Tell impactful stories with data

Transcript

Hello, and welcome to today’s webinar, Adobe Analytics, Telling Impactful Stories with Data. Today we’re going to be talking about the importance of data storytelling in your organization and how you can use Adobe Analytics to do it. My name is Alyssa Magruder and I’ll be your host. I’m a senior adoption and retention marketing specialist for Adobe Analytics. Today you’ll be hearing from one of our amazing Adobe Analytics champions, Amy Ard. Before I jump in, I want to address a bit of housekeeping. This webinar is being recorded and after the webinar we’ll be sending out a link to the on-demand recording, which will be hosted on the Adobe Analytics community. We’ll have a discussion post on the community forum where you’ll be able to find the link to the recording, the slides from today’s session, as well as any follow-up questions from the Q&A. There are a few things on your screen I’d like to point out before we get started. First thing, the counsel that you’re seeing on your screen is completely customizable. So you can resize or minimize any of the windows you’re seeing. Feel free to make the slides or the video larger or smaller, really depending on your preference. So feel free to test that out now. Most of the widgets that are included are self-explanatory, but to quickly walk through some of the specifics, we’ve gone ahead and shared several resources related to telling stories with data and continuing your learning journey with Adobe Analytics through Experience League. You can find those resources on the right side of your screen. Throughout the session, if you have any questions for our presenter, simply type your question in the Ask the Presenters box on the bottom left of your screen. We’ll do the best to get to all of the questions, but we have a few folks on the line today. So if we don’t, as I mentioned earlier, we’ll start a thread on the Adobe Analytics community to help answer any questions and really continue the conversation. And last but not least, there are several buttons that are located along the bottom of your webinar console. Outside of the widgets just mentioned, there are a few other things. You’ll find additional information about our speaker, a survey, which you’ll also see on the right side of your console. Please be sure to take the survey before you leave. It should only take a few minutes, and that’s how we pick our topics as well as presenters for future sessions. You can even give our speaker some love throughout the session with the Reactions button. So if you hover over that smiley face button, you can give things like a thumbs up or a heart. So feel free to try that out now. Now onto our agenda and what we will be covering today. Today we’ll be discussing how to tell impactful stories with data. Specifically, our presenter is going to review why is data storytelling important? What are some key components that make up a successful data storytelling toolbox? Her three-part narrative for crafting your story, as well as several additional examples of how to use Adobe Analytics to do this. As mentioned earlier, we have time set aside for Q&A, but feel free to ask your questions throughout the session. With that, I will pass it over to Amy to introduce herself. Thank you so much, Alyssa, for kicking things off, and thank you, everyone, for joining today. First, to introduce myself, my name is Amy Ard. I’m the Director of Analytics at Levelwing, and I’ve been with the Full Service Digital Agency for over 10 years now. And what you’re seeing here is just a brief glimpse into my life through a few images. I am based out of Charleston, South Carolina. And fun fact, I met my husband in college in Calculus 3 class. And we now have two children, a five-year-old and a two-year-old. They definitely keep us on our toes and make sure we’re consistently lacking a little bit of sleep. Living on the coast in our free time, we try to make it to the beach or on the boat as much as possible. But to give a little bit of background on myself, my undergraduate degree is in mathematics from the College of Charleston. And after graduating, I moved to Dallas and began my career in finance. And starting out, I was an investment advisor managing client assets via a stock options investment strategy. And then I spent several years in foreign exchange trading. Then I made the decision to move back to Charleston after several years in Dallas. And I began my journey at Levelwing. And before that, I had not had much exposure at all to the marketing world and prior to making that shift in my career. But there’s always been this common thread, and that’s using data to develop insights that prompt action. And to tell you a little bit more about our company, the genesis of Levelwing two decades ago was really in analytics and data mining. And this formed the basis of our fundamentals-first approach. And while we have evolved into a multidisciplinary technology, this approach remains critical in everything that we do. And to lift up the hood a little bit more and share more about Levelwing and what makes us tick, we’re endeavoring to build a company, to build a team, to build an analytics program that’s made up of individuals in which curious minds, big thinking, and caring people can create value. And that really is our North Star. And that’s creating value through great data storytelling. It’s creating value through great insights. And it’s creating value by simply being an extension of our clients’ teams. And we’re able to do this by having a team that’s extremely curious. And that curiosity sparks proactivity. Proactivity leads to boldness. And that boldness inspires creativity. And this is all fueled by the passion embodied in our team. And myself, as a leader in the company and our analytics program, I work to create an environment where the inquisitive analytical mind can thrive. And I do this by consistently trying to bring wellness into the forefront and give my team the ability to be the amazing storytellers that they are and bring that North Star to life every day throughout, again, their curiosity, driving some big thinking, and all that backed up by passion and also an authentic sense of self and what we’re working towards. And all of this allows us to produce elevated work as evident in the clients that we work with that you see here. So jumping into our message today, why does data storytelling matter? So here are a few stats that suggest why. The most popular TED Talks found that stories made up 65% of content. Stories have an attention span of about eight seconds, which is less than a goldfish. And visuals are processed 60,000 times faster than text. And messages delivered as stories are up to 22 times more memorable than facts. And really, I’m sure most of you won’t remember a single stat that I just shared. So let’s try that again. This is a more visual way of showing these same stats. My slide is showing blank, so I hope you all can see the slide here. But it’s the same stats shown in a more visually appealing way. And so I ask you the question, do you want your audience to engage with your message, understand it quickly, remember what you’re presenting, pay attention to your findings, and to you? So the key components that best serve as tools within your data storytelling toolbox are data, visuals, and your narrative. So data, it’s the