CCaaS executive insights | Turn CX data into growth with BI modernization

Welcome, and thank you so much for joining. My name is Marcy Rairdon. I lead TTEC Digital's global data and analytics practice. By way of introduction, I've been in customer data and analytics for my entire career, which surprisingly to me has been over thirty years. I've observed just an incredible amount of change over those decades. I can honestly say, though, that there has never been a more exciting time for data and analytics. Everyone gets it now. Everyone sees the immense importance that data plays driving business value, and of course, the technology itself has never been more intuitive. Today, I'll be sharing my thoughts and experience in the area of business intelligence, or BI, modernization, specifically as it relates to customer experience data. So this is about getting your organization to a best in class approach to the use and the socialization of data to really drive business value. Okay. So here's what we'll cover today. We'll start with the business imperative and then the CX analytics landscape. So that is a bit of a broader overview of what we're hearing from our clients in terms of what's needed in CX Analytics. And then we'll spend most of our time on BI modernization itself, what it is, the target state framework, and the activities that will get you there. And we're going to close with a maturity self assessment so you can see specifically where you stand, and hopefully that will help you prioritize your opportunities to improve. So let's start with the business imperative. According to CMP's recent research, this was a very exhaustive research study that they did, they found that improving customer data and analytics is currently the number one focus for executives. So think about that for a moment. Out of every priority that's competing for leadership attention, and not to mention AI, which I would have guessed was the number one priority, in addition to their financial pressures and employee engagement and product innovation and all of that, improving customer data and analytics rose to the very top. And this isn't analytics teams looking for this information and sharing it. This is really from the executives across all industries, recognizing that data is the foundation that everything else is built on. So when it comes to contact center data in particular, leading companies really understand that while interaction data from the contact center is crucial to things like enhancing service and creating personalized experiences. That's become pretty widely understood. But the contact center data is also the company's greatest export. So what that means is you know, context centered data can and should play an enormous role in improving other functions in the business, such as marketing, sales, product, and overall comprehensive business intelligence. So the mandate is clear. The question is no longer whether to invest in data and analytics, it's really how to modernize the way that we do it so that investment pays off. Okay, so let's talk more about this idea of taking the contact center data, or really the contact center itself, from a cost center to a value center. So I'm guessing your organization has probably already invested in a robust contact center platform that is generating massive amounts of data every second. So think of that as your raw material. Then think about everything that flows through the contact center every single day. So we're talking about the calls, the chats, the emails, the transcripts, the survey responses, the reasons that customers are reaching out and how they feel the way that they do. So that is one of the richest sources of customer truth really in the entire enterprise. And unfortunately, in most organizations, so much of it is rarely analyzed. It's captured, it's stored, but it's often thrown away. So, the opportunity is to move that data out of a siloed contact center and to turn it into a strategic intelligence hub. Make it a source of leading indicators for the entire business. So note the word leading. Your customers are telling you really important things like about a product defect, a confusing policy, something that's going on with the competition. They are providing that feedback weeks before it actually shows up in the company financials. So, if you're able to detect that signal, you're also able to unlock the business value. Let me give you an example of a client of ours, something that we recently were able to observe through the use of analyzing the interaction data. This is a healthcare payer, and the interaction analytics surfaced this big spike in calls about claims that were related to a particular change, a recent change in the policy. This was flagged literally weeks before any traditional reporting would have caught it. Once it's flagged, what happens is they're able to mobilize right away, write up all of the important information about that policy change, get that scripts out to the agents, make all of that information available so that this problem can be addressed practically in real time, like I said, where before it would have taken weeks to really understand what was causing this increase in volume. So, you know, in cases like that, the ROI is very measurable. Organizations that fully leverage interaction analytics, they typically see like a fifteen to thirty percent reduction in repeat contacts, plus all of the other important metrics like improving first call resolution, lowering handle times, and of course all of that translates very importantly to improved customer satisfaction. Okay, so if that's the vision, what could be getting in the way of that vision? There are four main themes that we're hearing over and over from our customers, and I'm going to just review those four themes and talk very briefly about the analytics solutions that address those four themes, and then you'll see that one of them is in fact BI modernization, and that's where we'll spend the rest of the time after I go through these points. So back to the themes. Okay, the first is experience inconsistency. This is when service levels fluctuate based on which agent happened to be picking up a call or some embedded misrouting that's sending issues to the wrong agent. There, what happens is there's really a scalability issue because typically, most of our customers are doing manual listening in order to address these problems, and of course manual listening cannot scale to one hundred percent, so then you wind up monitoring just a tiny fraction of the interactions, and usually you're not monitoring the right ones. So the prescription here is a solution that we have called conversation intelligence, which is a detailed