Day 2 Analytics Panel

in the chat so that I know that I am being heard. And I think Julian— Julian, you need to come back on the stage with me.

Yes.

Okay, just drop a note in the chat to make sure that my little reset did work. Great, thank you, Keron. And Julian, I'm going to hand over to you to pass us on to the next panel.

If you're hearing me?

Yes, I'm hearing you.

Yeah, but for some reason or other, you need to turn my camera on?

Click on those three dots underneath your picture, three vertical dots, it should allow your camera.

Don't understand why that is happening, because—

It should say audio and video settings.

My camera is there, should be there.

Ah, let's go, let's go forward with the, with the audio.

Yeah, fine.

Okay, I'll bring up the other, I'll bring up the panel.

Okay, so, um, let's, let's, let's then go to dealing with the, what we call the demand for data-driven insights for the new strategic planning. Everybody's very aware these days of the importance of data and managing data very well because it certainly is a very important guide to exactly what our strategies should be or could be, and also to, to, um, to help us review and conduct the kind of analysis that makes sense with every move that we make. So let me welcome Karen Ram Ganesh, who's going to moderate this session. And Francis will bring you in, I'm sure. And you can take it from there.

I'm going through the motions, Julian. Tell us some more about how analytics is disrupting your company as we wait for Carol.

I think, yeah, one of the— well, I suspect everybody is using analytics in one form or the other these days as a means of measuring the impact of our work, uh, certainly within the media industry, particularly with the online activity of, uh, of, of all the media houses that we're tracking. The, the availability of analytics is crucial because it is part of that selling tool that we make available to advertisers to say, look, these are the eyes and ears and so on that we do have. And therefore, when Aldo was talking earlier about the clicks and views situation, I said he was— he certainly is preaching to the converted in terms of the heightened interest that I have in, or we have in, in that particular area as media operators. So it's crucial. Gwen Garrison is there, I see.— Miss Harriet Mann as well. Good. All right. So we just need one— or Urit Atkinson is with us as well. Urit, how are you doing?

I'm fine.

I got that right, Urit?

Karon, I'm sending—

Yes, you did.

Sorry, guys.

Karon, I'm sending you the invitation. It should pop up on your screen right now.

Yeah.

And—

So you'll hear a great deal about Google Analytics these days because it certainly is one of the biggest measurement tools we have available to us, and it's become like a, like a Bible for everybody who wants to convince people, advertisers particularly, to support our particular work. Karen is there, looks like.

Karen, are you hearing me?

Yeah, we're hearing you. You can put on your camera now.

Yeah, trying to sort out my video again. Okay. Hi guys, how you all doing?

Oh, we're great.

Good morning, doing well.

Morning, morning.

All right, so good morning everyone, uh, to the audience as well as, uh, our panelists. So thanks to, um, to Julian and Francis for a very, uh, interesting conversation earlier. I dropped in a bit late, but, uh, but it really ended well. All right, so Alright, so good morning again everyone, and of course our industry panel is really here to share, you know, how analytics contributes towards, you know, strategic planning. And I have some really interesting guests with us today. I have, let's start from the top, well, who I'm seeing at the top, Dr. Gwen Garrison, full professor and director, Education Evaluation and Data Analytics, from the US. I have Ray, I'll say Ray Hiramwan, chapter lead, a woman in data, you know, Tableau Certified Associate Consultant from Trinidad and Tobago and doing all kinds of really nice stuff across the region. Welcome, Ray. I also have Mr. Atkinson, Hewitt Atkinson, Interactive Media Manager at Digicel, Jamaica. So welcome, sir.

Thank you.

I know that we have tons to discuss. I don't think we have enough time to talk about everything that we could possibly discuss. So let's, uh, let's kind of hit it off, uh, in general, uh, just to get some of your thoughts in terms of just a broad sense of, you know, somebody who may not even know or just maybe heard analytics in passing, just a general thought, uh, about analytics. If you have to tell someone who's totally new to this or never heard it before. So Gwen, I could probably just start with you, just a minute or two of your time.

Uh, sure. So analytics is, uh, is now a huge area, but it, it really is trying to harness the data that's inside and being captured by the organization or by the company itself and try to make sense of it in such a way that it can help support decision-making. So there's a lot of information that flows, and, and probably, um, if you think about it, one of the most and longest-serving data curators in an organization has been accountants. But now data is beyond just the financial components of that, and it's expanding throughout the organization, especially as they're harnessing things on the web.

Thanks a lot, Gwen. Uh, Rui, you want to tap into that?

Oh yes, so, um, I totally agree with Gwen. Um, analytics is all about harnessing the power of your data. I always say that, um, you know, I ask rather if people really see and understand what their data is telling them. And especially, you know, I would say within the last like 5 years or so, there was a statistic that came out in 2017 that said 90% of the world's data was created in the last 2 years. And if you think about that and you think about the amount of data that was created over the past 2 years with the pandemic, I mean, it's exponential. And, you know, we sit on a data gold mine. I tell people that we live in a data universe, and people, organizations, entities that really do not use data simply won't survive today. So it really is about being able to harness that, the power of your data, the potential of your data, being able to get those insights. But not only get the insights, but turn those insights into actionable insights. So, you know, many times there's a gap between what we get and what we actually do with it. And, you know, as you said, Karen, we could go on and on and on. There's so much, and that has— and analytics has transformed and has gotten so far. So, you know, the conversation could go into using AI, doing, um, automated, you know, automation, um, predictive, prescriptive analytics and stuff, as opposed to, you know, that traditional analytics that we were accustomed to now to this modern space of analytics. So yeah, that's pretty much it.

Yeah, yeah, that's— and that in itself is a lot, and we could take that and talk an entire day about it. Uh, you want to chime in?

I mean, I echo the same sentiments between Gwen and also Ray. Um, I think Ray had, um, she used a phrase just now that I wanted to say initially, which is Data— we use data to harness, to pretty much— let me start over. I'm a digital marketing professional and data is what I do on a daily basis, all right? And how I use data is pretty much to put into actionable insights, as Ray would have mentioned. And what tends to happen for me is there's an inundation of data coming in. And so what you have to literally do is mine that information and create decisions from that. So analytics is, is pretty much that. It's how we use the information that comes in to make decisions on, on the backend.