foundation of your story, and it includes the metrics that you need to support your point, your visuals that help your audience to absorb the data quickly and efficiently, and your narratives that give context to help explain the data and why the audience should care about it. And how all of these components really come together is you. That’s the human component. That’s what a machine can’t do, at least not yet, not very well. And because of that, that’s your job security. Data storytelling is where you increase your value and the impact that you’re bringing to your organization. And so that brings us to the three parts to tell an impactful story. It’s exactly what you see here. It’s identifying the problem, explaining through data and insights, and it’s offering a solution. And that’s it. That’s the main message that I want you to take away today. And sorry, I’ll flip back, but really it’s these three parts that you see here. And the intent here is to keep it simple. It’s to talk about this in a way that has wide application, regardless of the industry that you work in, regardless of the category, regardless of your audience. And while it’s helpful to know your audience and know what’s at stake, this is a winning guide regardless of the various factors at play. So to take a quick moment and recap, the problem or opportunity that we’re framing here today is that data storytelling, it can feel overwhelming. It can feel or it can be difficult to learn. It can be difficult to teach for that matter. And that’s because it’s this blend of art and science. But I imagine that’s why some of you are here today. But in the remaining time that we have together, I want to build on these three parts and I will share a few Adobe analytics tips along the way. And I will walk through a few real life examples. So let’s get to it. All right. So digging into part one, identifying your problem. This is one of the most important parts of your process. It’s because this is where you have the opportunity to captivate your audience from the very beginning. And really it’s the entire motivation of your analysis. So if framed correctly, it can help you communicate the value realization of your entire analysis to the wider audience. And it’s important to note that not all problems are suited for analytics. So with that, you want to ensure that the problem is fitting, that it’s generally a suitable problem to solve through analytics. It’s important to not rush to spend the time to fully understand the problem, think and discuss before you start and socialize it. Framing the problem should be a social effort. Get your team involved, get other stakeholders involved and really together align on a path forward and that will help to further establish trust in the data and yourself as an analyst. And make sure that it’s important to your organization. Also that it’s truly a problem that’s been unanswered previously. You don’t want to go down that path if it’s already been solved and that you have the data available, that the right data is available to you or you can easily obtain it. And that leads us to the chicken or the egg scenario. So what comes first? Is it the data or the problem? It’s important to know that this is a circular relationship between the data that you need versus the problem that you’re framing. And again, it’s important not to rush the process. Analytics is an iterative process and it’s natural to kind of cycle back and forth between these steps. And if you’re still stuck at this point, always start with a question. So for example, looking at the last piece here, seasonality, do you see significant data fluctuations around back to school or say in the spring when tax returns are hitting? Begin your process by digging into topics that you know to be relevant to your organization. All right, so jumping into a few Adobe Analytics tips before moving into part two. Make sure you have a deep understanding of your Adobe Analytics solution design reference. This is your EVARs, your events and how they map together their scope, their classifications. This could be an entire standalone webinar in and of itself, but it’s really important to mention here as your implementation is really the entire foundation of your data landscape. And I’ll take a moment to call out a fellow Adobe Analytics champion, Jeff Bloomer. We were on a coffee chat just in the past month or so and some of us began talking about the importance of implementation and how some of us started our careers or exposure to Adobe Analytics on the implementation side. And it’s really helped us propel our paths forward as an analyst and gain additional success there. So really important to understand your foundation and what is fueling your web analytics platform. Also, title your visuals in the form of a question. This will show exactly the problem that you’re trying to solve or the current situation that you’re highlighting here. That eliminates the work for your audience having to do that on their own and the data that you’re showing. All right, so part two, explaining through data. After you’ve framed the situation, use data to reveal the source of the problem or opportunity. Provide compelling visuals that show the relationship of your variables, but also keep it simple. Make it easy to consume, keep it short and don’t over explain, and show confidence in your results. Also, focus on the results that best explain the situation as opposed to the full analysis and process of getting there. And to highlight that a bit further, settling down, it’s helpful to mention here exploratory versus explanatory data and the process for both. So exploratory is your insight development process and explanatory is your process for data storytelling. And so Adobe Analytics is absolutely going to be your key driver and all that powers the exploratory side. But it can also be the process for your explanatory data and where you’re actually presenting your data story. Really, that’s up to you. And so now for a few additional Adobe Analytics tips. General organization. I recommend including a workspace summary for each new project that you have, including things like an overview or purpose. Include your intended audience, your creation date, for example. This will all be helpful for your future self when looking back on past workspace reports. Also, collapse and move to the bottom of your workspace. Any in-depth exploratory visualizations as we just touched on and focus on the results and not the full path that it took you to get there. That’s particularly important. Again, if you are telling your data story within Adobe Workspace, you want to just move everything that’s not super relevant to the bottom. It can be there for being in your back pocket. If you are presenting to a more analytical audience and the question comes up, you can quickly drop down and expand that section. And lastly, create simple, easy-to-read visuals that align with the type of data that you’re sharing. Use bar charts for categorical or binary data. Use line charts for numerical data and scatter plots to demonstrate relationships. All right. Now we’ve hit part three that’s offering a solution. This is extremely important because it allows you to keep your audience engaged by saving your recommendation for the finale. So during part one, when you’re framing your problem, you can hint at the solution, but be sure to limit