Increase your productivity - May 2023 APAC Adobe Analytics Skill Exchange

Thank you so much for that introduction. And hi, everyone. I’m really excited to be here to walk you through the last part of our session, putting all together. So far, we’ve learned how to get started in Analysis Workspace, how to analyze data in Analysis Workspace. And in this last part, we’re going to put it all together. My name is Maria Laura Lapointe. I’m a Senior Product Marketing Manager for Adobe Analytics and Customer Journey Analytics. This September, I’m turning five years at Adobe, and I’m really excited to walk you through this session and share our knowledge here. Feel free to drop your questions in the chat box while we’re going along the content. And this is meant to be interactive. So to get us started, let’s go over our learning objectives. So first, we’re going to learn how data is collected into your reporting. Then, I want you to walk away knowing how to create some basic visualizations. And finally, I want you to know how to share data with your business users at your organization and how to inform your organization to make data-driven decisions. Okay, everyone. So let’s get started on how Adobe Analytics works, and we’re going to get started with data collection. When you think about Adobe Analytics, our main value is our ability to differentiate through our vast customer intelligence capabilities. And we have four pillars. Our first pillar is data collection. Even though we come from a legacy of web analytics, we’ve expanded the channels that we can cover. And now we’re able to collect data from voice assistant, connected card, IoT, wearables, web, mobile, email, and OTT. So virtually any channel of interaction with your users, we’re able to collect and analyze. The second pillar is data processing. We’re able to slice and dice the data in a way that is going to be relevant for your use cases. For example, we have a capability called Context-Aware Sessions. This is the ability to break down sessions into a lens that is relevant for the different touch points, for example, mobile apps. Mobile apps, you will require a shorter session span than a long video format. The third pillar is machine learning and artificial intelligence. We have capabilities like virtual analysts, journey analytics, segment IQ. These machine learning capabilities that are powered via Sensei, they will allow you to surface insights that otherwise they will go uncover and also to save time to be able to make data driven decisions. And finally, is our ability to turn that insights into action through our integrations with audience manager and also target and campaign. We’re able to turn those insights into actions because we have bi-directional integrations with other Adobe solutions. Now let’s talk about the analytics cycle. The first phase of the cycle is define what are your business requirements. Your business requirements are what are the use cases that you want to solve when you’re using Adobe Analytics. Define what are your KPIs and what is going to be the North Star that is going to drive your experience. Second is design, your ability to translate those business requirements into a design and then deploy. You are going to have to deploy the collection of your data through SDKs and also to JavaScripts. So be able to implement the technology and capture the data that you need to solve for those business requirements. The fourth phase is analyze. Be able to bring together the reports and to dig into the data so you’re able to answer your questions for your business users. And finally is act. You will have to act based on the insights that you were able to bring from your reports. All of the cycle is an ongoing cycle because your customer experience are constantly changing and your business will constantly keep asking new questions. Next, let’s go into the Adobe Analytics value framework. The lowest level is data integrity. Second, we go into reporting insights and modeling. Data integrity means do you trust your data? Is the data that you’re collecting representing the behavior of your users? Then we go into reporting, which is creating the reports that are going to be able to solve your business questions. The third level is insights. Being able to make data-driven decisions based on your reports. And finally is modeling. This phase is where the data scientists and statisticians can make predictive modeling based on the data. Now let’s move on to data collection. When a visitor comes to your digital experiences such as web, mobile app, and also websites, the Adobe Analytics code is in your data layer of your digital properties and also through launch. The website is going to invoke an image and the Adobe Center servers will process the request recording the analytics data and return a transparent one by one pixel to the experience. Basically the data is sent to the Adobe reporting and you’re able to access this data through the reports. Now let’s cover how Adobe Analytics collects data into reports. So our most fundamental container of data in Adobe Analytics is called report suites. Report suites is basically a container that allows you to segment your data. We also have global report suites that aggregates all of your report suites. And lastly, we have virtual report suites. Virtual report suites are used when you don’t want a specific team to have access to all of the data or to all of those containers. Now let’s learn how to build some basic visualizations in Analysis Workspace. It could be a little bit tricky when you start using Analysis Workspace and creating visualizations because there are so many to choose from. So here there are some quick tips to get you started. First identify the most important data. So identify what are your metrics, what are your dimensions, and the segments that you want to analyze. Then choose which is the best visualization to represent that data. Make sure that you’re able to tell a story through