analysis of conversations and also automated QA for tailored training of the agents. The second issue that we see time and time again is that AI and automation has roadblocks. So, you know, the chatbots or the self-service is underperforming. This is typically due to an AI readiness gap. What that means is the data that's feeding the AI, it lacks structure, it lacks consistency, it lacks context, and the answer here is content intelligence. So, this involves a thorough and automated review of the knowledge content that feeds your AI, along with the remedies to fix that outdated, conflicting, or missing information. The third, again, this is probably the most common issue that we hear and the reason for this webinar, is what we call reporting band aids. Really, what I mean here is there's no single source of truth, reports conflict with each other, this all results in big mistrust of the data. People really have hesitancy in looking at and trusting the reports and making business decisions off of them. So, you know, the analytics gets very siloed, and it's all kind of a result of lacking governance. So that's what BI modernization solves, and that's what we'll address in further detail in a few minutes. And then finally, think of just blind spots across the organization. So this is when there are opportunities to address customer experience that go undetected. And this happens because contact center data is the last frontier. It's typically completely siloed, it isn't integrated with the enterprise data, and the fix for that is a modern data estate. But if your organization isn't quite ready for that type of investment, which is truly an overhaul and a modernization of your entire analytics environment. Really stay tuned here because we talk about integrating and governing the data that's most relevant to your analytics priorities, and we call that the Unified Analytics and Reporting Foundation, and that's really at the heart of BI modernization. So, let's zoom into that, BI modernization specifically. It is the heart of today's conversation. As I mentioned, the solution starts with a unified reporting foundation. So, data from contact center, your CRM if you have one, and the customer profile data, so all of the demographics, etc, or firmographics about customers along with transaction history, etc. Those are the key components that need to be integrated into one consistent, reliable layer. This delivers enterprise impact that connects customer information, agent information, as well as technology performance. Static dashboards become replaced with modern self serve analytics that adapt as the business changes. And the outcome, of course, is what leadership cares most about: a single source of truth that ends the debate over which number is right, you get better CX and performance decisions from one connected view, and a clear, sustainable path to scale your analytics. So I want to be clear about what this BI modernization is not. We're not talking about ripping everything out and starting over. It's not a multiyear boil the ocean migration before you see any business value. You know, it's really about building the right foundation and the right governance so that the tools that the business has already invested in and hopefully some ones that you can add as well will deliver trusted answers at the right speed. So I'll pause here and just kind of read through, case it's not clear, what our definition of BI modernization is, which is this: BI modernization depends on the technical infrastructure and guardrails that let self-service analytics deliver on its promise. What is that promise? It's a data driven culture where business users can independently answer their own questions using trusted, governed data. And to drive this home, I'll share a recurring theme that we took back from Databrick's Data and AI Summit that happened in San Francisco about a month or so ago. And they had a quote that was kind of repeated throughout the conference, it really resonated and it's very relevant here, which is this: The real fight in enterprise AI isn't about whose agent is the smartest. It's about who controls the rails that the agent runs on. Okay, so how do we get to the target state? There's quite a bit of information here, but it's very important information, so we're going to go through kind of step by step. This is really sort of the meat and potatoes of this discussion, and hopefully things that you can take away and make actionable in business. So, to get to the target state, there are four main focus areas. First, inventory and rationalize your existing dashboards. Then you've got an intake and governance model, integrating KPIs and do the standardization there, and then finally, your dashboard taxonomy and standardization for that. So there's a big theme about standardization, and it takes a lot of effort to be standardized across all of these different pillars. So first, let's talk about inventory and rationalizing your existing dashboards. That's where you have to start. Most organizations have way more dashboards than they realize. No one's quite sure which are the important ones and which aren't. A lot of people might have their own individual way of using a particular dashboard and getting to certain information. So it's very important to start with building a full inventory. What's in that inventory? You've got to capture the owner, the audience, the KPIs, the data sources, and the usage statistics. Next, score every single one of those dashboards on a simple framework. Is it used? So, many are not. That's normal. Does it drive a decision? Is it redundant with something else? Is the data trusted? And does it align to strategy? So, going through all that, every single dashboard should land into one of four buckets: either retain, consolidate, redesign, or deprecate. And every single one that survives has to have a named business owner. If there's no logical named business owner, then that means it's a candidate for deprecation. Second, put an intake and governance model in place. So this is how you stop, you know, all of those ad hoc dashboards, too many being built, and you've got this huge inventory of reports. So every new reporting request should come through a centralized and standardized intake. Things you have to capture here: the business objective, the decision it supports, the audience, and the required KPIs. And then it has to move through a triage, meaning a business value assessment, a governance review, and prioritization by a BI steering committee. So, you know, it stops becoming who's asking the loudest or the most times, and that's what's going to be prioritized. It's this very structured framework, and hopefully with a steering