your details so that your audience does not fixate in earlier parts on the next course of action, but really rather the analysis that you’re providing. And communicate your recommended action and make sure it’s clear on how you get there. The most effective way to do this is to quantify the possible impact through at least one critical API. And also provide details so that the stakeholders can make an informed decision. For example, any cost implications or required resources that it would take to implement the solution that you’re recommending. All right. More Adobe Analytics tips. So leverage classifications, leverage calculated metrics, leverage all the tools that you have available to you within Adobe Analytics directly to help quantify that possible impact. So think net profit or customer lifetime value, for example. And again, do this directly within the platform, within Adobe Analytics. Also use summary numbers and changes within Adobe Workspace to really end on a high note. And the example you see here kind of brings together all of the concepts that we’ve talked about up to this point. I know it’s a little difficult to see in the small text, but in that summary note at the top, we have detailed out the three parts of our story. We have the problem and opportunity. We have a summary of our data and insights. And then we also offer a solution. And this, again, screenshot here that you can’t see very well, we’re talking about product recommendations and looking at the average order value for best sellers and featured products and value products. And we’re looking at that trended over time and how that’s really resonating with consumers. All right. So now I will jump into a few full examples that walk through these three parts. So in this first example, you can see that this is focused on product search enhancements. And framing the problem here, you can see on the left, we’re really trying to dig into how we can reduce funnel fallout after a product search is performed. And on the right, you can see how we’re explaining through data and insights. So the biggest takeaway on the data side, again, in this example, is that nearly a third of our product searches for a particular category that we’re analyzing here, it was returning with no results. And we saw a lot of drop off and exits because of that. And continuing on with the same example, here we have part three where we’re offering a solution and we’re quantifying through missed revenue, essentially. So the solution that we were bringing to the table here is that the store lookup radius for nearby store locations appeared to that needed to be expanded to reduce the volume of no result searches. And to quantify that solution, we took the difference between the conversion rate for product searches with results against product searches without results. And we multiplied that by the average order value. And that equated to $2.24 million in missed revenue. In this example, even taking it a step further and showing our initial postmortem analysis after the solution was put into place. So you can see here in the highlighted portion on the right in this chart, the conversion rates noticeably increased. And that was within a week’s time after the solution was put into place. And you can also see the volume of product searches with no results that drastically decreased after the solution was put in place. So they had the inverse relationship as we expected. And what that resulted in was the highest weekly conversion rate over the last five months. All right. Moving into this next example, example two. So we’re exploring purchase behaviors of different audience segments. So you can see here that I have the full three-part story just shown condensed into one slide. The opportunity being that the average order value increases significantly following a customer’s second purchase and so on. And part two, explaining through data, we discovered that loyal customers, those customers that are making three or more purchases, they had an average order value of $500 greater than returned customers. And so the solution that we have here is to further analyze those loyal customers, the intervals between their purchases, and identify the ideal time to put retargeting efforts so that we can convert returned customers into loyal customers. And so to quantify this even further, if we were to say convert 5% of returned customers into loyal customers, this would equate in $1 million of incremental revenue. So again, quantifying the solution that we’re offering up here. All right. And moving into this next example, we explore the opportunity of capitalizing on the growing segment of electric vehicle owners. And that’s through the analysis of their attributes and behaviors online. So here we’re asking the question, how do conversion rates for electric vehicle owners compare to typical performance? And we took two of our top KPIs, in this case, the appointment conversion rate and tire conversion rate, and looked at that again for electric vehicle owners and again for non-electric vehicle owners. And further exploring, again, exploring data and insights, here we’re exploring further the year-make models for electric vehicle owners that’s represented in on-site traffic and to identify opportunities for audience targeting and potential tests that we may want to run in the future from the media side. And here again, we’re exploring more with data. We’re looking into which markets we should prioritize for increased electric vehicle audience targeting. All right. So wrapping that up, now we’ve identified the opportunity. We have explored a little bit from a data perspective. So now we have at least a better understanding of electric vehicle owners and their needs. We can help guide them with the right content and create a better user experience and ultimately increase conversions and revenue. So the three solutions you see here, we want to focus and expand SEO efforts and grow our organic market share. We also want to implement regional media tests to inform future growth opportunities and make sure that that also aligns with operational efforts in those markets. And the other solution, we want to leverage Adobe Target to serve up personalized website services for EV-specific messaging, guides, and imagery. All right. So wrapping up here, I’ve identified a problem and opportunity. We’ve explored it through data together. I’ve offered you a few solutions along the way. I wanted to close out on this note by sharing my philosophy towards analytics, which is quite simple, and that is to change behavior. So the purpose and value of everything that we do as analysts is to prompt action, and that results in changed behavior. So data, insight, and analysis really lose a good deal of their value without action. And I mentioned passion in my intro, but another valuable trait to show as an analyst is really empathy, because through analytics, you are working to persuade people to change behaviors, to take action, and change can be hard to accept. It can be hard to adopt. So it’s really helpful to have empathy too as an analyst. But what I want to leave you with here is that we should strive in every case to hold data accountable and hold it to its highest potential, and that’s through activation. All right. And at this point, I’m going to kick it back over to Alyssa to take us into the Q&A session.

