MindSpeaking Podcast Episode 12 - Nicholas Kelly, Principal consultant and trainer @ G&K Consulting
▶️ Watch the podcast with video on YouTube:
Timestamps:
00:00:00 - Introduction
00:02:50 - Tips for meeting new people for the first time
00:05:37 - Internship in Honolulu
00:08:19 - Understanding your audience
00:10:30 - Nick's Career Journey
00:15:00 - Combining business and a relationship
00:16:40 - The Adoption Struggle and how we can prevent it
00:25:01 - Definition of Value Mindset and the Production Mindset
00:28:44 - Tips for talking to stakeholders
00:30:22 - What do you see as the role of a data leader/manager vs. an individual contributor?
00:35:33 – IT and business in one room. Funny story 00:40:08 What's your definition of data storytelling?
00:48:15 – Data viz tip: The Data-to-ink ratio
00:54:00 - About the Dashboard Wireframe kit
00:57:59 - Fun/Quick Questions
00:58:47 - Practical tips for communicating as a Data Analyst
01:00:40 - Key takeaway of the episode
01:02:53 - How do you keep balance? What is your focus at the moment?
01:05:03 - Where can people connect with Nicholas Kelly
01:07:09 - Conclusion
🎙️Listen to your favorite channel:
Introducing Nicholas Kelly
Gilbert Eijkelenboom:
Nicholas Kelly is the author of the book delivering data analytics, and the hands-on leader in BI and analytics with almost 20 years of international experience and we have a lot of stories about that international experience as well that we're going to hear in this episode. He has done a lot of coding by himself, but also has managed large data teams of 25 data scientists, and now he has his own business with a focus on analytics, adoption, and user experience. He's running this business together with his wife, which I admire because I cannot imagine doing a business or running a business together with my girlfriends. But joking aside, what you can find in this episode is tips. For the data analytics, communication, about visualization and creating dashboards. And especially we're going to zoom in on how to increase adoption of dashboards and how to have conversations with stakeholders and end users so that you actually build something that they will be interested in and use. So he's another person, bring the data and the business together. So I'm excited to introduce to you Nicholas Kelly. Nick, great to see you again.
Nicholas Kelly:
Likewise, Gilbert,
Tips for meeting new people for the first time
Gilbert Eijkelenboom:
I'm really looking forward to this conversation. We had a conversation before I think it was two months ago or so. That's when we got to know each other and I got excited because we share a lot of interest about it. We'll look at creating dashboards, understanding stakeholders, understanding end users and data storytelling. And of course, I'm gonna dive into those topics. A lot today also dive into this wonderful book that you've written, delivering data analytics. And because, as I mentioned, the last time we met, I saw the passion for data for dashboards for user requirements, but also people focus. And what I felt in a conversation right from the beginning is that you made me feel at ease you made me very comfortable, you were interested, and that spoke from a lot of people skills and of course, that's a topic I'm interested in as well. So I was wondering, if you meet someone for the first time approach, give us some of your secrets or what have what are some of the things you do when you meet people for the first time.
Nicholas Kelly:
Awesome and Gilbert and thanks for having me on and it's a right back atcha love your book, loved loved the material and I think that's why there's a good affinity here. We're covering the same topics, maybe from slightly different approaches, but when I'm engaging people firstly, let me first admit that I'm I'm an introvert I did not like working with people. So I had to figure out what would work for me first. And so I ended up early on in my my journey with this was coming up with a set of questions. If you are meeting someone for the first time and you're going to like interview them and figure out what's important to them for their data and their business and their you know, their their day to day. One of the questions you should ask and Chris, there's lots of methodologies out there. I just usually ask with simple methods like you know, Hey, how's your Monday What is your Monday look like? Tell me about that. Right? And it's as simple as that it just grinds it into something that they can talk about whatever direction they want to go. They want to talk about something personal, hate great. Or they want to talk about, oh, you know, what, my, everything's manual. It takes me hours to figure out what's going on with my data and I'm fumbling around direction they want but the main thing that's happening is we're just breaking the ice you know, we're we're talking about something that's hopefully either a challenge for them, or it's just something they're really excited about. And so we can pick up on that and take that in the direction that we need to go. So there'd be some steering around it, but usually, it just some form of icebreaker. We're not getting straight into business.
Gilbert Eijkelenboom:
Right? I like that and what I like about is that is a very open question. You give people the opportunity to take it in any direction. And at the same time, it's also it is more specific than just how are you right, which many people answer with I'm fine. I'm good or doesn't really know what to answer if people ask me that. I'm also thinking, hey, how am I you know, really? I don't know. Fine, I guess. So. It's very tempting to have this, this auto response, right? There's robotic answer that doesn't really
Nicholas Kelly:
represent and it could go the wrong way. Of course, you know, like it because if we leave it, if we ask that kind of question, they say, Well, fine, you know, like are in Ireland, or they would, they would say, Grant and grant and it's like, you know, it could be hard to pick up after that, you know, pick up the energy after that. So, I completely agree. You're going something that's a little bit more requires a descriptive answer, rather than the one or two word answers. It's certainly going to be easier to drive the conversation the way we need it to go.
Internship in Honolulu
Gilbert Eijkelenboom:
Right, exactly. You mentioned Ireland. Tell us a bit about that. And before we dive into the data, career and more the business side, tell us about that. And I also heard you speaking about this internship in Honolulu, where they struggled with your accent.