committee, that puts everything through transparent roadmap. So third, focus on enterprise KPI inventory and standardization. This helps greatly with the trust problem. So if there's a centralized KPI catalog, which is a single dictionary with every single definition, the formula, the source, and the owner for every metric. And this makes it possible to eliminate duplication. Really, what you're trying to do here is align to one golden definition and certify it with governance. And then fourth, a dashboard taxonomy. So organize everything into clear categories. For example, you could have executive reporting, vendor reporting, operations reporting, technology health, so that people first know where to go for their answers, and the same view isn't replicated over and over. Okay, yes, that was a lot. But it's important, as I said. And then once that critical foundation is in place, these are the activities that bring it to life. So it starts with a semantic layer. What is that? It's a unified, governed layer of metrics with role based access so not everybody has access to everything, and certified data sets. This is what guarantees that when two or ten people are pulling the same metric, they're going to get the same exact answer. Next, usage and adoption analytics. So how many users, how many views, what's the adoption by role, which dashboards are going unused. So you're continuously rationalizing instead of just building. Then comes the AI enabled BI roadmap. You know, when people hear modernization, this is actually the step that they usually think of. And I am emphasizing here that there's so many foundational things before you get to kind of the cool AI enabled piece, but obviously that's very, very important. Think of this as three phases. Okay? So phase one is, you know, you've got smart summaries, you've got natural language queries. You know, this is where there's lots and lots of tools and capabilities out there that let users ask their question in plain English. They have trust in the answers. They get the right answers. This is really where BI is going as opposed to those static reports. Phase two adds predictive insights. So right reports, are predictions, rather than just reporting of the historical information, along with things like anomaly detection that provides alerts and tells users that something changed and is worth investigating before they have to go hunting and pecking for the answer. And Phase III is the prescriptive recommendation, which takes it even farther and gets to automated decision triggers. Finally, a newer and increasingly important component around reporting is AI observability. So, as AI becomes pervasive across the business we're all seeing that these days a whole new set of metrics is really essential. We need to monitor the model performance that's things like the accuracy, the latency, the cost, and the output quality. We need data and drift detection to know when production data no longer matches what the model was trained on. And we need governance and compliance guardrails that watch for the hallucinations and the bias and the policy violations. So this leads to operational visibility, the real time dashboards and alerts across AI applications. Really, in short here, the moment that you put AI to work on your data, you have to observe and govern the AI itself with the same rigor that's applied to everything else. So that AI observability is really a key, key component. Okay, so we're at the last section of this talk, and it's an interactive exercise. So let's make this practical, and let's turn the lens back on your own organization. So what I have here is a BI maturity assessment, and it goes across six areas. We've got data integration, reporting and dashboards, data quality and governance, user adoption and self-service, advanced analytics, and technology. So for each area, take out your pen, jot this down, score yourself from one to five. By yourself, I mean your organization. Or maybe you're part of the organization. So, a one is foundational. Data is still quite siloed across systems. Reports are primarily static. Governance is limited. Technology is mostly kind of legacy on premise. A three, of course, is in the middle. It's kind of things are developing. Some sources have been integrated, but not all. There's a mix of reports and interactive dashboards. There's governance on your key data, but maybe not all of the data, and maybe not a comprehensive approach, like I was describing earlier. And, of course, Five is a leader. You've got unified trusted data platform, real time self-service analytics, enterprise wide governance, and AI powered insights that are embedded right into workflows. So as you think about this, obviously be honest with yourself and where you really are today. Most organizations find that they're a mix, that they're strong in certain areas, they're foundational in other areas. That is totally normal. I don't know many firms that would rate themselves a five across the board, but what it does is it tells you where you should be focusing first. So, I'll give you another moment just to kind of think about this and finalize your assessment. Okay, so quickly add your scores together, and that total will place your organization on the maturity curve. Obviously, a lower total means that you have work to do, but frankly, it's not a bad place to be because that means that you have a lot of upside, and usually the fastest high impact wins are when you're foundational and you have some growth to do. So that's really not a bad thing. Of course, a higher total means that you're in a pretty good place, and you're in a position where you have opportunity to improve on an already strong foundation, and you probably also have the buy in from all levels at the organization. But, you know, the point of this exercise really isn't the number itself. It's the conversation that it starts. So, you know, think about what are the root causes that might be holding you and your organization back, and where to concentrate on your next investment. So that's what I'm hoping most of all that you'll get out of this discussion, is to be thinking about what you can bring back to your organization, where you might want to get started, and where you think there are some gaps that would really be helpful in addressing. So we have reached the end of our session. Thank you very, very much for joining. I wish you the best in your analytics journey. And have fun with it. And I really do mean this. Feel free to reach out to me directly. We have here both the TTEC Digital website, but also my personal email. And if you have any questions, ideas, feedback on what was discussed today, I'd love to hear from you. Thank you very much.