And, uh, and you know, just to add to that, because there's so much— you could see the, the sync among everything that we're saying, and data really is very important, but of course there's so many things that revolve around data in terms of, as Reid had mentioned, you know, the interpretation of it. As Gwen knows very well, you know, gathering it and being able to get the right data and the relevant data to use towards making those decisions. So analytics, uh, there's so much work around it and there's so much mystery, I think, as well around it. And I think that's, that's part of what, what we would want to demystify in our discussion today. To really get everyone to understand, listen, this is not some kind of rocket science type of thing that, you know, we can't do as a small business, or we can't do in this industry. And as Gwen rightly said, you know, it started in accounting and now it's everywhere. I do a lot of work in people analytics, and you know, it's been hitting off in the region as well. Um, you know, and HR is one of those places that's so soft, you know, uh, you'll never think that, you know, there's a ton of data in there that could really help us to help people. And so analytics really should be a staple in many organizations, many industries. And what we'll seek to do today is really kind of get people enthused about it and interested enough to see the value of integrating that into what they do on an everyday basis. And what I like to say, appreciate data. Appreciate data enough that we could use it for something and make it useful rather than just storing it somewhere and we don't know what's going on with it. Correct. All right, so, so yeah, so I'd like to invite you guys. I know that, um, each of you would have, you know, kind of prepared a little 5-minute presentation, uh, you know, on, on your, your views and perspectives on, on where analytics is and where it's going in various industries. So I guess, um, instead of me calling on anyone, maybe someone could volunteer to go first. And I'm not looking at you ladies for this one, you know, it could be anyone.

First, if I could figure out how to share my screen.

Um, if you look at the bottom of your screen, uh, right next to your video and mic icon, right in the middle, you'll see present to audience. Yeah, it's a square with an arrow.

Yeah, it seems as— do I have to share my— oh, I want to share window.

Sorry. Yeah, how do I do this window?

Oh no, I have to share my entire screen or PDF, so I'll go ahead and share my entire screen. I do apologize.

Sorry, not a problem.

Yep, let me go into PowerPoint. Can you let me know when you're actually seeing my screen?

Ah, we're seeing it.

Yes, we're seeing it.

Okay, great. So I'll start. So, you know, this morning's topic is, um, meeting the demand for data-driven insights and strategic planning. What I really mean, to break it down, um, is really how does analytics contribute towards strategic planning? And as I mentioned a couple of minutes ago, the statistic in 2017 said that 90% of the world's data was created in the last 2 years. And it speaks to the number of transactions that are produced daily. 2.5 quintillion bytes of data is created per day, and that was back in 2017. So if we think about how much data is being created now post-pandemic, or I would say, you know, being on the other side of the pandemic over the last 2 years, it's really exponential when you really come to think about it. Back in 2017 as well, in May 2017, in an article posted by The Economist, they said that data is the new oil. Um, I've disagreed with that since it came out in 2017. I've always said the data is the new air. It is more valuable than, than oil. It is literally a lifeline. So for me, I mean, we have tons and tons of data, but people really don't understand the value of their data. So, you know, the question that I posed was, are you really harnessing your data to its full potential? And Gwen actually echoed that before me. And in order to harness your data to its full potential, you really need to understand, to be able to see and to understand what your data is telling you. And, you know, in order to do that, and when we, when we try to align our analytics strategy, so organization's strategic objectives. Sometimes in trying to do so, we, we encounter many challenges, and I would like to focus maybe on one big gap that I found in my experience over the years as a data practitioner that I've seen, and that is really the adoption. Because many times in trying to align our analytics and our strategic goals, there's a huge disconnect. And in my experience, that huge disconnect to drive that data decision, data-driven analytics culture throughout the organization is getting people to see, to adopt analytics and see analytics for true value, for what analytics can actually do. And many times I've seen from the executive level that the executive team doesn't really have a clear vision for what its analytics, um, strategy or process is supposed to be. So that, that creates enough in and of itself a huge, um, a huge issue. Many times as well, especially with, with driving that data-driven culture and where analytics have now transitioned from the traditional approach where we use a lot of static reporting, um, you know, Excel, PDFs, etc., to interactive real-time, um, platforms that allow us to get information at our fingertips. Many people see that as a threat. And when we start the conversations on AI and machine learning and stuff, people are threatened because they see stuff like automation as technologies that are going to replace them. And that isn't actually the truth. Um, and as we continue with today's discussion, you'd see, or you'd realize, that all these new technologies, these emerging technologies, are really there to help you and not to harm you or replace you. A huge key in overcoming all of that and being able to drive that adoption throughout the organization, from what I've seen again as well, is education, you know, a lack of solid understanding between that traditional approach to analytics and the new modern approach, seeing the benefits or understanding the benefits of how predictive and prescriptive analytics, AI, automation, um, ML, etc., can actually help us and make us more efficient at our jobs and free up our times to do more strategic things. And I think once we demonstrate that and we're able to show value then, and we're able to educate our users at all levels throughout the organization, you find that the adoption process is a lot easier. A huge component as well is really trust, and this trust comes from trusting the data. One of the biggest gaps I've seen in that whole analytics strategy has been poor data quality. And in this real world, we never have perfect data. We'll always have data that needs some amount of cleaning. We need to use ETL tools, etc., but a huge factor in driving adoption is getting users to trust the data. And one of the ways that we— or the main way of fixing this is through a proper data governance policy. When there's transparency, when there's visibility, accountability for the data, people know what data makes up this predictive model, where the data came from, having that single source of truth, having that accountability as well, having you know, valid— validation for the data, making sure that the data is accurate. It's— it, it comes from one source, they can trust it. And once you start developing that trust for the users, then you find at all levels throughout the organization, they start trusting these dashboards, these reports that they're going to be getting, and they start using it more often because it's correct. Because the last thing that we want— and I've seen it many times, this happened to me as well in my previous role— um, two departments, you know, reports on analytics or reports on data, or they present reports. They are reporting from the same accounting application, but the accountant comes up with one figure and the IT manager comes up with a different figure. That was my previous life, all the time having multiple sources of truth. So that trust and the data governance is a huge component. And then technology. So, you know, we look at— we need to focus on the people and the process and that kind of thing, but people tend to always put the technology first and then focus on people and process. And I would say focus on the people, having the right people, getting the data in the hands of the right people, the people who have the questions, the people who know the business, the people who need to get insights right away. Those are the people who will drive your analytics and will get you those actionable insights that you need. While you put the data in the hands of the right people and you, you develop the correct processes, the right processes, be it your data strategy your strategic planning process, and you align all those together, then you employ the appropriate technology. And tech is, you know, many times I've seen where technology is really a disruptor. People don't adopt analytics because the first thing they say is, "Ray, that's too hard to use. I can't do any programming. I need to be tech savvy." But technology should be an enabler. So I think once you start with the people, you develop the processes, and then you find the technology that aligns with the people and the processes, technology now becomes your enabler, not your disruptor. So really, to summarize it, successful adoption I've seen is through the people, process, then the technology. And in between there, we have educating people, building that trust and stuff. I think that comes in between the layer of the process and the technology. So that pretty much is it for me. I will stop sharing my screen now. Let me go back in. Thank you, and I'll hand it over to the next speaker.