Nicholas Kelly:
Yes. So yeah, that's interesting story for you there. But, like born and raised in Ireland, and of course, you know, wherever you grow up your accent is, you don't think you have an accent. And so, part of my internship, you have to do six months in the university that I went to University College Cork, six month internship, most people in my year went to, either somewhere in Cork or Dublin, that the ones that were like very exotic might have gone to the UK. And not to be outdone, I wanted to go somewhere even more exciting. I wasn't really focused on what the internship was more of where it was. So, I ended up in Honolulu working for a virtual reality company, and my first week I did 106 hours. And it was ridiculous, like, getting getting getting up at, you know, four or 5am and coming home at like 12:12pm midnight, or 1am. And doing that every day of the week. And it's just really hard graft. And one of the things that was because as a virtual reality company, and this keeps sticking with me and keeps coming back, was it was really my first foray into visualization. So you know, of course in the 3d realm, and we always hear about, you know, 3d charts and 3d this and they're not optimal. And now when we're in the news, we have met her right for virtual bigger investments in virtual reality, and why didn't it take off? And I think just great parallels for us here because most types of information are best conveyed in 2d. And so when you're in the virtual realm, in 3d, yes, you add another dimension but you also have to add another action, a rotation to properly understand perspective of the 3d environment. And it struck me back then this is like late 90s 2000s 2001. Why didn't VR take off then? And why? Isn't it taken off now? And so it's very interesting, but during that internship, I had that my CEO of the company, he told me a few weeks into it, he said, Nick, you know, I only understand about 30% of what you say, and I was like, oh, that's awkward. So I tried to start making an effort. To try and speak like a normal person and not the Irish accent.
Understanding your audience
Gilbert Eijkelenboom:
That's funny. And it shows that you know, you need to you need to be aware, right, how you communicate and how people other people perceive your communication. I think that's such an important aspect. Also, for data analysts listening or listening or data leaders. Were talking to stakeholders or end users who might not be as data savvy as they are.
Nicholas Kelly:
That's a great point, like, like the, the movement of data literacy, you know, over the last few years. Really interesting. I think it overlaps very well with what you just said, is we don't want to assume and make any assumptions on someone's capability and understanding of how to use data. And it's one thing that I find that I almost always undervalue is when I'm doing any workshop or session with with a group of people that I have to preface it with some sort of training. Like, let's just establish a common set of definitions and how we're going to talk and but I never want to do that because I have done it. And failed. I never want to do it in the form of, I'm telling you what this is, right? It's like, I'm kind of treating you like a child because you don't know this and that definition. It's more, let's just talk about a case study. You know, and it's like this organization, you know, learned this outcome by using data in this way. Right and there's like, everyone's going, oh, yeah, that makes sense. Right. And as we went through their case study, you know, they learned some different definitions and things that they need to know in order for us to be successful. And so I agree there it is really important that we talk in a way that we properly understand our audience and who we're truly designing for, not who we think we're designing for, but really who we are actually designing for,
Nick's Career Journey
Gilbert Eijkelenboom:
right? Because so often also, the stakeholder we talked to is not the end user right of the dashboard. The stakeholders is a person that sees business value in the dashboards, hopefully. Hopefully it is not just a political reason or anything else. That doesn't make sense. But yeah, often those two people are different. And we're going to dive into that a little bit later as well, which are when we talk about your book. So before Before diving into that, I would love to hear a bit more about your career journey where you started and how you ended up here, behind the microphone and having your own company.
Nicholas Kelly:
Absolutely. So I actually started in I really wanted to get into VR, so lucky, kind of we'll carry on from that journey back there. And Mike The other option I had for an internship was actually working with a architectural visualization company in Oregon. So I have two options. I could have gone VR, AR, architectural visualization, both 3d but different. One is you're inside the 3d realm in VR. And the other is you're outside and you're using it for architectural visualization, which is you know, hey, let's build a model of a 3d building, put it in the environment before we actually build it so we can see what it looks like what's the environmental impact all those things? That's where I was going with my career. And so I kept going more and more 3d. game engines, game physics, right like code was very much into writing the code, but then also doing the interface. So designing the interface, but it got so much so that it became very apparent that there was major skills lacking in user interface design. And as I got into user interface design, I started to hear distinct user experience. What's this user experience thing? And, again, the projects I was involved in ditches was no one doing that. So I ended up being the one that was like, Okay, I'll try and figure this thing out. Like I'm not formally trained in user experience, you know, but just trying having conversations with people. I think, Okay, well, what do you need, and just kind of cobbling this stuff together. And so over the years, I got more and more into that, where I eventually ended up in Singapore, working for a mobile application company as the Director for User Interface design and user experience. And over there, I mean, I've learned lots doing all of that stuff. And I ended up being approached by Deloitte analytics at the time and they were just saying, okay, look, is there some way we can figure out how you might come and work in Deloitte in data visualization, and we ended up figuring out something that made sense right so we I honestly didn't know a lot about analytics, right? It was like a big gap for me. I knew Google Analytics. Right? But that was it. So I didn't really know what analytics meant. And so I ended up working in Deloitte and I need to like because my my wife worked there. Completely different department, but I was very nervous because I know they're like, top tier consulting firm. Right? So I was like, Oh, what's this guy from Ireland going to be able to do here? You know, and, but what I noticed was pretty quickly, there was the same gaps that was in at least that I noticed in the software realm, and user experience, or also in analytics, if not more so. I sort of the breakdown between what we what the business was getting from analytics and what they actually wanted, right? There's this big just a big gap or what the and what they needed. And so while there was like super smart people there, where I was able to add value is okay, well, let's try and just bridge that gap. Let's try and bridge the gap between what's going on with with data and and what's going on with people. And that's really where I found myself is then getting very much into data visualization, as a good means of being able to bridge that gap. So I ended up focusing a lot on the people side of it, just because that's where I thought I could add value and I knew I couldn't add value on the data side. Because there's like way smarter people than me they're doing that piece so and then that's ultimately I ended up moving with the firm from Singapore over to the US and then I did a stint with a really great company called logic 2020 as a consultant after Deloitte, and after that, then we went out doing our own thing, my my good wife, Maria, and I decided okay, let's let's give this thing a crack. And do our do our own business. So we do consulting to keep us honest. Right? I I'm a big advocate of yes, you can have the theory but if you're not applying it, then like it's stale. Or maybe it's just not applicable. You know, things change, right. So, we have we have that now. We're still working as a consultant but then we also do a bunch of other things like training and different products.
Combining business and a relationship
Gilbert Eijkelenboom:
And you describe yourself was very hands on right so that's what you mentioned with consulting and also, you know, actually doing the work. And what I what I was wondering because I have to ask because you you you've been working with your wife together for a long time. How is it to combine business and relationship?
