What happens between the dashboard and the decision?

(upbeat music) Welcome back everybody to the Dynamic Decisions podcast where we explore the strategies that drive successful decision making in the world of business. Why? Because the best decisions come from great conversations. I'm your host, Tisha Cable, CEO of Sea Model Data, and today I have a guest who came to decision intelligence the same way that many of us do. Not through a textbook, but through 15 years of watching smart organizations with incredible data get stuck at the same question after every review meeting. So what are we actually going to do? His name is Dishon. Welcome to the show. Thank you so much for being here. He is a data intelligence practitioner based in Sri Lanka with deep expertise across business intelligence, analytics, data science, and AI. And he recently published his first book called From Dashboards to Decisions. As a fellow decision intelligence practitioner, I welcome you to the show. Thank you so, so much for being here. I'm super excited to talk to you. Tell us more about you, more about the book, more about what you got going on. Thanks, Tisha. Thanks for inviting me. And it's a pleasure to be here. So I come from business intelligence and antics background as you said. I'm having a computer science bachelor degree and MBA. So over the years, I worked closely with organizations scaling across different markets, products, and operational complexities. But I kept noticing was something quite consistent. Companies were investing heavily in dashboards, data platforms, and now AI. But decisions were still slow, inconsistent, or something disconnected from outcomes. That gap is really what led me to write my book from dashboard's traditions. The core idea is simple. Data doesn't create value. Decisions do. And by organizations actually need this not just beta analytics, but a system that connects data, antics, AI, and human judgment into better decisions. Come on. That's what I refer to as decision intelligence. Yeah. I think in our space and decision intelligence is like, we know this. But typically, you've had to have been a practitioner from some time, like watching the problems head on and face on to recognize that the big gap is decisions that we don't think about decision making as a process the same way we think about a sales process. For example, we don't think about it as something that needs to be instituted and integrated up and down in an organization. So I kind of want to start with this line that stopped me when I was reading about your book. Organizations have never had more data. They've never had more dashboards. They've never had more analytics or AI tools. And yet, they still leave every meeting with the same question. That's the key one. What is actually happening in that gap between all of that intelligence and the decision that needs to follow? What is it? What is the thing that's happening? Yeah. So you know, because it's a challenge, sometimes I'm seeing-- I call it decision fragmentation. So and actually, I talk about this quite a bit in my book also. As organizations grow, different teams start looking at the same business through different lenses. Finastered defines revenue one way, commercial and other operations and other. Everyone has data. Everyone has dashboards. But decisions become slow and more difficult. So I've seen situations where teams spend more time debating numbers than actually acting on them. And that's a signal that the issue is not data. It's lack of shared addition context. So when it's come to scaling, suggestion, yes, increase complexity, sometimes it amplifies misalignment. You don't have a structured way of making decisions. So you identified seven gaps that consistently block organizations from turning their data into decisions. So without giving away the whole book, can you walk us through the nature of those gaps like, are they technical problems, organizational problems, or something else entire? As I said, I was thinking, so what is missing? So then I have come close to these seven gaps. I was always thinking, what's wrong? So then the first time I realized it's the disconnected intelligence. So it is like business intelligence, analytics, and AI, most of the time, exist side by side. But rarely work as a part of a single decision system. And the second one is decisions without ownership. So this is the most important factor, as I see. So the most of the time, the organizations have their decisions, but without their owners. So it's a main issue because without ownership, we can't really make the decisions. Sometimes I feel that the decisions are never happening. So we can't get any decisions without the real owners. And I think the third gap I was dashboards that informed, but do not act. We are always creating dashboards with every time. So because every team needs dashboards, every decisions. But we are not talking about decisions most of the time, but we are creating dashboards. So that's the other gap. The fourth gap is AI without decision design. So it's the most common one this day. Mainly with hype, every organization needs to have AI in it. But without a decision design, so we can't get more effectiveness with AI. So that's-- So it's good