Thanks a lot, Ray. You're welcome. You are here. I can go next.

All right, hold on a sec.

Gwen, he's given you time to drink that coffee.

Join it. Gwen is 2 hours behind us in Jamaica. It's now really 9 o'clock.

Oh yeah, she's in. Yeah, hold on.

So it's really early for Gwen.

Are you seeing my screen?

Not as yet.

Perhaps Murphy's Law is affecting me. You go ahead because I'm just seeing a buffer from my side. So you go.

I am trying to answer a chat, so I will do that. And then, okay, so I lost, uh, I lost the ability to share. There we go.

Okay.

So, um, I'm gonna share the screen. There we go. Okay. Hey, good morning everybody. Glad to be with you, and it's early on the West Coast, but I, I'm a lark, and so mornings are much easier than very late presentations. I'm glad to be with you today. Um, so as a way to just, uh, connect me to you, let me just say in the opening that Dale and I are longtime friends and colleagues for almost a couple of decades. And, and Dale's, uh, life took her to Jamaica. My life took me out of Washington, D.C. to, uh, to Los Angeles, where I am a professor leading a new program to try to build the capacity for all this analytic work and teaching about it and getting my students prepared to do the statistical sides, the data gathering, the data management side, how to actually do those things really well. And on the side, I have my consulting business, which I act as a principal data strategist to help organizations kind of think about the work that they're doing in data strategy., and trying to harness the data that makes sense in their area. Uh, one of the things I just want to say is it's a blessing to know Dale, and it is wonderful to be connected to Frances, and wonderful to be connected to all of you. Uh, in my profile is my LinkedIn connection if you want to connect with me there. I'm happy to further the conversation support after this. So let me— I just have one slide, and it's really when we think about the strategic endeavor, it really comes down to asking the right asking the right questions at the right time. And often what happens is there's a lot of questions that come up, but we ask them and we don't actually dive in at the executive level to really start to lean in to where we want things to go, or if we have capacity, or what kind of capacity it's going to take to be there. So there's usually, when I'm working as a data strategist, with companies and nonprofits and educational institutions, I start with kind of building on 3 questions. And I'm just gonna highlight those as a way to prompt your thinking and maybe to get you going around it. So the first question is, what is it that you can't do now that you'd really like to do in the future? And if you need some ideas about how to think about that is, You know, what would you like to know about your customers? What would you like to know about your supply chain that you don't now know that you would like to know for, say, future impact or growth in the future? So if you start there, and usually at the executive level, what happens is there's a lot of conversations, there's many, many questions, and you at the executive level have to get disciplined to focus on the one or two questions that you're going to really lean into. Otherwise, what you have is a lot of staff chasing a lot of information and providing you reports that end up sitting on the shelf. And I really want to iterate what Ray said. Analytics is only really valuable to a company if you can make it actionable for yourself. So you've really got to think about how is this connected to your future aspirations for your company or for the mission, to accentuate the mission. The second question is, what data would you need to build your confidence— and making a good decision. And I want to connect with Ray on this. This is about trust and having that confidence in those dashboards that they are telling you the kind of information that you can take as actionable insights. You can take the available and obvious information to start. And that's like I said earlier, accountants were our first data scientists in our organizations. We didn't call them that, but that's what they are now because they were harnessing the financial information. But the deeper and discoverable information flows come from a variety of sources. Let me just mention two. One is surveys, customer satisfaction surveys, which you're getting those, those are direct kind of information where you're asking your customers or your vendors about that. And then there's indirect And an example of the indirect is tracking website behaviors. Where are people concentrated on your products? How are they getting things into their carts? If you're in— if, if you have a presence of that way, how long does those things stay in their carts before they actually make a purchase? And that's a big thing Amazon has figured out, is just if, if they aren't prompted to close out those carts. People just have lots of things hanging out in their carts for a long time, and that doesn't actually close a deal. The third thing is what resources do you need to make it real, to make this question real? And there are a variety of things that you can do. One is programming. You might need the sophisticated technology. But a lot of times, a good spreadsheet where you have the capacity to actually roll up that information and provide it in pivot tables can be very helpful to understand two basic points in analytics. One point is the trend. What are you seeing as the common pattern? And the other is what's hanging out on the tails, and we call it in the statistics world, What's hanging out there all by itself? And is that hanging out there all by itself something we want to capitalize? Is it something that's emerging for us? And the final thing I want to say is, is human capacity building, and that's where I have taken my 30 years of being in, in industry and nonprofit worlds and building it back in the university where I'm trying to build the analytics teams of the future. You can invest in the people you have right now, and there are terrific tools out there. DataCamp, for instance, will help accelerate an on-the-ground team and really start to build their capacity. Or you can hire for the skill set that you need or for the strategic vision implementation that you need. So both of those things, there's many ways to tackle this, but it starts with asking a good question at the right time that you want to actually capitalize on for your future.

Well, that's great. Thanks a lot, Gwen. That's very, very insightful, and we could definitely see a lot of alignment with what Rui was talking about earlier, especially with regard to, you know, trust in the data and being able to, you know, start that process, you know, gathering data from the beginning. What are you capturing? Is it relevant? Is it required? And then Trust in the values that come in from the operational processes that will of course lend towards your decisions, uh, on a strategic level. Thanks a lot, Gwen. We'll definitely dig deeper into that just now once, once you are deals with Murphy here.