Nicholas Kelly:
I think it's part of it is just having clear definition of roles. We also have two offices. Or we're at right now we're in our in our studio. And so this is really where I spend my time and she spends most of her time in our office and right so and sometimes we will be there when we need to be there together. And then other times we're we're separated when I'm doing video or content or or other things so let's say a big house helps. And but but also, since we both worked for a large consulting firm, there's a lot of commonality there as well as like cultural similarities, right? So I think all that helps, because we are around each other a lot. And so I think it's a good question and we don't really experience any major friction because of it. Because of running the business together because she she's got her side where I am clueless, like truly clueless, finance terrible, with finances. And then I've got my side, which is the analytics piece. So there's a nice, clean division of roles and responsibilities.
The Adoption Struggle and how we can prevent it
Gilbert Eijkelenboom:
All right, and it's it's wonderful how you can complement each other den. sounds. Sounds great. So now we have some background write about you, where you where you're coming from and what's your what's your background and and some tips about how to connect with people when you speak to them for the first time. Now, I would love to dive into into your book. And I would love to start more at the beginning, where you talk about the adoption struggle, something that I recognize and I'm sure many other people recognize, can you briefly share what you mean with the opportunity struggle, and maybe already well, how we how we can prevent it?
Nicholas Kelly:
Absolutely. So we, we've, let's say like historically, we've overcome many things in the field of analytics, right? So we've started started figuring out okay, how can we use data, right to tell good stories and figure out what we need to do and we've done amazing work with data architecture, data structure, structure, performance, all of this brilliant stuff that has to be done. It's very, very important. We've moved on to data science where we're now like forecasting and predicting things right, very get very advanced with it. I think the next frontier I put it should really have been one of the first frontiers but I think the next frontier is getting people to actually use it. I'd like this stuff's brilliant. All of this stuff's needed. But if people aren't acting on it then we're leaving value on the table. And that sort of brings us into this question of value, as well. Why are people not using it? Maybe they don't realize the value there are hasn't been communicated to them. So one of those skills I think that is unfortunately lacking in many data professionals, is sales, which sounds nuts, right? But it's just can you sell what you've made, right? And if we look at a data, whatever we're doing with data as a data product, let's say you've made a data warehouse, so that's a fantastic data product. How are you selling that? How are you selling that to the people that need to buy it as a war? So the adoption struggle is really how do we get people firstly understanding the value that's there for them, and then getting them to realize the value so that they can use it more, and instead of people trying to be please use our data. It's more are our market going, Hey, give me more. I want to work with you more. This is brilliant. I love this stuff. I take my money, right? So that's the adoption struggle. We really want to flip that script from exactly and
Gilbert Eijkelenboom:
that's also when I when I talk to organizations many managers that director VP level they they tell me that this is exactly what they want to achieve right they want right now they're trying to push data and dashboards more on the on the business, whether they like it or not. And often they're frustrated that they're they don't like it right? Because there's a whole lot of hard work getting into those dashboards and predictive models. So I can imagine you're frustrated when people don't use it. I've I've had the same feeling too. But what they where do you want to go is that there's more a pool from the business that that who say hey, you guys need you need to help us these are our problems and help help us to solve them with your solutions.
Nicholas Kelly:
Yeah, it's like same as saying Gilbert like it can it can be very frustrating for people. And then like the challenge, then it's like, I'm not done, right. Like I I still have to get convinced people to use this. And they don't necessarily have the tools available to them the skills available to them to figure this out. And that was one of those aha moments for me some years ago, where it's like, oh, there's you know, I think I figured out this, you know, this approach to getting people on board and then I stumbled across over change management. Oh, yes. This is awkward. It's been around for years. gets there. There's lots of it. Lots of different methodologies. And I think, while we can't expect people to be experts in change management, one thing we could do, and if people get something out of this conversation is have a change awareness, just an awareness that whatever you're doing is probably not going to land the way you want it and it's not going to be adopted. And if you have that assumption, then you might approach things differently. Right? You might seek to get people on board earlier, you might say, hey, you know, Mary, what's the biggest problem you have? It takes me four hours to build a report and have to report every week. Okay? What if we made that five minutes or we change that four hours into five minutes? What would you do? What would it mean? To you? Oh, just be fantastic, will completely transform how I, how I work on a one day, I wouldn't spend four hours doing this, I would be able to do this. And okay, great, Mary, can we use you as a champion and get other people here in your story? And it's like, yeah, okay, great. All right. So we might approach things differently. Even just with a change awareness, because a lot of this stuff is common sense. Like, some of it's not, but a lot of it is right, and many of us have just experienced in dealing with people. And if we can tap into some of that experience, right and say so Oh, yeah. so and so said, This is great. Well, I'm going to do that as well. Right? Just social proof, right. So I think there's some of these things where we can look at and go adoptions hard, but if we know it's going to be an issue, we might approach things a little bit differently when we're getting into it.
Gilbert Eijkelenboom:
Exactly. And probably from the beginning as well right from the beginning before you create your stuff instead of working three months to make the perfect dashboard that no one would use, maybe involved in from the start from the beginning and ask all those questions and and I think what what what is the struggle of many data professionals is that they are a bit more introverted or not used to talking to people or being proactive and in communication. You need to have some social confidence right to approach one of the end user or stakeholders and have this ideation conversation with them. What about this value what they need and then think about, okay, how can I solve that problem rather than because it's way safer, of course, to think from the data first and then try to push the solution that you hope that people will use
Nicholas Kelly:
the risk? The risk is potentially great as well because if we go do something for three months, and then we put it in front of someone for the first time, and it's like, well it might land great if it does, but it also might not, right, and then even greater friction might be introduced, right? So if we can just chop up that friction, like you're saying there and maybe chop that up over time, right, like get them involved at the start, and maybe there's a little bit of misalignment but let's deal with it because we've not invested any time. But like you're saying often folks are on the data side a bit more introverted. And I certainly used to be and I remember early in my career in software, what I really should have done on gathering requirements was talk to the stakeholders. I instead sent them a form a survey to fill out and it just created more friction does like what are you doing? Sending a bunch of executives a form to fill out? Like, what are you doing? Right? And it just created all this unnecessary stress that didn't need to be there. If I just decided okay, look it's gonna be awkward, because I'm not comfortable talking to people, but it's better than the alternative. And I think there's really, there's not, there's not an easy answer around. If you are more introverted, are there ways you can elicit requirements from people and it's like there are right there. At some point, you're probably gonna have to get in front of people, but use proxies use like, something that takes the attention off you, right, like, Hey, we're going to follow this best practice. We're gonna follow this process. So it's not it's not about me. The tension is on this process, right and all your role is to guide people through a process or an approach, right, or a methodology, whatever, whatever it is, but it's one way that people that are more introverted, can help have a support, to go through this process of engaging the business and understanding what they really need.
