well, is this actually needed for the message that I'm trying to get out there? And how is that gonna impact the end result?
[00:23:16] Dr Genevieve Hayes: For listeners who wanna get in contact with you, David, what can they do?
[00:23:20] David Cohen: So obviously Vista our website, superposition strat.com. I'm also pretty active on LinkedIn as well. I post a lot of thought leadership on both data strategy, AI strategy, consulting in general, and just my thoughts on the data world as a whole.
[00:23:35] And I also have a YouTube channel that I post content on as well.
[00:23:40] Dr Genevieve Hayes: And there you have it. Another value packed episode to help turn your data skills into serious clout, cash and career freedom. If you enjoyed this episode, why not make it a double next week? Catch David's value boost a 10 minute episode where he shares one powerful tip for getting real results real fast.
[00:24:01] Make sure you're subscribed so you don't miss it. Thanks for joining me today, David,
[00:24:05] David Cohen: Course, happy to.
[00:24:07] Dr Genevieve Hayes: and for those in the audience, thanks for listening. I'm Dr. Genevieve Hayes, and this has been Value-Driven Data Science.