Um, Gwen mentioned website just now and it just piqued my interest. I got so excited. Anything web related, I'm all for. Um, Karen, I also sent you my presentation via email. I'm still buffering from my side. I'm not sure what is going on.

Um, if you—

are you able to bring that up?

Ah, okay, cool.

So I'm just seeing it here because from my side I'm on a very good internet. I'm not sure what is happening from my side.

Okay, no problems. Well, in the meantime, while I'm getting this up, um And you guys kind of share, like, what are some of the factors that you've experienced, uh, you know, dealing with teams, uh, on various levels that either, you know, influence them or hold them back from actually applying analytics towards some of the, uh, strategic items that they may have to deal with? What have you all experienced so far?

I think there's a fear of numbers. I think there's a fear of getting all that information and not understanding what to do with it. And what tends to happen is people tend to be a creature of habit, and they, they have a particular way of doing certain activities, and they sometimes, especially senior managers, want to remain as such. So there is a definite, um, intimidation of data, and, and rightly so. The data— there's a lot of information coming in, and what to do with it is always a question that both upper, medium, and lower staff levels actually have to grapple with. So that's just my take.

Karen lost this connection, it seems. Oh, but if I were to add to what you're saying, I actually think that, um, I agree with you. And I think another— well, one of the barriers to adoption, like I mentioned, adoption is one of the biggest hurdles. One of the biggest barriers, to add to your point, is that people don't understand what the numbers are telling them. And many times as well, this came as a huge surprise to me. Gwen actually articulated it in her introduction, and she said that, you know, the first users of data or the first carriers were accountant, and I actually had an accountant tell me just recently, she said, Ray, I always thought that data was numbers, but data is everything. Data is more than just numbers. Data is a fact. People don't realize, like, in terms of analysis, doing demographic analysis, doing cohort analysis— where are my customers coming from, who are my top customers, what are the cohorts, where are they purchasing from, you know, that kind of thing— even that type of data, people are still amazed by it. And I'm sure you could, um, resonate being in the digital marketing space there's so much data, especially in the marketing world. Because I mean, financial— my background comes from a very strong financial, um, a lot of financial experience, even though finance is not my forte. I've always been in computer science. But it's, you know, seeing in the marketing space how much data we really have to harness and how little of that people are actually harnessing is quite amazing.

All right, Gwen, I came from a history background. Um, it's just a fact that existed, and I'm not sure how or why I am here from a data perspective, but I'm glad. One of my mentors spoke earlier, um, and he made mention of several things from the marketing perspective, and I'm going to echo one of the things you also mentioned, Ray, from the segmentation. No, Where my customers are coming from, what are you to, to target? These are some of the things that happen, um, that I have to pretty much consider on a regular basis. And the good thing about it, I have the platforms that allow me to pretty much segment and target accordingly. But even with those platforms, just getting that data, getting that information, and passing it on to the, the key decision makers within my organization are things that I mean, they get excited about it because it's not just gut feeling. No, they can literally see the information, whether from a dashboard perspective or otherwise, and saying, okay, um, Segment A seems to be working, but Segment B, they're high-value customers. Perhaps we need to focus on them. So I don't want to Keron, are you hearing me?

Yes, I'm just breaking up a little bit, um, but yeah, I'm hearing you. You see that I'm sharing your, um, your stuff here, right?

Yeah.

Oh, finally! Oh, I can get Murphy off.

If it does, right? Okay, cool. So you want— if you want to kick off, um, I'll just control it on this side and you could just tell me when to go next.

That's fine, Karen, you can.

All right, perfect. Oops, click.

I had no idea it was on automation. Next slide.

Next.

Nice. All right, so, so I basically dissected, um, how we do a company standpoint or overall strategy. And this, this pyramid is, is probably adaptable across all companies. So just to look at the definition, strategic planning is a management technique that helps organizations set future goals and objectives. I'm going to stop there. From the marketing side, every single aspect of that pyramid has some element of data interwoven, from the mission to the objectives to the strategy, the approach, and even from my side, the tactics that we employ from the digital side. We have to be in a position where we, um, connect what the company wants to do with, with, with actionable insights with data that can allow us to get to that particular point.

Next slide. All right, let me just get—

oops.

Yeah, so I think I just made some problems with this.

All right, you can stop, you can stop there. And what I'm going to do is just, um, go through rather quickly. So from the marketing side, we are now in a world that is definitely data-driven. Everyone that I interact with, they want some amount of personalization. Every campaign that I launch, we— and how do I move forward?

Interconnected.

I think we lost you, Watts.

Yeah, you was breaking up heavily there. Oh shucks, we lost him there.

We lost him. Yeah.

Oh, all right. Well, I think, I think, um, Yeah, I think something's up with his bandwidth, but I like where he's going in terms of the relationship between strategic planning and of course how does marketing as well as any other type of process— because let me just kind of tap back on something that I heard you all talking about. And Ray, I heard you speak about governance, and that in itself ties back to that whole trust mechanism. And I think it's something that's happening in the region heavily, yeah, uh, where people don't trust using data. And I saw, um, Jocelyn in the chat actually mentioned that where managers don't even consider it useful, uh, to actually collect data for processes. To them, it seems like a time-consuming process, something that's taking them away from, you know, doing the job that they have to And I think we have to shift that mindset into, uh, something more useful for the holistic organization rather than being an extra piece of work that they have to do, you know. Uh, one of the things that, uh, let me just stop presenting, uh, until Ewart catches himself back there. Um, one, one of the things that I, that I have seen, uh, on many occasions is that that whole mindset that, you know, data collecting, this is just a whole bunch of extra work on top of us, and we don't need to do that. We have enough experience where we can make these decisions moving forward. We've— like, like Gwen was talking about, you know, seeing the trend— we know the trend because we have been here for the last 10 years, 15 years, and, and we're fine with that. We can make the decisions moving forward. Um, one thing that I have done in, uh, for some companies is collect data for one process or two processes and then show them that whole analytics, uh, process into how, uh, these trends can lead to something useful for them. And that way we get their buy-in for, for pushing on it from an operational level and then furthermore, uh, on a strategic level. So I think, I think we have to do this in phases because As, as Yvonne Rae and Gwen mentioned, where people are involved, we have to get people to appreciate data first before they even get to the point where they're data-driven and analytical. Even if you have small teams that are analytical that feed into strategic planning, they're not going to get the support from the people because that's where the data comes from, right? We have the operation, we have the process that generates data, but it's the people that will, will accommodate that process happening. What do you all think about that?