What do the Value Mindset and the Production Mindset mean?
Gilbert Eijkelenboom:
Right, that makes a lot of sense and, and once people buy into this philosophy or this mindset that you that you described and what when they see the risk, you know, building something and working on something for so long and not seeing it being used then I think people will be tempted to change and and in the book, you talk about the value mindset and the production mindset, can you share what you mean with that, and how do they fit in the in this context?
Nicholas Kelly:
Yeah. So when we think of the, like adoption and driving adoption, and we think of our everything has a product, right? So let's say my favorite topic dashboards, we think we're really in dashboard, we're building a product. And once we have that mindset, like most of us have not built products. And most of us haven't needed to sell a product but we really need these skills. Like they're just so important in the modern enterprise. Because in the modern enterprise, whether we realize it or not other people are selling their products to the same people, right? They're probably not competing products, but they're you're trying to vie and compete for attention. So people are going to react to whatever has the highest value our biggest pain right so Okay, CEO shouting at me I need to get this thing done. Okay, I'm going to deal with that right are this thing is going to net me the biggest benefit. So that's top of my priority list. I'm gonna deal with that. Once we take the mindset of a product approach, anything we build, we realize we're going to have to sell right so once we know we have to sell it, it just completely shifts how you go about doing anything. Because well the first thing is is like what if I have to sell it? Why would people gonna buy it? Right? Like so why would they buy it? Alright, so it's just one of the it's such a basic thing in sales. But we have to figure out what the market wants. Does the market actually want this product? Will they actually buy it? And who is that market? And to figure that out? Of course, there's plenty of methodology out there. We have to understand who we're designing for, what their goals are and what their challenges are. Right. So then we end up with the personas and figuring out okay, well, here's the different scenarios they have. Right? This is their day to day and this is how our product can help them. And then when we build our product, we're building it with the understanding that we're solving these problems for people, but also how we're going to communicate our product to them later on. When we want them to buy our product and buy a person most cases in the enterprises just use their time right? Like just I'm going to go into this dashboard and use it right. So that's a proxy for making a purchasing decision, right, but we want to keep them there. So once we do that, we take the product approach. It really transforms how we look at our whatever we're building, in this case a dashboard because we have we know we have to sell it later on.
Tips for talking to stakeholders
Gilbert Eijkelenboom:
Right? I really liked that because it forces you to dive into the motivations and pains and challenges of the of your end users. And I think that's that's a mindset that everyone needs to adopt. And and I think by by asking the right questions, able to call them business value questions before you accuse, you can get her dose those requirements, right?
Nicholas Kelly:
Well, I when we put like, because we could just say they could ask business questions, but I deliberately call them business value questions. So we can just both remind ourselves, okay, there has to be value in there. But also let the person we're talking to understand we're trying to drive towards value as well. We're not just like, we're not going to create lots of junk on our dashboards or interfaces just because you know, the data is there. We there needs to be value implicit in it. So that that's also why it's like a business value question.
Gilbert Eijkelenboom:
I like it. I think the focus on value is important here. And do you have any more tips next to the ones you share it to? For talking to stakeholders or to end users?
Nicholas Kelly:
No, I think one of the biggest ones for me over the years is having a sense of humor. And also not taking yourself too seriously. Because I've certainly done that, particularly when working with senior executives, where, you know, people have said things to me is like, oh, everything you just said is BS, you know, and it's like, well, I'm at the time I took it very seriously, right? Whereas a better people person could navigate those things a little bit more elegantly. So I think for me to two pieces of advice is if it works for you because you don't want to force humor, but if it works for your humor is great. It's a great way of getting to the truth of things. And then also try not to take your tips too seriously, because it can engineer friction and stiffness. And that's going to come across as well. Right? So people are just going to react to someone who's stiff or they just want to get through it and sort of mirror what you're doing right. So I think those two things is going to make your conversations and getting to the core of people's challenges and goals. That much easier. Right?
What do you see as the role of a data leader/manager vs. an individual contributor?
Gilbert Eijkelenboom:
If we're talking about a conversation with stakeholders and end users, sometimes I hear individual contributor contributors say, you know, as data scientists or data analysts say yeah, my product owner needs to do that or my manager takes care of all those conversations. What what do you see as a role, you know, if we talk about increasing adoption and change management, what do you see as the role of the data leader or manager versus the individual contributor? contributor?
Nicholas Kelly:
Yeah, it's it's a tough one. I mean, I think it comes down to you know, who owns adoption. And the challenge with that is okay, well, let's say ultimately, it's the director of analytics director of IT bi from in terms of responsibility, but who's the one actually doing that? I usually advocate for who's ever building the dashboard. In our example, whoever's doing the building needs to have at least an awareness of okay. We're building this so it needs to be adopted. Generally speaking, what I find works well is it's a, it's a partial role for someone. It's like, okay, within our analytics team, you are the person who's going to be responsible for measuring and driving adoption. For all our data products. It's not enough i Well, depends on the scale of your organization. But generally speaking, it's not enough for a single person's role. But let's say high level, the director should have assigned ownership and tasks related to adoption and at a minimum, put measuring adoption as a metric that they're going to measure themselves by to see if they're successful. So that you know, like, it's going to be a 5% adoption. Let's get up to 30%. Right, and even just that, right, just even just that is something but I think ultimately it needs to be assigned as a partial role to someone within the BI and analytics team.
Gilbert Eijkelenboom:
I like to bring this up because we in the data space are very passionate about numbers about measuring stuff. Why don't we become a bit more obsessed with measuring this piece right, the adoption How To which extent people use our work, isn't that a really important metric to to measure? Right?