those visualizations and also throughout the report. So make sure you have titles to the report and you’re able to connect the dots through a story. Then remove unnecessary noise. Make sure that the data is clean and the reader is not being distracted by unnecessary noise. Highlight what are your main takeaways. What are the takeaways that you want your audience to take away from the data? And finally make it easy to consume. Depending on your audience they could have different preferences to consume the data. It could be through a PDF or through a direct link. Now let’s go into choosing the right visualization. Analysis Workspace offers a wealth of visualizations from chartboards to scatter plots and beyond. When choosing the right visualization it’s important to keep in mind what is the job that you want to do and what is the takeaway that you want your audience to consider. We break down our visualizations into five categories. The first one is comparison, trends, parts to hold, relationships, and distribution. My favorite visualization is conditional formatting for my tables, also stack vertical, chartboards and histograms. For example histograms are very useful when you want to bucket your visitors. For example when visitors come from one to three times to your website a week, four to five or six to seven times. So it comes really handy when you want to analyze and bucket different behaviors of your users based on metrics. As I said like Analysis Workspace offers more than 20 visualizations and here I’m going to show you where to find those visualizations. If you are in Analysis Workspace you navigate to that left rail and you’ll see that visualization icon. Once you click on that you have more than 20 visualizations to choose from and you will be able to drag and drop that visualization into your project. Now let’s go over two quick tips to create visualizations. When you’re on any cell in Analysis Workspace you can do a right click and you have a menu of options to choose from and you can choose visualize. The second tip is when you have a table you go through to the dimension and hover over one of those dimensions and also click on visualize and there you’ll be able to see the visualizations and Analysis Workspace will make a guess on what is the best visualization to represent your data. Now let’s get started with some basic visualizations. A very common visualization is the summary number. If you go again to the left rail you’re able to drag and drop the summary number visualization. The summary number visualization is really useful when you want to highlight a metric or a north star metric that is very important to your business. Now something to keep in mind is that once you create a visualization you can adjust the settings of those visualizations and something that you need to pay attention to is that you have the ability to lock your selection of your data. So if you click on the gear button within that visualization you’ll be able to lock your selection of the data also make the legend invisible and abbreviate the value of your number. This will help your audience not pay attention to unnecessary details. Now let’s go into summary change visualization. The summary change visualization helps you understand what is the change between two numbers. So when you have a table and you select two numbers and you have the summary change visualization this will automatically represent what is the change between those two numbers. The first cell selected will be the numerator and the second cell selected will be the denominator.

Now let’s take a look at a quick demo on how to create summary number and summary change. So first I’m gonna find my metrics which is visits and I’m gonna find my weeks and drop it on their dimensions and I created my table to start with and from the left rail I’m gonna I’m gonna drag and drop the summary number visualization. As I click on different cells you’ll see the summary number changes. I can also change the name and the title of that summary number and I’m gonna call it visits in April. I’m gonna drag and drop also my summary change. As you see when I click on two different cells this number changes and I want to make sure that this number is locked so it doesn’t continue changing when I click on different cells from my table. I can also show the wrong number or I can choose to show the number as a percentage. Now let’s go into the line chart. The line chart is a very common visualization that we use in Analysis Workspace and you can analyze your data based on different granularities. You can choose from a minute to a year in the granularity of the data. The dimension of the data will always be time and this is a great visualization for example when you want to track your campaigns or if you’re running a special event. You again and again when you’re running a campaign you can track the different levels of granularity of your metric. Now let’s go into a quick demo of the trend line visualization. I already have my table built out and I’m gonna drag and drop the line visualization. So when I click on the different cells that trend line visualizations changes and again I can make sure that I’m using the right granularity that I want to analyze and I can also lock the cell or the data that I want to analyze in the trend line. Another rich visualization is the map tool. The map tool can help you analyze the behavior of your users across different regions in the world and for example you could compare how different users are behaving on a specific city across the US and it could also be a quick way to create segments based on the activity that you’re seeing from your users across regions. The next visualization is follow-up visualization. This visualization is very helpful when you want to analyze how your users are falling out of your digital experience. For example if you have a mobile app and you have a checkout experience you can drag and drop the different touch