Adobe Analytics Tips & Tricks

Transcript

Hey everybody, this is Jen Lasser. Thanks for joining us today for our Analytics Tips and Tricks session as part of Experience Maker’s Skill Exchange. Today we’re going to be going through all the greatest analytics tips and tricks that we’ve shared throughout the year for improving your workflows in Adobe Analytics. For those that haven’t met me before, my name is Jen Lasser and I’m a Principal Product Manager on the Adobe Analytics team. I’m going to be talking about the great things that we’ve done I’ve been with Adobe for about eight and a half years. I started in the consulting organization working with many customers like yourselves solving different analytics problems before I moved over to the product management team. I’ve traversed the country in my role at Adobe. I started in Chicago, moved to New York, went over to Salt Lake City for a little bit, and now I’m in lovely New Jersey. It’s super rainy out here today. During 2020, I’ve acquired a lot of new hobbies in addition to just having a little bit of a chat with my team and that’s in addition to just loving using Analysis Workspace and building features into it. I’ve been taking up baking, tons of puzzles, gardening, all of the things that you could do while stuck at your house 24-7. All right, so what are we going to be talking about today? We’re going to be sharing lots of tips to improve your workflow in the areas of analysis, segmentation, attribution, visualization, and sharing. I’m really excited to be talking with all of you that these are some of the common steps that you take to get an insight from Adobe Analytics and share it with your organization. We’re hoping by covering each of these areas that you’re all able to walk away with a few tips and tricks to improve your everyday work within Adobe Analytics. First step on our workflow journey today is answering business questions in Adobe Analytics and that always starts with analysis. You start your analysis journey by bringing in freeform tables, building out your business questions, and finding that insight that’s worth diving into further. I’m going to highlight some tips and tricks to get your analysis started by building tables as well as use some of the new panels that we’ve built in to help render some rich analysis for you. The first tip is to build tables automatically. You don’t need to know where to put dimensions, where to put metrics, where to put segments. As long as you select what you need in the left rail and drag it into the middle project area, we will render a table automatically for you. This is a great tip to highlight for any new users at your organization. They don’t have to know what to put where in the rows and the columns. As long as they drag it in, Adobe will go ahead and pre-build that table. This is something that we’ve seen be super successful. We see a lot of tables being built this way. Definitely take advantage of this if you haven’t already. Additionally, you can also use the Freeform Table Builder if you want to set up and customize your table before building it. This is an additional way to quickly get to insights within Freeform tables. Essentially what Table Builder does is it pauses the table so that you can get it set up exactly how you’d like. You can bring in your dimensions as rows or columns. You can also bring over your metrics and segments as well. In this view, you can get your settings adjusted, such as how many rows you want to show for each breakdown level and how many breakdowns you want to apply as well. These are things that haven’t been possible to date with the typical Freeform table that renders on the fly. Once you have all of your settings set up, you can then click Preview to see exactly how the table’s going to look before it is run and returned to you. This is a great opportunity to make sure you have it set up in the way that’s going to be easy to analyze. You can also monitor your progress through the status bar there up at the top so you know exactly how long the table’s going to take to run. Once the table’s rendered, it will be completely interactive, just like any other Freeform table. You can go ahead and drag and drop from the left rail and continue to dig into your business question. With the Freeform Table Builder, there’s some sneaky additional tips buried within it. You can also rearrange your static rows in any order that you’d like. We’ve seen this be really helpful when you want to have metrics or segments as rows, and you want them in a specific order to tell a certain story to your users. Use that Table Builder to manually rearrange the order of your rows when needed. We’ve added a lot of new panels to Analysis Workspace this year. Panels are prepackaged analyses built around a set of use cases that you have. We’ve added a Quick Insights panel, a panel for Adobe target users called our Analytics for Target panel, and also a Media Concurrent Viewers panel for our media customers. Let’s take a look at what each of these new panels brings to Analysis Workspace. Before I dive into any individual panel, I wanted to mention that panels can now be found directly in the middle area in a blank panel. So not only do we have the visualizations listed there, we also have the panels available as well. And this is in addition to where they’ve always lived, which is the left rail. Now let’s dive into the first new panel, and that’s the Quick Insights panel. This can be used to help answer business questions easily and on the fly. I like to think of this as essentially like a Mad Libs builder, where you go and you pick your noun, your dimension, you pick your metric, you pick your segment. And as you’re making these selections, the table and the visualization will render on the fly. So you don’t have to build from a table first and then add the vis. It’s just doing it all for you at once. You can build out your business question, and then you can interact with the builder even further if you want to compare or contrast segments or add additional metrics. In addition, you can go down and interact with the freeform