Analytics Learn Experience Makers Spotlight

Transcript

Now that you’ve gotten a comprehensive overview of the capabilities and potential of Analysis Workspace, we want to present some exciting use cases from customers that started off exactly where you are today. We’ve got Mandy George, who’s a Senior Analyst, Online Experience Analytics at Home Depot, and Kayo Walton, who is Director of Analytics for Voice of America. And we’re so grateful that they’ve made time to present for us. They’ll be joining us together for a live Q&A after we watch both of their presentations. So as you drop your questions into the chat, please be sure to direct your questions to the specific speaker. So Mandy George will kick us off with tips for how to organize your data efficiently so you can use it effectively in Analysis Workspace. They have a lot of great advice about the best ways to organize segments and metrics and how to use annotations effectively. So Mandy, welcome to the Skill Exchange.

Thanks for that introduction. My name is Mandy and I’m going to be talking to you about increasing your data efficiency with Workspace. I’m a member of the 2022 Adobe Analytics Champions Program and I’ve been with Home Depot since 2016. I’m a Senior Analyst on our Online Experience Analytics team and I’m coming to you from Niagara Falls, Ontario. I’ve got three cats, Merlin, Arthur and Morgana, who like to think that they’re helping me with analytics when they sit on my computer, but that isn’t always the case. So today I’m going to be going over three main tips to increase your data efficiency and workspace. The first tip is to start off strong by setting your default dashboard preferences. The second tip is to create strategies to organize and manage your custom metrics and segments. And the third tip is to use annotations effectively to identify factors that impact your data. So let’s start off with setting your dashboard preferences. The quickest way to get your preferences is from the link on the landing page at the top right hand corner. If you haven’t already switched over to the new landing page, you can use the toggle at the bottom left to turn it on. The new landing page actually has several advantages over the old one, including better filtering and sorting for looking at your existing dashboards and additional links to get to other useful features. Once you’re in the preferences, you want to click on project preferences. So these are what influence all of the new dashboards that you create. By setting these at the beginning, it can save you time when doing ad hoc data pulls. Of course, there’s also going to be times where you’re going to do full on dashboards, and you can always change these settings in individual dashboards when the needs don’t align with your default options. But when your stakeholders just want a quick number, having these default preferences set can save you time and make sure that you’re pulling your data correctly. So the first option is the density of the view. This changes how much white space is in between each of the rows. This is more important for larger dashboards than ad hoc data pulls, because having less white space means that you can see more data all at once, and there’s less scrolling that’s required. However, even for ad hoc data pulls, it can impact the readability of the report. Personally, I find that the compact view is a bit squished, but that the comfortable view is a lot easier to read. Of course, this comes down to personal preference, but it can make a huge difference. Additionally, having a standard selection across your organization can make it easier for stakeholders to move from one report to the next. The next preference is the color palette. You can select one of Adobe’s defaults, or you can create a custom palette using between two to 16 colors. To do that, you just need to enter the hex values separated by commas in the field below. Again, this one is more important for larger dashboards and reports that’ll be sent out, but if your organization has a specific color scheme, using it as your default can make your reports look better and make them all fit in if they have the same color palette. This next option is, in my opinion, the most important one. If your organization has multiple report suites or even virtual report suites, making sure that you’re in the right one will have a huge impact on your data. For example, if you have a regular report suite that has all of your website data, and then you have a virtual report suite that excludes your bot traffic, the numbers that you pull from these two could be very different, especially when you look at metrics like card additions or conversion. By selecting a specific report suite to be your default, it will ensure that any new dashboards that you create will have the right data without having to manually switch it. This can speed up time for those ad hoc data pools so the data doesn’t have to load multiple times from different report suites, and you can also reduce the risk that you’re reporting on the wrong numbers. In addition to the report suite, you can also set your default calendar time period. If you’re often asked for ad hoc numbers for a particular time period, such as month to date or the previous week, you can set this time period as your default. And if your business needs change over time, you can always update these preferences, but having them set at the beginning can be a real time saver. The