points that you want your user to follow. For example from being in the car and seeing a specific product all the way to submitting that order. Also here using the follow-up visualization you could compare how your mobile experience is to your web experience. Now we are able to have a complete project. We were able to use the summary number, the summary change visualization, we are able to incorporate a trend line, we can use also the map tool and also we can use the follow-up visualization. So now that we are able to create a full report and analysis workspace let’s start sharing that data and democratizing that data with our business users and our stakeholders. So I’ll leave you with this quote, an analyst job is not to pull data, your job is to translate data into stories that drive action and results for your business. So it is really important that you are able to help your business answer business questions and also to empower users and your organizations to make data-driven decisions and help drive the strategy of your company. So now let’s go over a framework that will help you take your data to insights. So first make sure that you understand the request from your business user. Make sure that you know how that data will be analyzed and what actions your user wants to take with that data. Then understand your audience. Understand how your audience wants to communicate, how they want to use that data and what is their preferred delivery method for that data. Speak that language. This means really understand what are relevant metrics for your business, how does this data fit within your specific part of the organization and what are downstream impacts in the ecosystem if they have the data available. Then know the value of your insights. Make sure that when you’re sharing the data that it is shared with context. So this means that you have a representation of how does that data affect your business and also that you’re able to offer a benchmark for that data or a comparison for that data. And finally question your assumptions. So when you start digging into the data, make sure that you are questioning yourself and try to start from a neutral analysis point of view so you’re not cherry picking information and making your analysis bias. Now let’s start democratizing those reports and insights. So if you are in Analysis Workspace and you go to the top part of the tabs, there is a tab that will allow you to share your project. So you can share your project and also here within this tab you will have the ability to curate your project and curate a component of your project. The quickest way to share your analysis is using a link. So here in this quick demo, you can go to share tab and then you share a link. You get a copy of that link and you can share it via email or slack. You can also choose to share specific sections of your report so you don’t have to share your entire report. Now if you don’t share the link, you also have the ability to download the data into a CSV file and PDF file and you’ll be able to send that data through email. And this will be a good fit for recipients that you know they aren’t likely to looking into Analysis Workspace. Then for recipients that don’t usually log in into Analysis Workspace, you can share the data through a CSV file and PDF. And here I’m showing you how to go into share, send a file through a PDF and then you’re able to send that file. You can choose to send that file on a frequency. It could be sent on a daily frequency or on a weekly frequency. And this is really helpful when you want to start automating those reports being sent out to your recipients. Now here I’ll show you how your recipient will receive if you choose to send that data through a PDF file. They’ll receive an email and the main difference between going into Analysis Workspace or receiving that data or that report through email is that when you receive it through a PDF, it’s not going to be interactive and then that information is going to be just static. Also when you are sharing your project, consider if you want to curate that project. That means that you don’t share all the components of that project because potentially for your recipient it could be overwhelming to see all the metrics and dimensions that you want to analyze. So if it’s appropriate, consider curating your project and to curate your project you need to save a new project so it doesn’t overlap on your old project. Another method to share data in Adobe Analytics is to use the Adobe Analytics mobile app that is called Adobe Analytics Dashboard. And here I’ll show you in Analysis Workspace you can start creating a mobile scorecard from the pre-existing templates and you can start drag and dropping the metrics that are going to be relevant for your recipient and you can continue curating the experience based on the metrics and the segments and the dimensions that are going to be relevant for your recipient. Okay so we’re going towards the end of our session and today we’ve covered how to create visualizations, how data is ingested in Analysis Workspace and also how to share the data with your organization and help your organizations make data-driven decisions. So now that we’ve covered all of that, I want to share some resources that will be handy when you want to continue learning your Adobe Analytics learning journey. So the first resource is Experience League Community. If you go to the Experience League Community and navigate to Adobe Analytics, you’ll find a very energetic community that is constantly asking questions. You can also ask your questions there and you have the ability to send your features requests through the Adobe Experience League Community. The other resource that is very useful is our Adobe Analytics YouTube channel. Here we post content constantly every time we have a new feature and also when we have Adobe Analytics webinars or customer journey analytics webinars, they get posted to our YouTube channel. So