Thank you, Amy. That was great. And the tips and examples that you shared were very insightful. We have a few questions that have come in from our audience that we will take now. And if you haven’t had a chance to submit a question for our presenter but you would like to, you can do so by using the Ask a Presenter panel that’s located on the left side of your screen. As a quick reminder, if you have to leave early, please don’t forget to take our short survey. It’s only three questions. And as mentioned, it helps us select topics for our future sessions, as well as speakers. With that, we will jump into our first question. So Amy, question from the audience. At my company, presenters tend to use the same presentation for various audiences, different meetings, and then just skip over things that are not relevant with the audience. Would you recommend the presentation of data be different based on the audience? And I’m also curious, are there any ways that you would recommend using Analysis Workspace to do this? Yeah, great question. So to touch on even the last part, for working directly within Adobe Workspace, I would say to save that for your more analytical audience. At least that’s how I proceed in those cases. That way, if you are going to drill down into the data a bit more, again, you can have that information readily available. You can save that towards the bottom of your workspace, as I touched on, and just have it collapsed and readily available if the question does come up. So in those cases, I highly recommend to leverage Adobe Workspace right within the platform, and just follow some of those best practices in terms of setting up your workspace in a way that you are prepared to walk through and tell that data story within the platform itself.

Going back to the first part of your question, I think that it’s really important to cater your story and your presentation and your visuals to your audience whenever you can. So I know there’s going to be some cases when you may not have a lot of insight into your audience. At least I experience that sometimes. Maybe if you’re on the brand side, not on the agency side like myself, maybe you already have a good sense of the audience that you’re presenting to. But in any case, yes, highly recommend getting to know your audience beforehand, knowing how they receive information and consume information, and really cater to that. So I would say that’s probably not really flexible in a templated version. I mean, there could be a framework in place, obviously, and as we talked about today, that highlight the three parts for data storytelling. But I would highly recommend catering that as much as you can to the audience that you’re presenting to. That’s great. Thank you for that. The next question that we have is from Corey. Do you have any recommendations for the best way to tell stories for the success of platform products? So, for example, outcomes that don’t necessarily directly impact conversion metrics or impact more subjective outcomes.