next group of settings are the panel type, number format, and CSV delimiter. So the panel type default can be based on what type of initial actions you usually do when you start pulling data. So for most of us, having the new panel set to a free form table will give us a place where we can quickly pull in different types of data. For the number format and the CSV value, those can be personal preference as it doesn’t change your data, just how it appears. My recommendation is that you coordinate with other users in your organization that use analytics and pick a standard type. By having the same values across all users, it ensures that when you show a stakeholder your dashboard, the format will be the same, they’ll be able to recognize it and make it easier for them to read. So finally, you have the option to turn on or off anomalies, sparklines, or annotations. You can also set these in individual dashboard settings, but having them set at a user level can save you a bit of time. So anomalies can indicate if there are significant changes to your data and quickly point those out. So it can be helpful to have those defined instances where something might be impacting your data like missing traffic or spikes during certain times. Sparklines are those trend lines beside the grand total numbers in free form tables. They give you a visual representation of your data over the time period that you have selected. Both of these can add another layer of data to your dashboards, but they can also be a little bit visually distracting. So it’s personal preference whether you want to have those on or off. Finally, annotations, which are something I will be getting into a bit more later, are manually added by users in your organization to indicate important factors that influence your data. For some people, this can help provide another layer of information when pulling numbers. But for some types of reports where you just want to see an overall picture, such as for high level executives, it might be a little bit too much information and could be less useful. The next tip I’m going to talk about is maintaining your segments and metrics. So in addition to the default setup by Adobe, you can add custom metrics and segments for any of your report suites. These can help you answer questions that are specific to your industry or organization. In order to maintain them, there are two ways to go about this. You can click on components in the menu at the top and select the type of component that you want to manage. Using this way, you can maintain multiple metrics or segments at once. But if you only need to maintain one at a time, you can always right click on that particular component in the left hand menu of the workspace and use the eligible action items there. One reminder about creating custom segments and metrics is that by default, they’re only visible to the person who created them unless they’re explicitly shared with other people. If we go into the components menu and select either segments or metrics, it will take us to a landing page for that component that shows us all of the values. For example, in this one, we’ve got segments. There are three tips that I have for maintaining your segments and metrics. The first is to create a visual indicator in the name of the component that you want to use. Because anyone across your organization can make segments and metrics, and because it’s possible to make them in a way that doesn’t pull the data quite the way you want, it is important to indicate which ones are the right ones to use. Also, if your organization has a lot of people that access workspace, having a standardized naming convention can help make data pulls simpler and quicker. In our organization, we put an AV at the end of the name of the metrics and segments that have been checked over by our analytics team. That way, all of the users know that they’re using the analytics verified or AV components when pulling their data. The second tip is to use the approve and favorite features. The main difference between these is that marking something as a favorite only does so for yourself, but using the approve option marks it as approved for all users that the component is shared with. And third is add meaningful tags. So you can add multiple tags to each component to help others find them. If you look at the left-hand side of this landing page, you can see a list of all the tags that are currently in use, and you can actually filter the list for the components that have a specific tag. The real benefit of using tags and favoriting or approving components is when you’re trying to find them to add to your dashboard. In Workspace, if you click the filter icon on the search components bar, it gives you a number of quick options. First, it has approve and favorite, which if you click one of those, it will automatically filter to all components with that setting. It does also then give you the option to look at either all segments or just all metrics or so on. But below those, it also gives you the options for all of the tags that you’ve created, and you can limit the component list to the tag that you’re looking for. Using these can quickly help you find the components that you want to add to your dashboard, and it’ll also help other users know that they’re using the right components as well, so pulling data is efficient and accurate. And the final tip I’m going to go over today is about using annotations