go there and continue learning all everything about Adobe Analytics. Okay so now let’s get to your questions. Final round, I’m back with Servi. Thank you for joining us again. So very happy to be here. I’m glad you’re here because we’ve got a lot of questions to answer. So let’s start off. First question that’s come through from Divya. Can you explain about unspecified data? Yes, so unspecified is a very fairly common one that we get typically. So there are different scenarios when you can get an unspecified value. Things like when an event gets fired without a conversion variable. So for example, a user comes to your website and makes a purchase and if you’re capturing that order in E-VAR1, let’s say for example, so makes a purchase without any E-VAR1 value, then there is no value to attribute to this order essentially. So therefore it automatically gets attributed to unspecified. Another example would be unclassified data in classification reports. So when viewing any classification data, any value that doesn’t have data associated with that particular classification returns unspecified. So one way to resolve that would be classify the parent variable value. And then third that I can think of is like breaking down reports when only breakdown of the reports when only one variable gets fired. So the second variable was not seen or even if it persisted from the previous hit. So that’s when the dimension item is unspecified. Okay, makes sense. Thank you. Another question has just come in. It’s regarding the use of data that is parked in a different data pool. Are we able to join with Adobe Analytics if we have a unique identifier? Coming from Esther. That’s again a customer journey analytics use case. So not perhaps so much on the Adobe analytics one. But yeah, with customer journey analytics, we can bring in data from other places and we can join it based on one of the identifiers and you can start analysing that data together with your digital analytics data from Adobe Analytics. Yeah, so all of that stitching we’re thinking customer journey analytics is that place to go? I mean, basically, if we have different channels as well. Yeah. Yeah. Yeah. Makes sense. So many people in this group might want to go and check out one of those customer journey analytics tutorials in Experience League. Cool. I have a question actually that someone’s raised and I think this is something that a lot of my customers ask about as well. And it’s something that is front of mind for many people at the moment. It’s AI and ML. So we’re thinking like about efficiencies here and how can we use our tools to do those kind of easy jobs for us, I guess. So what are some of the AI and ML capabilities that are available in Adobe Analytics? So good question. So there are quite a few AI ML capabilities. So one would be the one that we just talked about in the first session was anomaly detection as well. So anomaly detection lets you uncover unexpected trends. So when you when we are doing a trend line or time series visualization, anomalies get detected and gets highlighted essentially. So that’s one of the ML capability. And then next question would be, oh, why is this anomaly coming up? So that’s where we have contribution analysis that helps you understand what’s moving those metrics. So as in when you detect that anomaly using anomaly detection, you would want to understand what factors are contributing to that anomaly. And that’s when we can start using contribution analysis to understand what factors are leading to those anomalies. Next one is, of course, we have attribution IQ to attribute different touch points leading to those outcomes, which marketing channels are leading to a sales order or a purchase or form downloads. And last one is also like segment IQ to understand why different segments are performing differently and understanding their overlaps and stuff like that. So there are a few different ones. Again, Experience League would be the place where you can understand a bit more about those features. Excellent. So lots, lots in there just to help us do our jobs more easily. Yeah. Lots of smarts. A question that’s come in around EVAs and props. You mentioned EVA before. Could we take that one and just, I guess, go to those very basics and explain the difference between what is an EVA and what is a prop? Yeah, sure. So one is basically when you are trying to do a traffic variable and the other one is basically when we are doing a conversion variable. So EVAs persist throughout the journeys and props are essentially when, props do not really persist throughout the journeys. Yeah. Makes sense. I’ve got a good question here around cross verification and just doing some checks to make sure that our analysis is correct. So the question is, can we view the underlying data in Adobe Analytics while we’re creating segments just for the purpose of cross verification? Not 100% sure if I understand that correctly. Is it about like the raw data that we are creating or is it like, hey, we’ve created this segment. Does also, okay, so if I understand this correctly, if it’s basically you can drop that segment and if you’re capturing any member IDs or customer IDs, the other way would check that would be once we have created that and do some behaviors, do a test and track those customer IDs against that segment and check if it actually is those segments and they have visited those pages or not really. So there are different ways of checking some of those scenarios. But again, it depends on the volume of data and when are you doing those tests? Yeah. Okay. Makes sense. We’ve got a question that’s come through. I’m not sure if this is a little bit in depth for our session today, but let me ask it and then we’ll see what you think. So question is how do we eliminate the impact of GCL IDs, the Google click IDs or Facebook click IDs that are recorded after this CMP ID, the campaign ID, so many acronyms that affect the grouping of the marketing