So let me just add that I think some— one of the things that's a hurdle that most people don't appreciate in the same way is what is something that Ray has alluded to and what makes up data. And in the data world, we bucket data in two ways. We call it structured data and unstructured data. And the best machines, computers, operate best on structured data. So whether you use, you can count the number of sales, you can put a date on something, which is a data structure, we can do that, or you can count it. You can use a, a Likert scale in order to measure satisfaction. All of those kinds of things are structured data, and computers are built to handle and process structured data. It becomes much more difficult to handle the unstructured data. And often what we have in computer science and computer engineering is people who are taking those unstructured data pools and creating and harnessing them around data. So when, when we say the organization is awash in data and we're trying to answer a question, sometimes where the mismatch happens is we use a past structured data pool that actually is not close enough to actually answer the question with some level of confidence. Confidence and satisfaction. And so that's why when you, you start with these questions and you do exactly what the pyramid that you were just talking about, you really look about what is it going to take to actually make this real and to make the truth of this trustworthy. And it takes a lot of planning in order to make it come through so that it really can influence the decisions. And that process really needs. It needs the people in the mid-level management and the technical specialists working closely with the executive group.

Yeah, I agree with that. Actually, um, I was going to say as well that, um, you know, in order for success in any analytics project, and especially when it comes to aligning your analytics to your strategic planning. There are really 3 key personas that I see, and many— I say personas because many times in organizations I've found that one person wears multiple hats. But you have the business user, as I mentioned earlier, the one who understands the business, who knows the business rules, who understands the caveats of the business and needs information, wants to get insight in a timely manner. Then you have um, the what traditionally used to be the MIS person, but I now call that the subject matter expert on your analytics tool, because that's cool. As we've seen, has now transitioned from the IT department to now having, you know, business intelligence departments, um, data support units, etc. So you now have the person who understands your analytics tools, or the subject matter expert with the analytics tool and that domain area that you're speaking about, because it could be marketing, finance, HR, etc., as we know. And then the other persona is the one who understands the data, the database administrator, the database technician, whoever that is, that knows that this data sits in this table. These are— this is, this is where we'd find these components. In order to do a cohort analysis, we need to connect to these tables. These are the fields we need to connect to. And you find that when these three personas come together is where you have that magic happening, like I say, and you create those moments when people start to see their data and get excited about it. Because with that is where you ensure— where you really achieve the success. You have— you now have the person who has all the questions, the data person is able to find that data for them to answer the questions, and then the subject matter expert with the tool is now able to take that— those questions and put it into a visualization for them that makes them, you know, I get that, right? Hands off those light bulb moments. So those personas are really integral to achieving success in any analytic project.

Yes, that's very true. They become the translator, basically, uh, take any business needs and use any tools and technologies to actually provide some kind of insight, you know, to help drive the decisions that these guys have to make. And, and Derek actually in the chat brought up something very interesting where he was mentioning Users don't really understand everything that they need to get. And then we have a lot of these vendors and providers who provide these tools and solutions, you know, the off-the-shelf, and, you know, they'll customize, but it doesn't always give the user everything that they need. As you were talking about, you know, getting the right data, the relevant data, sometimes an external vendor don't really understand all the intricacies of a process and what they really need from it. And they just kind of provide what they can provide, you know, because the users who are involved in that process will always know best. Even from a strategic perspective, they might just look at it from an operational side. This is what we do, this is what we collect, but nobody really looks at it holistically. So I think a recommendation there would be like having the proper teams in place when you're setting up analytics tools or even some of those solutions that even collect any data because we need to get, you know, get it right the first time, get the inputs well. Because if we follow any, um, SIPOC model, uh, the inputs must, you know, match what, what you're looking for as your outputs. And I think that's, that's a lot of the problems we get, because people get fancy solutions and they can't get any insight from it. They have a lot of data, but even if they try to analyze it, they don't get enough insight. And, and Gwen, like what you mentioned, you know, the the right question at the right time. It's, it's tying back to, you know, my stuff in knowledge management, where we is to get the right information to the right people at the right time to make the right decisions. And, and it's all, it's all connected in terms of that holistic process, right? So, so yeah, I know, I know that we all speak in the same language. And, uh, when we think of organizations and how they really want to use data are they really collecting the right data in the first place? And are they using the right tools? How do we get them to really appreciate what they need to do to get to the point of strategic planning? Because when you think of strategic planning holistically, it's about making the right decisions every day, right? As somebody I mentioned earlier, strategy is all about that pattern of decision-making. If you're making the right patterns of decisions, you could get there.

Yeah, go ahead. Kieran, I think it is a lot about taking incremental steps. Steps and getting quick wins, quick and sufficient wins to create that trust cycle. And so what we have learned in very sophisticated companies, we have multiple operational systems that are guiding the practice and we're harnessing very sophisticated teams are joining one side and the other and, and putting these together. In smaller companies, they don't have those sophisticated systems. Systems operating behind them. So they may be working with bank statements, they may be working with various different places where they're, they're just tracking information, inventory information, simply working from invoices. And so what we have to think about as we build these teams is how do we get data in a position that it can serve us? And sometimes that will, will take a new skill set, or sometimes that will take building on a particular skill set. So I think some of the obstacles here is actually, is here's our question and here's our variety of data buckets in variety of structures. And then we have to figure out what are we going to do to get it across that gap. Before it's actually making use for us. And sophisticated organizations have already begun to think about that and build that capacity. Other organizations are still coming up, they're still trying to figure out this process. And so the back end of this is, is it's not— you can't do it in a snap of a finger. It has to be deliberate, intentional, and an investment.

Yes, and, and I, I also will answer that is that when, when we think strategy, everything analytics, I think many companies think that it has to be organizational-wide, you know. They think that, okay, if we're gonna get into this analytics thing, we're gonna have to do it every way for it to work. But I think as you talk about the incremental approach, you have to do it in certain segments. I think, I think the organization needs to decide where are our critical operational areas and then start tackling those areas first, generate the value. And once they start seeing the value, then the snowball effect occurs because they will now be more willing to trust it and make the investments necessary to get there.