Sorry, you cut out at the very end. Apologies. I was trying to find the text to see if I could read the question myself. No problem. So recommendations for the best way to tell stories for the success of platform products, for example, outcomes that don’t necessarily directly impact conversion metrics or impact more subjective outcomes. Yeah, so I mean, I would say that maybe there’s opportunity to do a little bit more digging and see if you can really dig into the KPIs that do impact or the data that does impact one of your most critical KPIs. That could be, you know, an opportunity where you’re continuing your analytics process and going through that cycle of investigating through, again, data and framing your problem and going through that cycle. But I would strive in every case to make sure that you are going after, again, the impact of one of your most critical KPIs.

Great. And our next question comes in from Michael, who’s asking, do you have any tips on changing companies’ behavior from always wanting quick ad hoc data reports to wanting more sophisticated self-service dashboards? Yeah, that’s a great question. I would say it’s helpful to really educate on the process of analysis and what it takes to get there, particularly if you’re asked to turn around something very quickly. It may just take a moment of educating and helping others to understand your process and the level of effort. There may be some quicker wins, too, if you are working under a time crunch to deliver information along the way. That’s not without, you know, avoiding the fact of providing something that’s, you know, unfinished, but that may be an option as well. And it’s really just creating that analytics culture, as mentioned, you know, in the question of creating these self-service dashboards and creating a centralized data hub and a way to make sure that you’re not in a situation where you’re not getting the best results. So, you know, it’s really important to have a robust data hub and centralized resources so that everyone is working off of the same playing field and there’s that single source of truth.

Great. That is very helpful and hopefully helps answer a question from the audience. So, you know, we’ve talked about this a little bit, but we’re going to talk about this across different business units. What tools of Adobe Analytics help facilitate that? And are there things kind of similar to what you were just discussing that are self-service reporting or, you know, what tools would you recommend for this? Yeah. So, to service different business units, I highly recommend making sure that you’re using the same type of software across all business units. And I would recommend utilizing that segment as like a filter within your workspace so that you can look at things from a holistic standpoint and look at performance comprehensively across all business units. But then also there’s the opportunity to drill down into, you know, one specific business unit or, I think that that’s the best way to handle that type of information. But yes, make sure that you have all of that data classified directly within Adobe Analytics itself.

Great. That is a great tip for our audience members to take away. Kind of relating back to one of the earlier slides you discussed with which comes first, the problem or the data, do you have any techniques that you use or that you can recommend to those on the call to help recommend generating a hypothesis that really sparks a test? Yeah, absolutely. So, I highly recommend, again, making it a social exercise. So, when my team maybe has, you know, troubles identifying that starting point, we talk closely with our client services team and really try to understand what conversations are happening on the client side, the organizations that we work with, to understand what’s top of mind for them. Always leaning into, you know, seasonality, any promotions that are going on. Those are some, I think, areas of low-hanging fruit that you can really tap into. But again, going back to it, it really should be a social effort to make sure that you are honing in on the areas that matter most and, again, that are important to your organization. That’s great and circles back to a lot of the three parts that you discussed earlier on in your presentation. So, helpful to see that all tying together. Another question that we have, in your current experience with different clients, how often was offline data being brought back into Adobe Analytics for creating a holistic dashboard for the business? Yeah. So, we try to pull in offline data in every case that it’s applicable and available. So, we try to make Adobe Analytics the source of truth in that case. And if it’s not the source of truth overall, depending upon, you know, if there’s any limitations in place, we make it the source of truth, at least from a digital perspective. And there could be another outside BI tool where everything is further coming together for that holistic reporting. But my recommendation and what I always default to is pulling in that offline data, yes, directly within Adobe Analytics itself. Because if you don’t have that, you don’t have the full story, again, if offline data is applicable for your organization. That goes back again to what I mentioned from framing the problem, making sure that you have the right data available or that you can easily obtain it. You want to make sure that you have all of that living in one place and that you can tell the comprehensive data story.