effectively. So annotations are actually a relatively new feature in Workspace, having just been introduced in the last year. These are actually a really powerful tool for you to add visual indicators to your reports when there’s something that is impacting your data. All annotations are manually created by users in your organization, and similar to segments and metrics, they are by default only visible to the person that created them, but they can be shared with other users in your organization. Annotations won’t show up in the components menu on the left of your dashboard, because they aren’t something that you drag into the tables to pull data. So to edit them, you need to click components in the menu at the top, and then select annotations. The annotations landing page is very similar to that of segments and metrics. Again, you have the option to add tags to help sort and identify them. You also have the option to add them to your favorites. This way, when you’re looking through them, trying to keep your data up to date, it’s easier to find the ones that you’re looking for. The main difference here is that annotations don’t have an option to be approved like the other types of components do, but you can still share them with other users in your organization so everyone has access to them. So when do you want to use annotations? These can be used to indicate any internal or external factor that’s going to impact your data. Say, for example, you had an analytics bug and lost tracking on part of your site for a couple days. You can create an annotation for those days, that way users who are pulling data will know that something out of the ordinary happened. You can also use them for stuff like new features being launched, particular sales that you’re having, or other external factors, like in the example here, store closures that happened during the pandemic. So for those of you that are outside of Canada, when the pandemic was going on, the province of Ontario closed all retail stores for a period of time and gradually reopened them, meaning that customers had to shop online because they couldn’t go into store. So this had a huge impact for a wide variety of retailers and is a great example of how we can use annotations to indicate there is something that will impact our data. Here are four tips for using your annotations effectively. The first is to have a concise yet descriptive title. The more descriptive it is, the better. You want to give your users as much information as possible, but an overly long title can be difficult to read, so keep it simple and short. This is because as the second tip, you can also add a description. This description can go into more detail about what happened, how it’s impacting the data, or any other relevant information that you want your users to know. Using a description can help you keep your title short, but still convey all of the important information to your analytics users. The third tip is to include tags. So similar to metrics and segments, using tags can help you keep your annotations organized and let you find and manage them quickly when updates need to be made. And the final tip for working with annotations is to pick a color scheme and stick to it. So right now, there are nine different color options that Adobe has available for annotations. And if you create and stick to a color scheme within your organization, it can help your analytics users know what kind of factor is impacting their data. So for example, you can use green for new site features launching, red for data loss, yellow for external factors, and so on. If all of your users follow the same format, it can make identifying the types of factors that are impacting the data a lot easier to notice in the dashboards. You can even take this one step further and use an external source like Confluence or a shared Excel file to keep track of which colors are being used for which type of data impacts. So to conclude, the three tips I shared today were to one, always start off strong by having your project preferences set to your most commonly used settings. This can speed up your ad hoc data pulls and make sure that you’re looking at the right information. Two, keep your segments and metrics organized so that users know which ones to use to pull their data. This can be using tags, the approve option, or adding something to the name of it to indicate that it’s the right one to use. And finally, make use of the new features that Adobe has added, such as the annotations. You can use these to keep track of internal and external factors that are going to impact your data so your users know why their numbers might look off if something has happened. Thank you all for attending today. It’s been great talking to you and I’ll be back for the Q&A after the next presentation. Thanks, Mandy. We’re really excited that you’ll be with us in just a few minutes for Q&A with our next speaker, Kea Walton. So as we’ve established, Freeform tables are a powerful tool in Analysis Workspace. And in Kea’s presentation, you’ll see what you have to do to create them, how to narrow down the data in them using filters or segments, and how to compare data across different date ranges. Remember, Kea will join us along with Mandy for Q&A right after this. So if you have any questions or comments, drop them in the chat window and address them to either Mandy or Kea, if you will. Now please welcome Kea to the Skill Exchange.