Yeah, and I, I don't want to diminish the, the need for understanding how change dynamics work because sometimes we, we want the answer and we don't want the the answer, because that may actually reflect on us as a manager in a particular way. And we have to, we have to build that change and capacity for change in our organizations as we go, because data can be very threatening. It can reveal things about ourselves that maybe are, you know, we kind of go, gee, I do, I wish I wasn't It wasn't quite like that. I didn't know I was the lowest salesperson on the totem pole there. We have to think about the emotional relational dimension that these answers are going to have an impact in what's going to happen with them as they roll out. I think that Rae has a really good command on that perspective.

I think to echo what Gwen was saying with the incremental steps, I've seen many times, especially in large organizations, quite interesting, where they say, well, you know, we can't start our analytics process as yet, or we can't start using analytics as yet because we have to consolidate all our data sources, or we need all the companies within the group to be using the same platform, etc. But what I always say is start where you are with what you have. So, to your point as well, Keron, you know, you start in one area. You may have data maybe just for HR alone, and start using that HR data, start harnessing the power of that data, developing, you know, those, you know, picking those low-hanging fruits from it, getting the value. And then once you can demonstrate value within one business area, one unit, one department, then you start to gradually and incrementally grow. And as you grow, you would realize the need for— and then you start, you know, looking at your analytics roadmap. Do you have a proper strategy? How do you collect the data? What are the parameters that govern how you collect data? What are your business rules? Because I've always seen in that whole analytics roadmap, um, where we— where I've walked clients through, you know, that, that process, the biggest hindrance has always been with the data quality. And many times within the organizations themselves, they do things because— I think you said it, um, Carol— this is the way they were told to do it, or this is the way we do it at the operational level. But making that connection, understanding the business rules Many times people don't understand business rules, and when you ask questions, being the subject matter expert, and you ask the right questions, then you're like, oh, so the reason why we do this is we only calculate processing time from 8 to 4 on a Monday to Friday. We don't include weekends, we don't include public holidays. So when you break it down to a level where they can understand, then, you know, you start growing. And as you grow, those incremental steps, you start developing your policies, you start putting in that analytics roadmap for your analytics strategy. You start defining a governance policy, etc., is where you get there. So many people think that I have to have it all ready up front before I could start using data and start using analytics. No, you don't. You just start where you are right now with what you have.

Yeah, yeah. Uh, you what, you what, are you hearing us? Uh, maybe you could put on your audio and see if you're getting through with that.

Um, I mean, I'm hearing you guys loud and clear, but I'm a little worried that if I go on, Murphy will happen again. But I have been listening. Um, I'm unable to provide a, a point right now simply because I literally just got on, so I'm not even sure where you guys are.

Not a problem, buddy. You just stick around with us and we know you'll add value at some point in the discussion. Thank you. Certainly. Um, when I like that, um, answer, I was going to talk about that whole noise versus signal question that, uh, that was raised. And I like, I like your perspective on it. Um, Ray, that, that question that, that asks, you know, how do you separate the signal from the noise? I'll just add something before you answer it, uh, because when you spoke about— yes, because we do need to pay attention to all data that comes in, and you know, those outliers usually give us some indication of things that, that may happen, you know, that, that, that will help with forecasting, you know, where trends are changing. Where demand is suddenly shifting, etc. Um, I'll also add something as well, you know, how do we separate the signal from the noise and how do we do it well? I think it has more to do with the capability of your analytics team, you know, getting the right people there to look at your data, not just any riffraff, you know, not just a boot camp and telling people, hey, this is how analytics works and, you know, you'll go run free, you know, go forth and analyze You know, we can't do that. We need to ensure that we get the right skills in place so that they can actually detect and separate the noise from the signal and then be able to maximize, you know, the varied data that come in. Rui, what are your thoughts?

So I would echo what both you and Gwen said, because with Gwen, both are important, really. You need the noise, but you need to pay attention to the signal as well. Um, and in what you're saying, having the right people— many times you find that people use analytics, and I go back to what I like. I, I actually just love the traditional versus modern approach, and people do stuff and they do it methodically. So, you know, we do reports, we do whatever, but then we really don't take that and turn it into actionable insights. And it really takes initiative, creativity, conscious thinking to really take what you found and be able to use that to make an informed decision. And sometimes you find the gap because we, we are creatures of habit. I think somebody mentioned earlier, and we just do things mechanically. We don't go that next step. So being— so to your point, Karen, having the right people, the people who will think further, because many times executives tell— tells me, well, you know, really use the analytics, but they don't really bring any insights. They make no suggestions. So you produce this fancy dashboard, it's all well and nice, but what are the steps? What are the contingencies? How can— how can looking at my sales on a scatter plot, my sales versus profits on a scatter plot, really help me to increase profits and, you know, double revenue, etc.? They don't go that extra way. So it really is the right people, like you said, Keron, the people who would take it further for you.

Correct. And, and You know, strategy, as I said, is a pattern of decision-making, and we see that connection now because to make the most informed decisions or the best decisions possible, you need that data, right? And to get that right, uh, coming to you, of course, the analytics process must be appropriate. Uh, we talk about tools, uh, a lot of people talk about, uh, even in the chat, this dimension, okay, we need these fancy tools, we need an MIS, etc. But I'll tell you flat, I've helped a lot of companies, whether it be, you know, coming in and looking at what they're doing, even in HR, in IT, etc. Simple tools just like Excel. You know, we have Excel, we have that Excel system sitting right on our desktops, and we don't realize we could use Power Query, we could use pivot tables, and we could crunch some of the numbers that you get it from even the fancy software applications and, and be able to generate sufficient insight. You know, we could build small dashboards or even just a couple charts in Excel, and with the right storytelling, you know, with the right context, we can provide that insight that will help drive decisions. So I think, uh, I think that was mentioned earlier, that, you know, people come first. And I always, being a technology person myself, I'll tell you flat, I put technology last because technology really is a tool that you harness. You know, technology amplifies what you want to do. So you don't put that first. You have to put your people with the right skill sets. You have your process generating the appropriate data, and then you use the tools now that could crunch that into what you need. And, you know, so you don't really need big fancy tools, you know. And many of the organizations, many of the executives, when they really— when it comes down to it and they get a dashboard, the usefulness of that dashboard doesn't depend on whether or not you do it on Power BI or Tableau or Excel. It really depends on whether or not the charts and visualizations that you obtain can actually give you enough insights to help drive your decision-making from a strategic perspective. So that's really what all they care about.