Nicholas Kelly:
It's such an important metric, like because there are ones out there that you know, ultimately, let's say we are want to use data to increase our revenue by 5%. And any organization that's already like aspiring to put a percentage or a measure to a outcome of using data is great. Like that's already fantastic. But in lieu of that, often, we're not there. We're not there at organizationally as a team as people willing to commit to something like that. So a really good proxy is let's measure adoption. Now. The downside of that can be waterfall, we're measuring the wrong things. And what if, you know, just the metrics here aren't useful and we're driving adoption? Right. So now we're getting this in front of people, a bad product in front of people, right, that could long term impact adoption, so we have to be really careful about it. So I recommend measuring adoption. Once we figured out the you know, the data's the data side of it. Are we showing the right things? Are we showing the right metrics, even if they're not super valuable? Initially, maybe the data doesn't support that. Right, that we can have really good metrics, but we have something there, but it's well thought through. Then we can focus on the adoption piece, right but we want positive adoption right for it's going to drive more and more adoption, not like, Hey, we've got 30% adoption, but now more people just got exposed to bad bad insights and bad data. Right? So that's going to be more frustrated people, and it's gonna be even harder to work against it. So it can be a double edged sword. So I think once we figured out we've done everything right, we should measure ourselves on adoption. I completely agree as data professionals, we we need to be keeping ourselves like, yes, we're measuring the business in the sense by giving them access to data, but how are we measuring ourselves and our own performance and adoption metric is a good one, not the only one, but it's a good one.
Gilbert Eijkelenboom:
Right. And you also already mentioned a few times like struggles and difficulties with, you know, with these, this process and difficulties with stakeholder conversations. You also you've traveled a lot. You've lived in Singapore, us, Ireland. You've been to many places. I'm curious, do you have any stories, any stories about stakeholder conversations that went bad or dashboard adoption or, or workshops that didn't work out?
IT and business in one room. Funny story
Nicholas Kelly:
Like there's definitely like cultural differences I've noticed working in Asia and then there's just like proper expectation setting like for me, always the hardest work is working with people. And like, by far, like it's not even close, like give me bad data any day. If that's your only problem then fantastic, right? But the hardest part is working with people. So like I have had several workshops and I really shouldn't admit this because it's on me, but we're, you know, we may be we're supposed to have 2030 people show up and it's like one or two people show up, you know, and there was one instance in Jakarta where we had, we're supposed to have like a large group from it. Join a large group from the business join. And I was there I was ready. I had everything ready. Like, whenever in a workshop, like I do all this prep beforehand, print out big posters, right. So you know, okay, here's what analytics is and here's what Analytics does in your industry. And here's how we should be thinking about these things, all these kind of breakout groups who were going to have the food's ready like catering, all the stuff no expense spared and start school start at nine o'clock bypass nine. I've got three people in the room at 10 past 911 More person struggling and I'm like, oh, what's going on supposed to help like maybe 3040 50 People in this one? And I ended up finding out that it really hates the business. I mean, probably no big surprise, right? But they really hated them. And the feeling was completely mutual. So when they looked at the invite for the event, they just saw Oh, so and so's going. This is going nowhere. I'm just gonna you know, my time is better spent doing something else. And it was like a complete waste of time. That was supposed to be a few hour workshop. And you know, we'd flown over from Singapore over there as expensive client to paying for it. Maybe they gotta pay for it anyway. Right. But we had to reschedule it. And what we did the next time around is we just did a free lunch. And what was interesting about that was they had no restaurant in this building. So if they want to eat food they're getting in their cars are a bus and going to the nearby mall to eat are they're bringing their food with them and so it was it was somewhat novel, say, oh, cool, this free pizza. Right. Great. So um, we did two invites, we sent out two invites one was to the IT people and one was to the business, but it was in the same room. And so as they were recovering in our we just stayed there just to make sure Okay, yeah. Go over there. Get your pizza, right, like, no focus on who's there anything like that, right? And I just made sure no one was leaving. And people had conversations, not about the project. But they realized that the people across the other side and it there were not terrible. They had, you know, their own struggles. And, you know, people heard from it. Well, you know, the CEO wants me to clean out this broken mess. I have to get back to him. You know, and I've got my priorities are like, you want me to do this dashboard. Last minute, right? So they got to hear and it actually took months to get them back together again. So we can actually work on a project. But you know, those things, the people things and understanding the challenges and you've mentioned politics or your Gilbert, right, like having to deal with politics, being aware that we have to deal with politics. And then finding out what the politics actually are. Like, that's some of the hardest stuff and working on working in data is because data can surface a lot of things that potentially people don't want to surface. We have to navigate this and figure out okay, well, maybe we can't do everything we want with data right away. Maybe we just have to see this as a journey. And that's the perspective that I take that I've learned over the years and dealing with people is, look, it's all just gonna be a journey. You're not going to get the results you want in a week, maybe two years, maybe three years, right. But if we take that approach, right, we're not just and I love this idea of like data storytelling, because there's the storytelling in the data, which is really important. But then there's the storytelling that goes on of the people around the data, as like, how do we bring them on our journey, the kind of longer journey of driving adoption and getting them on board. So yeah, just over the years of just time to deal with people. I've had to reset my expectations on how long this stuff actually takes. And I think when we do that, it makes it easier. Right? And it's more realistic, we get more realistic timelines.
What's your definition of data storytelling?
Gilbert Eijkelenboom:
Right. And I'm completely with you I mean, numbers are easy, right? But working with people is a challenge always, you know, put people in the room. And definitely there will be some some hassle at some point. And it's also the fun stuff, right? It's by being curious. I think you can solve those problems most easily. And still, they're challenging. But I agree working with people is always a challenge. And I love the idea about bring data storytelling, not just to explain what the numbers are, but also to bring to facilitate the change process and to bring people on board to get their buy in and when you mentioned data storytelling. I'm also curious what what's your definition of data storytelling, because a lot of different people have different definitions or different views about what data storytelling actually is.