channel or other attributions that rely on that campaign ID. So the question is that now it currently makes that data group to under unspecified. So is there a way to eliminate that impact? Might be a bit in depth. Yeah, that’s a bigger conversation. There can be multiple ways to address it. There can be, yeah, I would say that’s a really good question to post it on the community group and we can start discussing those questions perhaps in there. Excellent. Okay. I’m going to come back with some more info on the Experience League communities. I think you’ve given me a nice segue into that, which we’ll come on to in a minute. So thank you for sharing that question and let’s see if we can get that answer for you through our communities. I have a question around page names and character limits. Are there any character limits for when you’re putting in those page names? Yes, Adobe Analytics with most of the variables there is a character limit, of course, just to make it much easier for customers to read and stuff. So page names, I believe, has like a hundred character limit. And then again, while we were talking about those conversion variables, which was EVAs, there are two hundred fifty-five character limit, I believe. Yeah. Again, you can double check that on Experience League. Okay, cool. Thank you. And another question, which kind of relates back to the question that we had earlier around verification of data. Oftentimes when you’re working as an analyst and or you have many people in your company that are testing experiences or want to test through and they’re putting test data into the system and we want to try and keep that as clean as possible and make sure that what we’re analyzing is actual customer proper data. Is there a way to exclude the company’s test data? Yes, again, there can be like multiple ways to do that. One, you can exclude your IP addresses. If you’re in a company and building, there is a feature in Adobe Analytics where you can exclude those internal website activities from like site testing and employee usage. So that’s what is called IP address exclusion. And it would be in your admin guide section that you can get some more information on. Perfect. A question as well around exporting Adobe Analytics data. Are there different ways to do this? Yes, there are three or four different ways of doing exporting your Adobe Analytics data. One, of course, using the workspace reports that we were talking about from sharing perspective and we can download the Excel. But then there is a limit of 400 rows, I believe, from the analysis workspace report download. But there is data warehouse export that we can do. So it would be in your admin section or a component or tool section. I’m just getting confused which section it is. So data warehouse is one of the options where you can download your data. Then we have data feeds where you can feed in like raw data feeds every 15 minutes or something like that to your any of the SFTP or S3 bucket. And then I believe there is another feature to extract that data. Let me have a look and come back. But yeah, those are some of those different ways to download and export your data to your S3 instance, SFTP or Excel.

Perfect. The question here, which is probably really relevant to lots of the customers on the call today about loyalty. So we want to understand who are our customers that are visiting time and time again. So is there a way to analyse how many times a customer has visited? So if the same customer is visiting multiple times? Yeah, sure. Again, it depends on your implementation. But there is like you can try again the frequency number and the page numbers as well. How many times a customer has visited your website? What’s the frequency? Are they visiting? There is another visualization in Adobe Analytics, which we would have touched on is the cohort analysis so that you can see if a customer visited the page on week one, how when are they actually completing that transaction? Is it week two, week three or day wise? So there are different ways where you can start analysing those data. And then from that analysis, if you right click on that, you can create a segment if people have visited more than four times, if the frequency is more than seven days or something like that, seven times, then you can start creating those segments and start analysing. Or if you want to activate that elsewhere, that’s how you can do it.

Perfect. And then final question, I think because we’re running close to time and you just touched on different visualisation, I think this kind of ties in about trend lines. Is there a way to add a trend line so that you can kind of try and foresee the future? Not really a future, I guess, but yes, we can do a trend line. And I believe that was covered in Ashok’s video. So if you have a free form table where we have, let’s say, page views and days, and this brings back to my first question as well, like which ones do you start using upfront? So if you have like a page view free form table, if you right click on that, click on visualize, you can start a trend. And that’s when you can see a trend, bar line trend of how the visitors or page views happened over the month or year at a daily basis. And in that trend, you can see the anomalies if there were any anomalies. And from there, you can start doing contribution analysis to understand what factors are leading to those dips and anomalies as well. Perfect. I guess we can’t really see the future, but that’s going to help us understand the past, right? Yeah. So, Luxemi, this is a wrap on this portion of the Skill Exchange. Thank you so much for spending your time with us today. You are such a font of knowledge. And I really think that all of your insights are going to be really valuable to our group of learners today. So thank you for your time. Thank you. It was my pleasure. I hope everyone enjoyed the session and perhaps see you in one of our Experience Leagues forums. Thank you again.