Yeah.

And I wanna, I wanna add to what you're saying, Kieran, on that. Um, the human brain, uh, needs to be trained to read a table. Um, and we do a lot of that work in our primary and secondary kind of data literacy work, that's what we call it now, but that when we were coming through our math classes, the human brain is really set up to look at visualizations. And we can make very quick judgments. We can get into the data when something is moved to a visualization. And Excel is very powerful in making visualizations for us. And then our data visualization tools like Tableau and Power BI, really helped to harness multiple complex data flows coming off of lots of systems in order to actually visualize data. So what I want to say is that you can bring an organization along when you take that additional step to try to make a visualized data story out of it, because it is that data story that kind of hooks us in our brains and in our hearts to try to make better decisions. And it gets us going from just being an analytical kind of activity that we do in our brain to say, oh no, this, this is where we need to go. This is the segment of people that we want to reach with this product, service, tool that we have. And I just, I think that's where there's so many powerful stories, because when we see that happening, then we can build on our success.

Yeah, just, just a quick point though, um, Excel is something that I use, but in, in my area of expertise, um, I think I alluded to it earlier, I don't necessarily have to use Excel or Tableau. I use a platform called Klipfolio, and I mean, having worked with all the platforms, um, they have their, their, their analytics component built in. What I find is that even when you provide the data to— whether it's, it's a senior or somebody else— there's not enough data interpreters. So yes, I can visualize it, I can put it in such a good format, but once they receive it, it's like, okay, this looks cool, what next? So even as data scientists, we have to, we have to literally delve it or extract it so granular that, okay, okay, this is what I need to do next.

Yes, yes, I totally agree with that.

Like, um, you know, sorry, let me just, let me just ask the audience, uh, what tools are you all currently using? Do you all have any specialized tools, or are you all kind of, you know, exporting your stuff out and working in Excel? You know, uh, what are you all using and, and how are you all applying analytical processes, you know, within, uh, your various industries. So if you could share that as well, uh, when you all share it, we could probably add some tips in terms of how you all could probably improve, uh, how, how you all could probably, you know, shift how you're doing things, and we can provide you some with some direct suggestions. So yeah, Ray, you could, you could jump in.

Yeah, so now I was going to say that, um, data visualization is a huge component in, you know, successful analytics, um, being able to to see the data, to see what that data is telling you. And regardless of the tool, being able to use charts, graphs, visuals that are easily understood. So, you know, people look at a bar graph and they can tell what's happening. You look at a line graph and you see like it's a time series analysis. You can see your data over time. You can see patterns, trends, where you had dips, where you had, you know, what times of the year you had the most sales, et cetera. And then being able to, Ewart's point, to articulate that. So the data storytelling component is a huge component as well, because you can have the best dashboard, but interpretation is quite, um, interpretation varies from person to person. So what success looks like for me may be totally different from what success looks like for Gwen, and we can look at the same dashboard and have different interpretations of it. So as the person presenting that data or sharing that data with with the audience, um, it is really important that you can tell that story and not just tell the story, but, um, to my point earlier, being able to tell that story, um, in a manner that is effective, in a manner where you are able to take the insights and make them actionable. So this is what it's showing us, and based on this, if we were to continue 'If we were to increase revenue by 8% every month, then we can meet our annual target of 21%, etc., etc.' So being able to see the data, visualize it, understand the data, and then articulate that to your wider audience helps you to get the buy-in from the executives. It's really very, very important.

And one thing to add here, and I think Ewart might be able to add to this one with his marketing book is that while we may make the best decision possible on, on what we're seeing from the data, we still have to review the outcomes. And that brings us back to the data because it is, it is a cyclic process. Uh, not every decision we make may pan out the way we expect it to. So part of the whole analytics process is actually looking at the results, comparing the results with what we expected, looking at those gaps, and then making further decisions along the way. Decision making doesn't stop, right? We make decisions with every breath that we take. What do you think?

Sorry to cut you off, Kieran, but I got so excited by, by, you know, the trajectory of the question. No, what is happening in my field is who I report to, they now have a KPI called return on ad spend, similar to return on investment. So in terms of the output that you just mentioned, every single dollar that I I have to be in a position to say what amount of revenue was gained from that $1. So I now have a KPI that, say for instance, 5 to 1. So in terms of first crafting the segments that I'm going to target, second ensuring that the message resonates, and then how it resonates comes from the return on ad spend. Because more, if more persons are buying, if more persons are taking that action on e-commerce, it means that my data from start to finish has panned out. Those are some of the things I literally have to look on. But the outputs that you mentioned, it's absolutely on point. We have to be in a position where we know that whatever we would have created from the start, we know that the outcome pretty much works simply because that's how you would have crafted that data in the initial stages.

I'd like to build on that about how it is so important at the executive level to manage expectations around the data storytelling. I'd like to share a story that impacted all of us. When I say all of us in the United States, I'm thinking that you guys in the Caribbean were watching just as closely. But at the beginning of the pandemic, when the United States and pretty much the world goes under lockdown, Dr. Fauci comes on with this data visualization that says, look, for 2 weeks, if we all do this, perhaps 4, we're gonna flatten the curve and we'll all be back to work. And that is absolutely what did not happen. And of course, things exploded. I mean, we became restless, and 2 years later, we're kind of all kind of emerging, and life is fundamentally changed for all of us. And my point about telling this is when you use a visualization and it's used in that forecasting mode, when you say to somebody, this is what's going to happen, and you don't manage that expectation around it, you can totally then set up a a backward step into nobody trusts you.

Yeah.

Because in fact, you have set up yourself for this unrealistic picture. Now I can tell you, in January of 2019, my statistics colleague and I, on the back of the napkin, were calculating the spread information coming out from the World Health Organization in January. We had it pegged. About the worldwide situation in 30 days, what we were going to have just on rates of spreads, and we were doing that on the back of the napkin, and we all sat there and go, this is going to shut down the globe, and no, none of our leaders know what to do with it. And in fact, then you come out with Dr. Fauci, and then what happens is Dr. Fauci's credibility is completely eroded as a result. So this factor that, that Rain has brought up and that You are to talk about, about building trust is so important from the executive ranks down and from the people coming up who are providing the data. Managing the change process in the analytics world is a vital component to this piece.