management. So after spending all of the time that you've spent in these organizations, working in BI and Analytics, right as a practitioner, what does it actually look like when a very well built dashboard fails to produce a decision? Can you kind of give us a scenario that maybe most of our listeners might recognize? Yeah. So there are some, let's say, how Dishonel making walls-- it's actually quite interesting. So in early ages, Dishons are fast and intuitive. So you are close to the customer, close to the product. But as organizations grow, they invest in dashboards and reporting, expecting better decisions. But what happens is the opposite. More data creates more confusion. So that's exactly the transition I described in my book. Organizations move from intuition to dashboards, and eventually need to move to structured Dishon systems. So at scale, it's no longer about the data says. It's about how Dishons are designed, own, and executed. I can tell you one sentence like growth at data, but scale demands Dishonel design. Yeah. So the centerpiece of your book is this model that connects business intelligence, analytics, AI, human judgment, right, into a unified decision-making architecture. We kind of call that decision intelligence. But most organizations really treat those four things as separate investments. Why is the connection between them the key and what breaks down when people decide to operate in silos? It's an interesting question. So I have a model, we call it Dishon Intelligence Reference Model. The co-idea is that I call it Dishon Loop. So it starts with clearly defining the decision because most of the time, that's not even clear. So then you define the context, which is what we are optimizing for and what are the constraints. So from there, you use data to generate insight, but the key is not to stop there. You move to action, measure outcomes, and then learn from it. For example, in Fising or Emerging Based Dishons, you rarely have perfect information. You make the best decision possible, but you design it in a way that you can quickly learn and adjust. So that loop of decision, context, data, insight, action, outcome, and learning is what creates long-term effectiveness. So you don't need perfect data. You just need that decision loop. So AI, we know, is changing the inputs into every step of that model. The BI layer, the analytics layer, and the judgment layer, all three. But from a decision intelligence perspective, is AI accelerating better decisions, or is it accelerating the same broken? A decision patterns a lot faster. What are you seeing? I think the biggest challenge is not adopting AI. It's managing AI driven decisions. Organizations are increasingly using AI for things like Fising, Focusing, and Operations. But the real question is, do you understand how those decisions are being made? If not, we risk scale-indition faster without control. So I think the next stage is really about gowners, ensuring decisions with the human AI driven. transparent, align and continuously improve. That's where I think addition intelligence even more important. AI doesn't just scale intelligence. It scales mistakes if unmanaged. Amplify both good and bad dishes. So without proper oversight, mistakes can grow just as quickly as success. - And then in your book, you also make the case for what you call intentional decision design. So that phrase does a lot of work, right? What does intentional decision design mean in a practical sense? And what does the absence of it really truly costs an organization? - As I said, when I, when my model, I describe it like we do need these seven layers. It's a design, so it's architecture. It was even a product without a proper design. You can't go with that. You will not have a long run with that. So that's the importance of a decision design. If you really care about your decision, you need to have this loop as I said earlier. So the context is very important because most of the time, if you don't know what is your context with your decision, it will not succeed very much. So it's the main thing. And then you need to gather relevant data. So you need to know what data you should see with your decision. It's the second major. And then it's come to the BI analytics and AI parts, with the data, what you should do. So what insights you should get to have this decision made. So it's very important. And then you are going to the outcome. And with the outcome, only you can learn your system. So with the decision systems, sometimes I feel like there are two types of decisions. I've seen many organizations that make these two types of decisions. They are not nominally doing these two types of decisions, but in a wrong way, most of the time. So those are repeatable decisions. And there are strategic decisions. So repeatable decision, you need to have a design decision. So otherwise, so it will come, let's say daily, and let's say sometimes monthly, sometimes weekly, sometimes yearly, sometimes time to time in your company life. But you need to know what are those repeatable decisions? And what are those ad hoc decisions? For ad hoc decisions, you don't need to have a decision design. But for the repeatable decisions, you should have and you must have. This is good. So