Hi there. My name is Kea Walton. And today we’re going to learn all the fun new things about Freeform tables and Adobe workspaces. Before we jump in, a few things about myself. I am the Director of Analytics for Voice of America, a US federal agency that broadcasts news around the world. I have the privilege of working with newsrooms and journalists that publish content in over 40 languages and reach millions of people. I’m also a proud alumna of the University of Maryland system, Go Terps and Retrievers. And in my free time, you can find me on the sidelines at the soccer pitch cheering for my all-time favorite player, my daughter. Today’s session is all about Freeform tables. I’ll walk you through how to create them to get the data you need for actionable insights and next steps when it comes to your digital platforms and campaigns. You will also see how you can narrow data to the things you want to focus on using filters or segments. Because trends are important, we’ll also go through how you can use tables to compare KPIs from different date ranges within a table. In my opinion, Freeform tables are one of the most powerful visualization widgets in workspaces. As you’ll see in today’s session, you can easily get so many insights from tables alone, but you can also use tables to generate helpful widgets and charts like summary, number, line charts and stack charts. The key to getting meaningful insights is simple. Understanding how the Freeform tables work and figuring out how your different variables, metrics and segments can fit into those features to get you the insights you need. So let’s start with creating a Freeform table. You can select one of the metrics and drop it at the top of the table outline. This will automatically create a time-based table for that metric. If you want to replace the metric, you can drag the new KPI over the table header until the current metric is highlighted. The word replace will also display. To add more metrics to the table, drag the metric next to the existing headers. This will also display the word add when you’ve hit the right spot. To change the table from being a time-based table to analyzing specific variables, drag the variable to replace day. You can limit the number of rows by clicking on the highlighted number next to the row. You can also browse through the different pages by clicking the navigation links next to pages. To replace the granularity of the time-based Freeform table, you can drop day, week, month, year and hour to the variable section. You can use time-based tables to assess the growth of traffic to your websites or sales of your products or services over time. This is great for identifying anomalies such as surges in traffic because something went viral or drops in sales because something went wrong. At VOA, we use time-based tables to compare growths and drops in content consumption such as article views, video plays and audio plays to the rise and fall of traffic. This allows us to understand if a surge of traffic is specifically connected to a certain news product. You can also display specific dimension variables by using the filter feature in the dimension header. You can remove the data row for unspecified or enter a keyword to narrow down your table. Unfortunately, you can only query one keyword in this method. What if I wanted to show data for Canada 2? To display specific variables that may not contain a common keyword, you can go into the dimension menu, search for variables you want and drop them into the table. For example, if I also wanted to include Canada in addition to the countries that had United in their name, this is the method I would use to create that table. You can change the order of the rows by clicking the column headers. Using this method helps me eliminate any irrelevant variables I do not need in my analysis so I can focus on the stuff that matters. This also helps me create a table with specific target countries or traffic sources that our newsrooms are interested in. Those tables allow them to understand how specific users are engaging with their content. Segments can also be used as dimensions in freeform tables. Segments are a powerful asset in workspaces because you can create a user segment based on multiple variables and metrics. Simply drag and drop the first segment into the body of the table. To add more segments, you can drop them into the header. Here, I’m trying to compare different segments based on the devices they use and what platforms they come from. Using segments can allow you to analyze different types of users based on the digital platforms they are on. If I wanted to see the kind of traffic or content consumption for users on one of our websites versus our mobile app, using those segments are a great way to make that comparison. I also lean on segments when there are minor errors in data collection or entry. For instance, if a journalist misspells their name, their byline, I can easily create a segment to include that mistake so they do not lose that data associated with that wrong byline. You can also dive deeper into tables by breaking down dimensions with other variables. If I wanted to see what social media traffic sources generate visits to our websites, I would drag referring domains to the data row to show the top referrers for that segment or dimension. Another way to do this is to right-click under dimension, select breakdown, and find referring domains. This trick helps me organize traffic sources by their referrer types so it’s easier to see what sources are effective or not. It is also a great way to group top content or products by a specific variable, such as a journalist byline or specific TV program. If I wanted to see a newsroom’s top program, especially which episode is generating the most views, this is what I would use. You can also break down metrics by variables. Drag one of the dimensions under the metric and it breaks visits down by the top five variables for that dimension. If you’re interested in breaking down by specific variables instead of the default top five, you can go into the dimensions and drag and drop specific variables. You can also use segments as variables to break down metrics.