It is for me, that especially, especially in driving that change process and in maintaining the trust because once that trust starts to drop, you find that you have a lackluster approach to analytics, to everything. And it's like, okay, the interest drops, people don't really use it anymore. They begin to question the credibility of the person telling the story as well, you know. So it's a whole chain reaction, a whole ripple effect if that is not managed properly.

Yeah, I'll also add another story, Gwen. Well, I'll A lot of positive story this time. Uh, I, I had recently joined an organization that they actually had a training school for new recruits and providing them with all the resources and that kind of thing, but turnover was pretty high and the organization was not doing anything about it. I came in as a, as a senior HR officer and I did the calculations, and I think it's one thing that we need, especially like in, in people analytics as well as with other departments, the ability to quantify the process and quantify, you know, the value being generated. I think that's, that's what the executives really need because they speak that language. You have to speak their language to be able to get through to them. And I did an analysis of it over, let's say, a 6-month period when I was there. And, um, we had presentations to senior managers, and HR had a presentation to do, a presentation, and I went to do that. And my first slide was We have effectively, uh, lost $2 million, uh, in the last 6 months. And all the guys who were falling asleep in the afternoon of that all-day meeting actually woke up and jumped out of their seats and were asking all kinds of weird questions. And then of course I did the breakdown in terms of, you know, the excessive cost of recruitment. We had a bucket with a hole, water is leaking out, the pipe is filling the bucket, but hey, it's gonna never get full. Because we're wasting water, we're wasting time and effort, resources, etc. And everything comes at a cost. And that actually broke through to the executive who sat down within the next week, and all the recommendations I've been praying and pleading and begging for the last 6 months were actually implemented. So being able to quantify using analytics, you know, the value generated within a process or the losses incurred within a process very, very important to help with planning and of course strategic planning, because you need funding, you need, uh, you need that investment internally, both with people. Yes, you know, getting buy-in, getting the data appreciation, all well and good, but we also still need funding to get these things done. And to get it from the executive level and to get that buy-in, I think is very, very important. So, so adding the dollar value towards, you know, the data being generated and these things, it's, it's essential and, and of course critical to getting going, uh, for the organization.

Karen, just a quick point on that. I think you spoke their language, um, and it happens to me too. Once you mention money, once you mention where it hits, it piques everybody's interest. And I've done that before where I've provided some amount of data, even if it's visualized in the, in, in the, in the best way 'But once I start to speak dollars and cents, everybody's ears perk up and they want to know more in terms of how to stop that hemorrhaging.' So that's, that's, that's how you do it.

Yes. Yeah, so get any support not only from those guys in the process, you know, um, of course that, that in itself is an uphill battle, right? Because they consider it more work. Or what, what I've done is try to integrate it in certain steps of the process. So just, you know, as you're doing this, just log it. As you're doing that, just log it. Or we find some way to automatically capture it, just like Gwen with learning analytics. We capture the intrinsic behavior of your learners. You're not going to ask your learners how long you stayed, but we could capture that through the system, and then we could correlate that with, you know, their performance and figure out, you know, if there's a possible causation effect or if it's actually just correlated, you know, and it may not necessarily be you spend a lot of time here, that causes you to perform well, you know. But if we find simple ways to do it, and I think the real story and the real takeaway for most of the guys in the audience is that analytics is not like a 0 to 100% kind of journey. It's an iterative, gradual thing. When you think about change, change is erosive. So you want to slowly erode it, you know, you do segments, you do portions, you look at the high potential, processes, the, the high-impacting processes within your organization, and you start hitting it there. You look at your tools, you look at your people, you find the right people with the right skill sets, you use the tools that you have at your disposal. As Ray said, you know, do what you have with what you have. And then you start deciding, okay, as you start generating value, we start, um, quantifying it, which will give us now the input to us to say, okay, let's make the right in investment. It's not an expense, guys, it's an investment. Because as with all projects, we have the data to show it. As well as analytics people, we all have the data to prove you get 10 to 13 times your investment value back. And, and that, that has data behind it. So, you know, yeah, yes, sir.

Yeah, if I may add to what you're saying, Karon, um, I totally agree with what you're saying, but if I may add to it as well, I think in developing or designing that process especially now in what we like to call the new normal. That process definitely has to be iterative, but it has to be a lot more agile than we used to, um, than we were accustomed to, especially since, you know, over the last 2 years, everything that we've seen. And not only agile but adaptable, um, you know, from— I think I speak for everyone— the pandemic disrupted our lives. It disrupted every area of, you know, digital transformation, if you will. It literally brought or drove that digital— um, was a catalyst to really move that digital transformation conversation along. So people were forced to— like I said, if you don't, if you don't use data, if you don't adapt, you're not going to survive. So, you know, your, your, your process going forward has to be adaptable, it has to be agile, it has to say, you know, is it disruptive or is it You know, you, you need to factor in all these things as you go forward now. So I just wanted to add that to the point that you were making.

Yeah, great. Gwen, I know that you always have something valuable to add, you know, so I'm looking at you.

Um, you know, start with your question, start and build incrementally and develop the capacity in your team. Your team has a lot of good insights They, even your frontline folks, have a lot of good things to tell you. So, and it is an investment, but it's really worthwhile and it can be very satisfying. Manage the change process.

Yes, guys, it's about that time, and I'd like to thank you so much for deep and practical conversation that's gone beyond, I think, anything I've heard online here in the Caribbean with respect to this particular topic. So, um, to our audience, please join me in giving kudos to our team, our panel. And guys, for our audience, guys, you should know, if you're listening in, these folks aren't— they didn't just get together 5, an hour ago and start talking. They've been preparing for about a week and a half to put this panel together, and you want to know that there's a whole lot of work that goes on in the background to get something of this quality. So please join me in thanking them and thanking all the panelists actually for their ongoing investment.

And, and Francis, if I could just merely request, we To finish this conversation, we'll just need another 12 hours of your time and beer.

Well, thanks, Keron, Ray, Ewart, and Gwen. We're going to end this session and then— well, actually, I don't think we need to end it. I'm just going to have Julian introduce the next group, the next