Nicholas Kelly:
I think the part of the challenge is is the medium that we're going to deliver in the forum. So when I when I first got into analytics and data visualization, I first thought of it as okay, I've heard this term data storytelling. It wasn't a big thing back then. But my assumption is okay, it's something you do in data visualization, right? It's related to data visualization. And then for me because everything sold at a dashboard, I just assumed it's like, okay, data storytelling happens in a dashboard. And it really depends, right? So let's say we're going to deliver a presentation to you know, a bunch of executives and in a business, we're not necessarily going to be presenting it in a dashboard. We might be, we might be, it might be in the form of slides. Right. And so, for me, firstly, delivery matters. Okay, how are we trying to tell the story is it through a dashboard that people are logging into, let's say once a week once a day, whatever? Are we trying to do it in a curated experience, where I've been able to create and craft the visuals to tell a specific story of something that I found in the data? For me, I think there it's two different things. Right. So firstly, they're both important. And it's it's a matter of, okay, well, what's the delivery method? So firstly, I think there's that then if we dive down into either those, so can you do storytelling? With slides? Yes. 100%, right, like 100%. Can you do it in a dashboard? Okay, it depends what type of dashboard we're talking about. Right? So if it's a, you know, just a typical operation dashboard, just look, here's what's going on with the business fine. Sometimes there'll be a story sometimes there won't, right like, whatever, Everything's just fine. Nothing of concern going on the business then. Alright, there's nothing really alarming. So there's no story, right? But more often than not the way I designed dashboards, which is probably different to most types of dashboards that are out there because I think most dashboards out there are more like a report. Right? It's just like, I agree. It's a report right? Like, and that's fine, right? And that's the place and it doesn't mean we have to do data storytelling and every single scenario reports have their place, you know, like, but for me, can you tell the story in a dashboard, I think you can. If you properly engage with the business at an appropriate cadence. So what's an appropriate cadence? For me, it's three months. A dashboard should have a life to it, and a cadence and an iteration to it. That both affords a proper adoption change to happen. And for people to react to the data that's in the business so that their action and reaction to what's in there is impacting their environment and the business and we need a feedback loop. So that what their actions are doing to the business and the environment are being taken, because they're moving the business forward. So that the dashboard now reflects that change in the environment in the business and the people and in the data. So all of these things coming together. So it becomes a living thing. So why when we take that approach, I think absolutely. You can tell the story in a dashboard. When you have a story every single time. Probably not. I'd like let's say yesterday, there was something calling your attention and your dashboard is like, Oh, we're out of inventory and that store and we take action on it. We restock the inventory. We're out of inventory for XYZ reasons, whatever. The story is there because we've designed our dashboard that way. Well, I look in today and it's like okay, no story, right. There's nothing there. It does, like, everything's kind of great.
Gilbert Eijkelenboom:
There's no complication or no no conflict right now. Exactly. So there's no story.
Nicholas Kelly:
So so what I what I've noticed is some people might say that's not storytelling. Well, in and of that moment, it's not storytelling. But I think it's a definitions challenge that we have is what is data storytelling. That's one part of it, but then why are we doing it? Right? Why are we doing this because we want people to take action. We want people to get value from their data. And that's really what you know, I think that the true purpose of storytelling is it's a vehicle of getting there. It's a means of getting there. It's not the only means of getting there, but it is a means of getting there. So I think sometimes we potentially lose sight of that, that the storytelling piece is really to drive behavioral change and behavioral outcomes. And it's a vehicle a very good vehicle for doing it. But it's not the only one.
Gilbert Eijkelenboom:
Yes, absolutely. It's a powerful vehicle what I see, to share with my Bombay perspective, what I see with many people is that they see just a visualization just one graph as a story by definition. I don't think it's true. You also need some some conflict, some attention in a story and weather. And I like your observation about the dashboard. Right? I think my view is on was when the dashboard is very hard to create to have stories because you can't really control the narrative, right? That might pop up a challenge that you then can solve, but then maybe it's outside of the dashboard. Maybe it's only the problem that pops up in the dashboard. But I agree you know we need to there's a lot of focus on storytelling and sometimes organizations come to me and they say, Oh, we we need to have a training about storytelling and ask them what do you mean you know, what can you tell more about the challenges? And then they talk a lot about Yeah, we don't understand what is the business problem and, and about persuasion skills, the sales skills that you mentioned, the business value, mindset, those things that are things that they don't explicitly mention often, but they that's actually the challenge. So it's not just the data storytelling during the presentation or saying, telling a story about the data, but also, you need to give a presentation that make sense in the first place. So I like that you emphasize that point because the process is so much bigger than just the last piece of the product, right? The dashboard or the presentation that you give
Nicholas Kelly:
100% And I think that's the challenge is we expect a lot of data storytelling. I think it's like the expectation. That's still there of data visualization is like what is data visualization? Well, you could say it is the representation of data in a visual form. Yes. But what should it be doing? Well, it should be influencing decision making based off what the environment is saying. But data visualization doesn't say that as a definition, right. So I think it's important definitions problem, in part a misunderstanding problem. Because like, like the example you gave there, okay, they're struggling with, you know, understanding what the business needs, right? Well, that's not really a storytelling problem. That's a Yeah, like you're saying, like, it's changed management, its requirements. It's, you know, right, like user experience, right. It's all these other things. That in by by not having experience or knowledge in other in those areas, people kind of gravitate to the term data storytelling as a means of getting there. Right, like so we are sort of expecting a lot from data storytelling and what it can and can't do.
Data viz tip: The Data-to-ink ratio
Gilbert Eijkelenboom:
Right? You You make a lot of contact on content on LinkedIn and also YouTube, about data. visualization as well. And today, I benched your videos a bit, which were very insightful, so I would definitely recommend people to to to visit our YouTube channel. I'll include a link in one of the videos you you talked about the data to ink ratio of ad worked after, can you briefly explain for people who don't know what that is? Because I think it's a powerful concept.
Nicholas Kelly:
Yeah, for sure. So there's two types of ink we use in a visualization. There's ink used to show data, and then there's ink used to actually show the visualization. What we want to do is have as much ink showing data as possible and minimize the amount of ink we're using to show the visual. So what do I mean by that ink to show data, imagine bars, right? So you got bars in your chart that's showing data. You might have lines in your chart to help you scan left to right. What's on to the scale but if we just put the number directly on the bar, we don't eat the lines anymore, right? So we're increasing the data to ink ratio by minimizing the amount of non data visual visuals or ink to represent it. Now, for me there's a there's a there's a balance as well, because there's also good design and so we don't want to completely forbid non data Inc. If it means the design is going to continually look very very bland, and I do this myself many of my dashboards visualizations are very bland because we want to highlight and draw attention, which is the purpose of the data to ink ratio. We want to draw people's attention when we need to. So there's huge value in doing that. But we do need to balance it with having some ink there so that it doesn't look completely bland. Your neck can just be font it could just be right just how we size the different placements right but but still minimizing it but yeah, that's in a nutshell, the data in ratio exists so we can draw the users attention to what really matters is the ultimate intent.