This feature is a great way to see the increases and decreases in traffic or events from specific countries, traffic sources, and other variables over time. It is also helpful in creating pivot tables. At VOA, we use this to understand how many people in specific target markets are watching certain TV programs.

Did you know you can apply segments to the entire table? Simply drag and drop segments to the top of the panel. You can also stack segments, which means applying more than one segment at a time. You can drop the second segment next to the first one.

If you want to remove a segment, simply click on the X at the right end of the segment box.

To create a dropdown, hold the shift button while dropping your segment into the panel. This will create a null filter option and the different segments you added to that dropdown.

You can also add labels to the segments or segment dropdowns. In this case, I wanted my workspace end users to know that one dropdown is for segments based on what devices they use and the other is for the traffic source they came from. You can change the date ranges of the data in workspaces by clicking on the top right. Here you can select the specific date ranges for your data. There are also preset date ranges you can use such as this year, this month, or the last 30 days. You can even set the time range within the day. This feature is really helpful for folks who want to monitor a time-based flash sale. The rolling date feature is another one that is incredibly helpful. At VOA, we use this to create reports for specific events, for example, US election season. We set a fixed date and then have the rolling date as the end date. Date ranges are also available as elements in workspaces. I like to break down metrics by using specific date ranges. In this case, I’m breaking it down to today, yesterday, and two days ago. This is a quick way to compare a three-day trend for specific variables. In this case, I can see how certain refers have changed over a short period of time. Another way to get a time-based comparison is to right-click on the metric to select time periods. You can either select the previous time period to the current date range of the workspace, or it allows you to select specific dates. I really like using this feature because, as you can see, you can quickly get a visual on which variables have increased or decreased. To wrap things up, freeform tables are one of the most powerful tools in workspaces. You can see trends in variables or metrics over time. This can also help you immediately identify which date ranges to focus on or see if your digital efforts surrounding a marketing campaign or event are effective. As you’ve seen, you can break down variables and metrics with other variables such as the date range, the date range, and the date range. You can also break down variables and metrics with other variables, metrics, or segments to give you a better understanding of your digital audience. Tables are also the main source of data for other workspace visualizations. Your charts are only as useful as the data they are based on. Thank you so much for joining me in this session. I look forward to chatting with you in the Q&A. Great. Thank you, Kea. And thanks to Mandy and Kea for being here with us today. Guys, welcome to the Q&A section.

So, we are going to, it looks like we’re still waiting for some of the questions to come in, but we’re going to jump right in. Kea, I think this one is for you. I think we’re going to say, we’re going to have this one for you. It says, I’ve heard that using segments for things like bounce rate and exit rate can be tricky because you need to take into consideration hit level versus visitor level. Can you speak to that? Is it true? You know, what are your experiences with that? So, that’s definitely something you have to take into consideration when you’re setting up. When you’re applying segments to your workspace or to your tables and what metrics, what kind of metrics you’re looking at. So, if you’re looking at hit level or visitor level, that may not necessarily apply for bounce rates, level segments. That may not necessarily apply for bounce rates or exit rates because those are session-based or visit-based metrics. So, it’s just fundamentally looking at what type of metrics you’re putting into your workspace or your tables and then what kind of segments, what are the settings for those segments and making sure that those align. Okay, great. Thank you. Yeah, it can be tricky, you know, with the different buckets in the Segment Builder, right? You have to make sure that you’re really kind of picturing, am I asking about a page? Am I asking about, you know, a visit or a person or whatever, you know? So, thank you. Okay, awesome. So, Mandy, we have a question for you. What is the difference between setting your default report suite as most recent or specific? That’s a really great question. So, when you have your report suite set to most recent, that means that when you open up a new project, a new panel, it’s going to automatically set it to the last report suite that you used. However, if you use more than one report suite but you have one that you use most frequently, you can set that as your specific report suite. And that way, whenever you start a new project or open up a new panel, it will always open with that specific report suite that you want. And it’s always important to make sure that you’re using the right report suite because if your organization has multiple different ones, some of them might have maybe virtual report suites that have different filters applied to them. Making sure that you’re pulling the right data is important. So, by setting it to a specific one, you always know what report suite it’s going to open into.