accountability on this is one of the places where decision frameworks often like fall apart inside of organizations. So people are real comfortable with visibility and foresight. But when it comes to like who owns the decision and what happens when it doesn't work, when it goes wrong, that's where things get a little bit murky. So how does your model address accountability and is that a technical problem or a leadership problem? It's a leadership problem most of the time because accountability should define well. Because otherwise, let's say now there's a decision, do we need to go to this customer in the next year also? Now you see now a customer's journey. It's a decision. Company should decide who owns this decision. Otherwise, it is only a discussion. Most of the time in companies, we do have only discussions, decisions. Most of the time, companies think, let's say, when having a meeting, we are going to close a decision. But the issue is sometimes people think, now we are aligning to making this decision, but of alignment is often approached in wrong way. Many organizations try to fix it with more meetings. But what I have seen is that alignment comes from shared decision logic. So if different teams are optimizing for different things, such as revenue, margin, and efficiency, so without shared additional context, misalignment is unavoidable. - Yeah, you can't even approach technology building, the same way that you approach sales. They're different things. Right now, there should be collaboration, there should be endpoints, inputs, and there should be considerations, but you wouldn't approach your decision-making in the exact same way. One optimizes likely for a customer experience and whether or not they're going to do it again and the other is going to optimize for a different part of that experience. So I mean, I think there's alignment again, back to your point, necessary, but those decisions, they should have a different structure, a different design in some cases. - These seven layers should align to the decision. So in my model, it's decision is in the middle and it's surrounded by these layers, because let's say ownership is one, because it should align with the decision and also the insights parts like KBI, is what analytics we should do. It's also directly go with the decision itself. - Yeah, absolutely. Okay, here, we're going to jump into my favorite thing. I'm going to assess your decision-making persona. So I'm going to walk you through four decision-making personas and I'm going to ask you after each one, how much the way that I'm framing it resonates with how you make decisions and you'll have a scale of zero to 100%. Okay? - Okay. - Now, you only get 100% across all four, all right? - Okay, okay. - Okay, here's the very first one. It's called the feeler. It's a values driven leader. So this persona centers decisions on purpose and human impact. They want to know not just what the data says, but what it means for people. So when I'm evaluating whether a new analytics tool or AI system is actually worth implementing, the metric that matters most to me isn't the technical capability. It's whether it genuinely helps the people who have to make decisions with it. If it creates more noise than clarity, it doesn't matter how impressive the model is. I won't recommend it. How much does that values anchored human impact first framing resonate with how you evaluate and recommend technology? - Yeah, that's an interesting one. I feel if I know the other three S field, I can divide it to the-- - No, but you can't know the answer. - Yeah, no, no, it's a game. So then I'd say I'm probably around 15% feeler. - 15% feeler, okay. Let's see what's next. - Yeah, so go ahead. - Yeah, because I want to give you some context for that. So because I was very data focused, everything was numbers, dashboards, metrics, but over time, I started realizing that decisions don't live in dashboards. They play out in real life with real people. For example, you might make a perfect logical decision to optimize pricing, oh, reduce costs, but if it impacts customer trust, or internal alignment, the long-term effect can be very different from what data suggested. So now I continuously think about that human side, how additions affect relationships, teams and customers, yeah. But I wouldn't say I lead with emotion. It's more something I lay into the decision to make it more company. So yes, I like that. - I like that, maybe one company. - Yeah, but-- - But in a supporting role, you know? - Oh, shit. - All right, I like it. Okay, here's the next one. It's called the planner. It's a strategic visionary. So this persona thinks in long arcs. They're always mapping where the decision architecture means to be, not just where it is today. So when I started writing this book, I wasn't just documenting what I'd seen. I was trying to build something that would still be useful five years from now when the tools have changed again, but the decision problem hasn't. That's how I approach my work. What structure do we need to build now so that the next wave of capability doesn't break us? So, Dushan, how much does that forward mapping architecture first approach reflect how you think and plan? - So I think plan is definitely