Awesome. Thank you for that answer. And kind of similarly, if we take a look at the different kinds of data that you can be looking at, are there things that you recommend keeping in mind for exploratory data? So, when trying to be mindful of customers’ feelings or attitudes behaviors, but using that to communicate to those who might be higher up in the organization. So, especially when you’re considering things like getting funding or time to dive deeper into products and projects. The first part of that was, is there any recommendation I have around the exploratory process? Is that right? Yes. And keeping things in mind for exploratory data, things that are more qualitative and how you’re using that to communicate throughout the company.

Yeah, absolutely. So, I think it’s really important, again, with Adobe Analytics to make sure that you’re leveraging your EVARs as much as possible. And that’s your dimensional data, right? To help you really show the qualitative side in addition to the quantitative side. That also goes back to knowing your implementation very well and the scope of your variables at play. And making sure that you know how they map with one another and how to use your EVARs in tandem with your events and the scope of each of those. If you’re not on the implementation side, I highly recommend befriending the team that is. And again, just making sure that you have a tight understanding of that information and that side of things so that really, at that point, fuels your exploratory process and gives you a leg up in understanding, again, what data that you have available at your fingertips. Those are very helpful tips, and it’s always good to make friends within the organization to help those things as well. I’m sure people can relate to that. One question. When would you present directly from Adobe Analysis Workspace versus a different method? Yeah. So I touched on this a bit earlier, but I default to Adobe Workspace when I’m speaking to an analytics team or someone, again, that’s more ingrained in the data and may have more questions on the data side and digging in a bit further, particularly those conversations that are those situations that are more conversational. And I know we’ll get into the weeds a bit more and need to explain more of the path, again, of how we got there. But I personally prefer, because in most cases where we’re speaking to an executive team or C-level team, that’s something that we typically put in PowerPoint so we can be a little more flexible with our data storytelling. And also, there’s likely information that’s coming from other sources as well outside of Adobe, and we want to mesh that story together across all of our departments, at least at Levelwing, where we have client services bringing in their strategy and our media team bringing in their strategy and our creative team so that we can really spell that out and make sure that that’s clear. And that we tend to do within more of the PowerPoint slides as opposed to in platform in Adobe Workspace. Great. Thank you for that. We have a question from Shauna who’s asking, sometimes data only tells part of the story. I wonder if you have any recommendations for integrating findings that don’t always have data to support it. Yeah, that reminds me of an example that we were talking about actually leading up to this webinar, and I’m probably going to butcher this, but Justin, who’s also on the call on the Adobe side, had shared this with me. And I think it was a story about World War II planes coming back, and they were trying to study, based on the bullet holes in the plane, understand how they can better construct the planes. But there were certain portions of the plane that did not have any bullet holes, or the planes that were coming back did not have any bullet holes in this particular area. But then it was discovered that, or someone thought, hey, maybe the planes that are being, you know, shot in this area, those are the ones that aren’t coming back. And so just by looking at the data that’s available on hand, again, it may not tell the full story. So you really have to, I think that’s where the art side of it may come in. I mean, that’s where the human component comes in, right? And why, again, we’re all employed, that machines aren’t taking over our jobs just yet, because you have to consider other aspects as well. You have to pull in all of your information and make sure that you’re looking beyond surface level and look into the areas that may not be so readily available, and ensure that your solution does truly make sense, and again, that you can quantify it as much as possible.

That’s a great example, and I think really does showcase what you were discussing. What you were discussing about that blend of art and science really truly with data storytelling.