Gilbert Eijkelenboom:
And when I look at the average data analyst that can definitely benefit from this concept because what I see and hear from people is that many data analysts tend to include too much you know, they want to include too many details because they think they're important then then it goes back to what do does my audience find important? You know, to which extent are they interested in all these details? And maybe you still want to include more detail, but put them out on four separate slides, right? So you control the narrative and show it step by step. So I think that's that's very helpful and of course, it's, you don't want to delete everything, because there needs to be something there. But I think the main challenge for many data analysts is more to the Remove rather than to add.
Nicholas Kelly:
I think it's partly also with design struggle, because you could, you can, you can look at the data ink ratio, and that applies to the field of analytics. But you can also look at minimalist design in like many user interface fields, right. And it's similar concept, but it's like, we just want to minimize clutter where appropriate, but use it where it's important, right? Of course marketing does it all the time with brand, right? So it's a similar concept is we have people looking at something visual, we want to control where we draw their attention. And the data to ratio is a way to do that. Like To your point, right, like a lot of people in any field, don't know how to design and so the data Inc ratio is a methodology to help people in the analytics field design. Well,
Gilbert Eijkelenboom:
exactly. And also make me think about this image that I saw on the internet, where they first showed an average employee Employee Application from a big enterprise right with 500 fields that you probably don't need, but you actually need to fill them in, which is totally not, not a great user experience. And then next to that, a picture or image of Google right Google the homepage, this one bar, we can type in, and a search button. And that's it. So it's, it's a beautiful way to show that contrast and also to convey the message of simplicity and the power of that
Nicholas Kelly:
it's so important. What user experience can do for analytics is you know, like your example, like imagine how we used to search the web, it was by category, right? So like, you just have all of these things you could click right. And when you look at who developed that interface, it's a software developer, right? So what's important to them features because they built the features. So the interface lo and behold has features all the features that can fit on the interface. But then you come along with a user experience professional, they say, what does the user need? Completely different perspective. They're Worlds Apart. A user just needs to find what they're looking for. Right? And so Okay, well, how do we best facilitate that search bar, a extensible, scalable search bar that I as a very basic user, I can type in a simple natural language term are, if I'm more skilled and capable, I can query it right with with in different more detailed, sophisticated ways, so I can narrow my search down even more. It's a brilliant example of what happened to the realm of software and interface design, and what's happening needs to happen. In the analytics field, we need a similar shift.
Gilbert Eijkelenboom:
Absolutely. One of the ways to get there is the dashboard wireframe kit to collect those requirements early. It's something you created. It's kind of a, it looks like a card game. But of course, it's a professional tool. Can you tell us a bit more why why you created it. And I think it's always fascinating when people use kind of gamification or creating game of that facilitates the conversation, right? And if it doesn't happen, we have this crazy application with 500 fields. So to facilitate that conversation, and create data products that are actually people want to use. So tell us a bit about the dashboard wireframe kit.
Nicholas Kelly:
Absolutely. So when I was working, doing workshops, so I've been doing like day in and day out doing workshops with executives on what do they need to get from their data, and where's the value? And when I started doing that, I was like, hey, awesome. We've got these really high tech touch screens. They look fantastic. We got this view of you know, the Marina Bay Sands in Singapore, no expense, spared beautiful space. And I was all about using the technology. And to my, to my disappointment. Over time, it's just became very obvious that people like using sticky notes a lot on whiteboards, so like very low tech, and it's like, Guys, we're working with data this is cutting edge, stop using the low tech stuff. And but eventually it was like like, Okay, I give up. Let's just go low tech. And I started finding there's like, no matter the industry, no matter what people were working with, there is a set number of steps you need to go through on a whiteboard, right? That you can quickly go through a set of requirements, get alignment from people and agree on a set of next steps and basically, specifically to dashboards. And so, over the years, I started sort of capitalizing Biden ends, like usually they're big posters, and I print that poster, stick it up on the walls, and we work through it. And then it started getting smaller and smaller and was like, actually, no, I could just do this one on one. So why don't we explore that? And so I ended up doing that. So a few years ago, I did a crowdfunding on Kickstarter for the dashboard wireframe kit, which is just like a distillation of what I do in workshop. Very simple distillation of just keeping it low tech, and because just people responded to it, and they're like it. It really works, which is the thing that's surprising, like board games. If you look on Kickstarter are the most popular thing that is funded. And so I was like, let's do a board game version of, you know, wireframing. And so that was fun. But I'll announce here, Gilbert, I know it didn't tell you. But we're doing a second edition. And we're gonna we're gonna go funded and as well, it's not available yet. But I learned so much from doing a Kickstarter about analytics. What did you learn? How do you sell a product? If you're, if your life depends on it, your livelihood depends on it. And you're building a product. How do you make make it successful? And then I was like, maybe should do that for dashboards in general. And that's why I stumbled on I want to give a plug for a book from Jeff Walker. It's called the Product Launch Formula. Anyone in analytics should read it. Because if we if we agree with the notion that any data, any data you're doing work you're doing is ultimately building the data product. You need to know how to launch it and sell it. And that's a really good book that condenses condenses it very well that's the Product Launch Formula by Jeff Walker, brilliant book. And I use that stuff all the time with my own dashboards that I'm building for clients. So but but I also have to read that to do the dashboard wireframe kit, how do you productize something and sell it and get people to buy it? So it was a great learning unfortunately, I got to apply it back into by my work as well.
Gilbert Eijkelenboom:
That's awesome to hear how all those lessons you know, unexpectedly maybe fit in, fit in your own products and fit in the story of stakeholder conversations. What I would love to do is is have a few short questions and that we make we have carousing until the end, I'm not sure if that's the right expression. But anyways, we're approaching the end. We're going to have some quick questions now and then Salma some last questions. So first question is rather easy? USA or Ireland?