strong for me. Probably around, let's say, 25%? Let's give it a shot. Yeah. Well, it's been a man who just won't solve problem once. It's my nature. I want to understand how we can solve it consistently. I remember working on recurring operation or other types of decisions. And every time the same issue would come back, I had the same discussion again. So that's when it hit me. The problem is not the decision itself. It's the lack of structure around it. - It's the structure. - I started. So I started thinking more in terms of systems. How Dushan's are made, who owns them, what inputs are used and how outcomes are tracked. So that shift is actually what led me to write my book. Because I realized, aggression don't just need insights. They need structured ways of making decisions. For me, planning is really about designing better decisions, not just making them. - Yeah. All right, here's your next one. It's called the counter. It's the challenger of assumptions. So this persona makes progress by questioning when everyone else has accepted. it is given. They find the gap that others walk past. So everyone around me was focused on improving the dashboard. It's better visualizations, faster data pipelines, more real-time metrics, and I kept thinking, none of this is the actual problem. The problem is what happens after someone looks at the dashboard. That's the part that no one was designing for. That's the gap that I couldn't stop thinking about. How much does that assumption challenging gap finding instinct actually describe how you move through your work? Yeah, interesting. So I think count is like, let's say, now I think we will be almost 40 percent. It's like 20 percent. I naturally tend to question things a bit, not in a negative way, but to understand credo's, for example, I've seen situations where a decision looks great from one angle, like increasing revenue, but when you look deeper, it might reduce margin, or increase operational complexity, or affect customer behavior. So I often ask, what are we optically simple and what are we potentially giving up? So that counter perspective helps avoid very one-dimensional decisions. Because reality, every decision is a trade-off. It's not about right or wrong. It's about choosing what matters the most. Yeah, so I think that question in mind said becomes even more important as organizations still. For sure. I wholly agree. So here's the last one is called the data driver. It's an evidence-based operator. So this persona anchors every recommendation and evidence. They don't just trust instinct without interrogating it first. So I don't write something in a book unless I've seen it fail in an organization first. Every gap I identified, every framework element I proposed came from a real problem. I couldn't explain any other way. The practical experience is the evidence. That's what earns the recommendation. How much does that evidence-grounded experience as data approach kind of defined how you build and share your expertise? Yeah, I think that's definitely my strongest area. So I'd say now I have 40%. I'm glad that. So you know my background is in BNLTIC. So naturally I rely on data to understand what's happening and reduce uncertainty. Especially in complex environment, the data gives you a solid starting point. Without data, you just another person with an opinion. Edward's jamming said that. So I thought I'd be leaving that and I don't like to be that person. But one thing I learned and something I talked about in my book is that data alone doesn't make the decisions. So I've seen an organization with great dashboards, great data, but still struggling to make effective decisions because data tells you what is happening, but not necessarily what you should do. So what I've moved from being purely data driven to being more decision-driven. You've seen data as a foundation, but combining it with context structure and judgment. So I would say data is where the decision starts, but it's not where it ends. This is great. Well, one, thank you so much for being here. I just appreciate you sharing your expertise with us. Would you let our listeners know briefly where they can find more about you, where they can find the book? Yeah, and thanks for having me here. It's a great time we had. So I think the easiest way is through LinkedIn right now. Okay. I'm quite active here. There are a share of thoughts around vision, intelligence, data and AI in practice. And if anyone is interested to coin deeper into these ideas, my book from dashboard solution is available to Amazon. I really want to really bring together everything we discussed today. How organizations can connect data, analytics and human judgment was important into better decisions. I love it. I love it. And human judgment. Yes. This is great. Well, thank you so much. And to our listeners, I want to thank you for joining us on the Dynamic Decisions podcast. If this episode resonated with you, we would love for you to subscribe it, rate it, share it with a data leader or a decision maker in your network who might need to hear this. And let's keep the conversation going on social media using Dynamic Decisions podcast. And until next time, keep questioning, keep growing and keep leading with purpose.