On a similar sort of wavelength, are there any tips that you have for handling telling a story where data from two sources might conflict? So specifically, perhaps on-site data that you might see in Adobe Analytics versus other data from social or display? I think that goes back to just making sure that all stakeholders are looking at the same source of truth, and that’s something that I think you need to establish in the very beginning of any, you know, the kickoff of any relationship, whether, again, you’re on the agency side or you’re partnering in any case, or, you know, on the brand side, if you are, again, coming together to really understand data collectively, everyone needs to agree on, you know, a source of truth, and it’s not to say that, you know, one platform is more correct than another. It’s really looking at data through a different lens, right? If you’re looking at an ad-serving platform like Facebook, for example, Facebook is a walled garden. It’s going to see everything that’s happening from a social perspective. It’s not considering all of the other data sources. So I think that’s, you know, just, again, another moment for teaching and educating those around you and, again, establishing that analytics culture and your source of truth that everyone is working from. Great questions. Great advice, and I think most people can find that definitely very relevant in working with different stakeholders and cross-functionally, so that’s very helpful. On a more lighter and perhaps relatable note for our audience, have you ever had a time where data storytelling might have fallen flat or even failed? Yes, there is a situation that does come to mind. It kind of ties back to what we were talking about earlier in bringing offline data into Adobe Analytics. So there was a situation with one of the clients that we still work with today. We worked with for several years. I think when we were first starting off the relationship with them, we were a few months in and we were presenting to the CEO and we were talking about our media performance and the impact that it was having on e-commerce sales. And we had passionate analysts on the call talking about this and really hyped up about what we were seeing in terms of the results from the marketing perspective and e-commerce perspective. And the CEO was kind of quiet for a while and then he mentioned, you know, that’s great from the marketing side and the e-commerce side, but overall, sales are still down. And that really stuck with me because at the time, we didn’t have insight into offline data and what was happening in store. So yeah, it kind of fell flat essentially because we didn’t have all of the data available at the time. So it was definitely a learning lesson for us and we do have that in place today with them. Like I said, we’re still working with this client and have a great relationship and we have a holistic reporting solution in place that looks at all data sources. So that definitely was top of mind when you mentioned the question.

Yeah, that’s very relatable and I’m sure many people understanding not always having all of the data that you need. So that was a great example. I know we’ve asked a lot of questions and we have a lot more, but we’ll just focus on a few more knowing that we’re coming towards the end of this session. But bringing it back, how do you recognize really when an observation that you have in the data is something worth a data story? Yeah, that’s a great question. And I was reading an article, I think it was just within the past few days from Brent Dykes, who is with, I think he started Analytics Hero, and he was talking about what an insight truly is. And he talks about it and that it’s truly a shift in your mindset, that it really challenges all of your assumptions leading up to that point. And I think that’s a really concise way of thinking about it, that that is when you can discover that this is something that is worth telling a story around, that it also has impact, it has wide impact that you can potentially quantify too through data as we discussed. So not everything, again, as I mentioned, is suitable for analytics or building a data story around, but I think that that’s something that comes with experience, it comes with time as you build your experience and move further down your path as an analyst.

That’s very helpful. And it sounds like, along with most things in data storytelling, practice makes perfect. One question around some of the data that you might use for your different analysis, how do you address questions or concerns that might come up related to data hygiene in your reports and your presentations? So data hygiene, data integrity is something at leveling that we address from the very beginning. It’s something that we kick off every relationship with, is really establishing the data foundation and ensuring that we have solid data integrity that we can trust and believe in. Because otherwise, anything that you do further downstream is, it really, you can’t lean too much into it. I mean, it’s not, it can’t be trusted. So I would recommend doing, going through a similar exercise and making sure that you have full trust in the data that you have available. If there are any gaps, dig into that, understand where platforms may not be speaking to each other. Again, if you’re not on the implementation side, make friends with those that are, and really dig into and understand where the data is coming from and how you can truly use it on the analyst side and that you’re using it in the right way.

Yeah, that’s very helpful. And we will take a few more questions. One we have from Caroline, do you ever use journey mapping to identify improvement opportunities based on what your clients are feeling using your workflow or your solution? Yes, absolutely. We were actually just talking about this in, we have every Thursday with my team, we have an analytics huddle and we talk about things that impact our industry, impact our clients. And the topic today was CRO, conversion rate optimization. And we were talking about leaning into like RFM modeling, recency, frequency, and monetization, and really understanding maybe the 20% that could have the 80% impact, if you will. So again, we’re talking about the impact, if you will. So identifying the clients that are, sorry, the customers and your audience that’s most valuable and understanding their path and how they’re moving through the website to really hone in on that and remove any potential friction points along the way. And really, again, be able to understand what may be keeping others from hitting that conversion that you’re aiming towards and making sure that there’s seamless transition throughout your conversion funnel.