Yeah, that’s great. I mean, it also, I think, speaks to making sure that the friendly name of the report suites is really well described, right? I mean, if it’s really vague or whatever, then people aren’t going to know what data they’re pulling from, right? So, great answer. Thank you. Okay. Kea, we’re going back to you. Let’s talk about time-based tables and stuff like that. How do you know when to break a time-based table down by day, week or month? So, one of the things I like to think about before building a table is looking at goals. What’s the purpose? What type of questions I’m trying to answer with that table? And part of that is date ranges or the breakdown. So, for example, if I’m looking at specific programs, so at my job, if I’m looking at a specific program that aired at one o’clock, I might want to look at data and break it down by hour. If I’m looking at specific events that range across days, then that’s when I have to look at daily breakdowns or if it’s across weeks, weekly breakdowns. So, it’s really fundamentally at what you’re trying to get at, what questions you’re trying to answer, and then using the different time segments that I showed in my presentation to break down your table. Yeah, thanks. I don’t know. From my experience, I’ve also noticed that sometimes stakeholders ask questions, and you can tell by their question that they really are asking something else, right? What you’re really asking is, what you really want to know is this, I think. So, you probably get, I don’t know if you guys get used to that, kind of like maybe translating the questions that stakeholders have for you and going, okay, I know what to run to answer this question. It’s a little different than how you asked it, but you have to be able to do that translation. Do you find that as well with your experiences? Absolutely. I think part of being an analyst is also being part psychic and understanding and speaking your stakeholders’ language. And so, that’s definitely aligned with what we do. Yeah, great. Okay, awesome. Mandy, I want to come back to you with this one about anomalies. How are the anomalies actually calculated? Thanks. So, the anomalies are automatically calculated based on the history of data that’s in Workspace. So, they’re not something that you’re going to create manually in your Workspace. Once you turn on the anomalies, the Adobe Analytics system is going to look at the trend of data and find anything that falls outside of the normal range and automatically mark that as an anomaly in your data. Oh, cool. Okay, great. Sometimes, I feel like it feels a little bit black boxy to people. So, thank you. Appreciate that. Okay. All right, Kea, back to you. So, if I’m starting my first Workspace, we’ve talked a lot to people today about in the beginning stages. It says, how can I build a report that would be helpful to me? Maybe just right out of the gate. Where would you start? Going back to goals and looking at the questions that you’re trying to answer. Of course, I’m a little bit biased because this is what my presentation is about. But always start with a free form table and experiment based on what kind of data you’re going to get to see if you can get your answers from there. But usually, that’s a really good one. Quick Insights panel is also another good visualization where you could start. But definitely have goals, have questions in mind. If you’re building your first Workspace for a stakeholder, no pressure. Definitely start to use those psychic powers to understand what they’re trying to get at, what they want from the data, and then use that to build your Workspace. Yeah, I love that. Hard psychic. Yeah, that’s awesome. Okay, Mandy, coming back to you with the question here. Can you edit components that other people have created? So, yeah, this is something that’s really important to keep in mind. Components, unlike overall workspaces, typically can only be edited by the person that creates them. So when you make an entire dashboard, you can share that with other people and you can set up it so that way other people can edit your dashboards. You don’t really have the same ability to do that with components. You can edit components that you’ve created. And that’s basically it. The only exception to that is if you have admin privileges. Anybody that has admin rights can go in and edit components made by other people. But other than that, you can only edit components that have been made by yourself. It makes a lot of sense. I mean, you can imagine creating some components like that and then going in and somebody else has changed them. So that could be a problem. So, yeah. Thank you. Great. Great. Okay, Kea, here’s one for you. Can you just download just a table? We’ve talked a little bit about downloading and stuff like that and things like that. Can you download just the table itself? Absolutely. This is a great question. So if you just want data from a specific table within a workspace, you can right click on that table. And one of the options, one of the right click options will be just to download that table as a CSV. And you should be able to download that and import into Excel and integrate it with other forms of data. Great. Yeah, that’s awesome. Always nice to be able to take out that chunk that you need and not have to have a bunch of other stuff in there as well. Okay, awesome. Well, thank you guys. Really appreciate you coming in today and helping us and giving these presentations and kind of teaching us what’s going on in the real world. We’re always excited here at Adobe to share how things should work and how we think you should use them and those kinds of things. But what really matters is what’s happening to you guys out there where the rubber meets the road. So thank you for your presentations and the Q&A time here. And we appreciate your time. Thank you so much for having us today and looking forward to speaking to you all again, hopefully. Hope this was a helpful session and thank you for having me.