Nicholas Kelly:
USA,
Gilbert Eijkelenboom:
okay, unexpected. You travel a lot what country has given you most stories?
Nicholas Kelly:
Singapore, and if you started, you started your career in data and dashboard design all over? What is the first soft skill that you would learn?
Nicholas Kelly:
facilitating workshops
Gilbert Eijkelenboom:
And what is your favorite tip for creating dashboards?
Nicholas Kelly:
wireframing you got to wireframe that's gonna save you so much trouble in the long term.
Practical tips for communicating as a Data Analyst
Gilbert Eijkelenboom:
And then lastly, you do have some practical tips for for communicating as a as an analyst.
Nicholas Kelly:
Yeah, I think a really important really important advice here is go talk to people. It is it's the biggest thing you can do. Just go have conversations and just get started doing it. Also, another tip would be don't try and solve everything in one go. Set expectations appropriately and iterate through what you need to do. So if a building dashboard, right, people want to answer 50 questions from the business. Okay, well, maybe version one, you can answer three and version two, you're gonna answer five. Right and you but you let people know. So they're aware. Version one, eight. It's not going to be everything. But if your wireframe version 10. Right, they're going, it's coming. Thank you for painting a picture for me. We'll get there. Right. So that's the value of wireframing is can make this so much easier to communicate these things. And then it's not about you. You're using different tools and methodologies to maybe take the attention off you and put it on a best practice. So if you don't feel confident doing some of these things, leverage best practice and make it clear you're leveraging best practice because very few people are gonna go Oh, no, I don't want to follow a best practice. So I can take some of the attention off you.
Gilbert Eijkelenboom:
Great, great tips. Thank you very much. And when they see the picture in the future, right where they're working towards, I think it's so much easier to get their buy in from stakeholders or end users and and keep it right and build it step by step instead of working on something for so long. And then discovering that the adoption is low.
Nicholas Kelly:
For the present, people will usually only give their proper feedback once they see something visually. Right. And so the faster we can get there, the better.
Key takeaway of the episode
Gilbert Eijkelenboom:
Yeah, absolutely. Great. So it's a fun conversation. I learned a lot and I'm sure the people that are listening to what is what is one big takeaway you want people to take away from this episode.
Nicholas Kelly:
Okay, I think Stark, you need to start somewhere. You should start where you're most comfortable and that might not be going having conversations with people. If it is great. Keep it light. So what I suggest is always focus on simplicity. When you want to start down this path and you know, Gilbert obviously it's a it's a path that you you do as well. Right is understanding people. So what I would say is, the easiest thing to do is have a change awareness. Just that if you're going to do anything, just start thinking you need to have a change awareness because what you're doing with data is but should have an impact should change the business. If we're if it doesn't, then why are we doing it? Right? So have a change awareness is my my one takeaway, I think that is going to have a large impact for people. And because when you have a change awareness, you also then have to have a product mindset. Right so I would leave it or change awareness. Great, great.
How do you keep balance? What is your focus at the moment?
Gilbert Eijkelenboom:
And then the last question, I know you work on so many things. You mentioned consulting, right you have a book this one, highly recommend it. You have courses live seminars, you have this product next next version you just told me first of all, sell this question. I'm curious, how do you combine all this? And how do you keep balance? And second, what do you what is your focus at the moment to give people some context?
Nicholas Kelly:
I thought, yeah, combining that I usually just start early. So I usually get up at 430 in the morning and then I've time where the kids you know, are still asleep. And so I can get through stuff like that, but I know where like near like the, the level I think of where I should be are where I could be the more efficient because there are things that I end up doing that are probably better. 100 people should do, right. So it's I'd say there's some struggle in in letting go. Right. So but I think there's challenges around time allocation time management. You know, I know you you advocate for, right, like a kind of ranking of what's going to have the biggest impact right and spend your time there. I probably still struggling with that. So in part I struggle with that. And what's next is I love building products. I love building physical products. So we have our Kickstarter for the second edition dashboard wireframe kit coming up I will I'll still keep building products, but I'll also shift more into sharing and distilling into training courses. So I'll be doing that but I think the fit that there's just too much excitement, and it's too enticing to not build products, because you learn so much by doing that by going through the process. And And invariably, what I keep finding is you can just apply it back to whatever it is you're doing. Right so for me learning how to sell of course is important because it's important, how we sell our consulting services, but it's also important how our clients can sell what they're doing to their internal stakeholders as well. So the better I get at that the better it is for my clients and and my business. So yeah, I think next step is just doing more at scale of what we're doing. building more products and putting more of that knowledge into courses.
Where can people connect with Nicholas Kelly
Gilbert Eijkelenboom:
Awesome. Now, where can people connect with you or follow you? I mentioned your YouTube channel.
Nicholas Kelly:
I'm probably most active on LinkedIn. So if you just look up Nicholas Kelly on LinkedIn, and our website is delivering data analytics.com So if you just want to follow anything that's going on. We pretty regularly put out free templates, free training materials to anyone on our mailing list. So and that's another place so either LinkedIn, YouTube or website delivering data analytics.com
Gilbert Eijkelenboom:
Fantastic. Well, Nick, thanks so much for for coming on today. It's I when I when we spoke last time I expected to have you know another connecting conversation with with lots of insights. After conversations. I read your book and got only more curious and excited about about today's conversation. And I'm not disappointed. So thank you very much. It was it was it was a pleasure today. Thanks for sharing all the insights and also thanks for you know, being another person helping to bridge the gap between between data and business and you're doing some fantastic work out there. So thank you for coming on today.
Nicholas Kelly:
Awesome, Gilbert. Thanks for having me. I love what you do. Thank you for doing what you do. It's really a pleasure to talk to someone who's, you know, has to not just talk the talk but walk the walk as well. And it's, you know, anytime we talk, it's like, it's like, you know, you're you have similar stories and similar experiences, because of the focus on people and you know, just see seeing your work and your, what you're able to achieve as well from your background is really a pleasure. So thank you for having me.
Gilbert Eijkelenboom:
Thank you very much and speak soon.
Comments