Awesome. Thank you for that. And this is a question that we have coming in. Do you think that there is a pressure within marketing departments to construct positive stories as opposed to negative ones when data storytelling? That’s a great question. I think it’s, you know, it can be difficult to truly tie back at times the marketing, the impact that marketing is having on sales, whether it’s, or particularly if it’s offline sales. And you have to find what is right for your business and making sure, again, that everyone is looking at the data in the same way. So whether that’s a matching process that you’ve implemented to be able to tie online marketing efforts to offline in-store sales, whether that’s a digitally influenced revenue model where you’re using average conversion rates to, again, to tie back to maybe something that’s happening offline, or if you have a marketing mix model in place. And that is, you know, at least in my recommendation, that’s where you need to be, is building out that marketing mix model and honing in on the channels that can really bring you incremental growth. And if you have the right tools in place, and again, everyone, all stakeholders are looking at the data through the same lens, that at that point, you’re just, for the most part, relaying facts. It’s, you’re agnostic to the outcome in a way, but you can also, again, understand and pinpoint the areas that you need to address and the gaps that are there to have better outcomes.

Yeah, that’s helpful, especially understanding, you know, all the different sides of the problem that you may need to address. A question that’s related to that topic, when specifically looking at reporting an analysis workspace, do you include the unspecified bucket in your reporting, since it can be a mixed bag of what’s included and can be hard to analyze? Or what would your recommendation be there? Yes, the unspecified bucket. Yes, so no, I typically do not include that, not in a data story, data storytelling exercise. That may be in your exploratory portion, where you may want to see that, because that could indicate that maybe you need to update your classifications, that something’s not properly falling into the classification that you would expect it to. So you want to keep an eye on that and make sure that that’s not the case. Also, it could give you the indication that you may not have a great understanding of the scope of an EVAR and may need to better understand, again, how that data is being collected. So I think the unspecified bucket is something you have to keep an eye on and make sure that it’s monitored and dig into it. But it’s not something that I generally would include in a data storytelling situation. I’d want to investigate that more and really understand what that bucket is representative of before putting it in front of a larger audience.

Yeah, that makes sense and appreciate the answer there. When you’re taking a look at just these data storytelling exercises and really planning for that, one question that we have is, how many hours does it take, really, overall to find, analyze, and prepare these data stories? Is it as time-consuming to find your real insights and not necessarily just observations? Oh, interesting. Yeah, that’s difficult to put just a general time around these exercises. I think it really depends on the type of presentation or conversation that’s happening. If this is an ad hoc analysis that’s coming up and something that you’re proactively going after, or whether it’s a QBR, a quarterly business review, it could look quite different in terms of the time that you’re putting into this. But again, I’ll hammer home the fact that it’s so important to properly frame your question in the upfront. That’s something that is, I mean, it’s intuitive, but it doesn’t always happen. And if you don’t properly frame your question in the upfront, it can really hurt your end goal or when you’re making a recommendation or you’re showing your data and explaining through data, as we discussed. It could cause a disconnect if your problem is not properly framed. So yeah, but a timing perspective, I mean, I would say to put some guardrails around it, like make sure that you’re not just going down deep into a rabbit hole, that you have, again, a framework around what you are digging into from a data perspective. And if it’s a larger conversation with other teams involved, I would make sure that your story also can fold into the larger story as well. So a lot of things to consider, again, based on the situation and the audience that you’re sharing it with.

Great. And I know we are coming up on time, so we will end with a fun one. Who is your favorite data storyteller? Oh, nice. I don’t know if he considers himself a data storyteller, but if I had to name one person, I would say Adam Grant, who, for those of you that tuned into Adobe Summit, I’m sure you got some exposure to him. I know he was a speaker this year, but he is an organizational psychologist. I just learned of him, I think, in the past year, but he does a great job of just breaking down complex situations into really relatable, simple stories. And he’s also really easy to follow, and his voice is really, I guess, soothing in a way. And that can help, too, when you’re telling stories. So he is just great. I highly recommend, if you are not familiar with Adam Grant, checking him out. He’s got a great podcast, too, called Work Life. But yeah, I’d have to say he’s definitely a great data storyteller.

That is a great answer. And I’m sure most people, if they aren’t familiar, will definitely be checking him out after this session. But I really just wanted to say thank you again, Amy, for your time today and for sharing such great information. And thank you, everyone, for joining us. As mentioned, the recording for the webinar, as well as the slides, will be shared on the Adobe Analytics community as a follow-up, as well as any questions that we might not have gotten to during today’s session. So that is all we have for today. But I hope you all have a great rest of your week. Thank you so much, everyone.

Thank you.