table. Once it’s rendered, you can bring in additional things from the left rail if you’d like. The visualization also has a vis-type selector kind of up-leveled from the vis settings. So you can change from a bar to a line or an area chart on the fly if you’d like. Now, the second panel I wanted to highlight is the A4T, or Analytics for Target panel, which is useful for our Adobe target customers. The A4T panel allows you to analyze your target activities and their experiences to pick a winner for any test that you’re running. In addition to having some helpful information like activity impressions and activity conversions, we’ve also now built in lift and confidence into this panel. This is something that previously was only available in reports and analytics, and by having it here in Workspace, you can have a more seamless workflow when you’re analyzing your tests. Another advantage of this panel is that you can analyze up to three different success metrics and bring in lift and confidence for those as well, something that you haven’t been able to do in Adobe Analytics to date. In addition to the table, we’ve built in a conversion rate trend as well, so you can look at conversion rate by each experience over the course of your date range and your test runtime. In addition to the panel, I wanted to touch on just a couple backend updates we’ve made. We’ve also taken the steps to add support for auto-allocate and auto-target activity types as well. So now you can analyze your sensei-based activities right here in Analysis Workspace. That just means even more data to analyze within this panel. The last panel I wanted to highlight is one for our media customers, and this is the Media Concurrent Viewers panel. This gives you a way to understand where peak concurrency occurs and where drop-offs happen to provide valuable insight into the quality of your media content and your viewer engagement. Like with our other panels, you start from a builder state where you can bring over many different dimensions to analyze. You can also extend this to date ranges or segments as well. You choose the granularity that you want to look at concurrent viewers across, and the panel will then render. In the rendered panel, you can interact with these new views just like any other project today. So you can turn the legends on and off. You can hover over the graph here to see the min or the max concurrent viewers. And you can even build in additional tables to this panel as well if you want to dive into a specific use case even further. Now, something that all three of these panels exhibit that hasn’t been present in Workspace before is the ability to not use the left rail. So we’ve actually built a new component. We’ve built a drop-down where you can pick from a list of compatible dimensions or metrics, and you can use that to build out these panels instead of the left rail. If you prefer to use the left rail, though, to build, you absolutely can. Drag and drop from the left side into these panels. This new drop-down component is something we want to carry out through the rest of the UI. So be sure to keep an eye out for this, and you’ll see it pop up in more places in the future. The goal of this new component is really to reduce the learning curve of Analysis Workspace so that any user can come in here and start diving into different analyses that they need without having to have the big barrier of understanding all of the components available in the left rail. So hope you guys are as excited about that update as we are. Now, we know that Analysis Workspace solves a ton of your analysis use cases, but sometimes in your workflow it’s necessary to take data out of the UI for further analysis or sharing to other systems. In Analysis Workspace, we offer a few different ways that you can get your data out. The first that I wanted to highlight is the ability to now download up to 50,000 rows of data. This has been a long time asked, and we’re super excited that it’s in the product now. In the UI, you’re able to see up to 400 rows, but if you want to see all the rows available for a particular dimension that you’re analyzing, you can right-click and download items as CSV. Here we’ve downloaded the 2,436 pages that were available for the page analysis we were conducting. Now, a great thing about this option is that it will also apply any breakdowns above that dimension as filters, and it will carry over any segments that you have as well. This is a huge step forward for data export because this type of export is fully compatible with everything you see in Analysis Workspace. You can get all your calc metrics out, you can get applied attribution out, there’s no segment limitations. So this is definitely a good step forward for getting compatible data out of Analysis Workspace. Now, another great way to export data if the UI falls a little short of your expectations is to use the 2.0 API. And one way you can do this is by learning how to build API requests using the Workspace Debugger. The Workspace Debugger can be enabled under the Help menu, and once enabled, it will record all of the queries that are being run to populate the UI. Now, the great thing about this is it records each individual query in the exact format that you would need to write the same API call to get the same data. Like the 50,000 row export, there is full compatibility with what you see in Workspace. You can see the allocation models that are applied to metrics. You have your breakdowns available, you have calculated metrics available. So if you haven’t used the API before, I highly recommend enabling the debugger under our Help menu and just starting to learn how the different things that you see in the UI are built from an API perspective. For those that are extra interested in using this debugger, there’s a couple other cool things that you can see. When you first turn it on and pop it open for any particular visualization, you’ll actually see the different types of calls that need to be made. For example, for free form, you have to make calls for the data itself, for the time comparisons, for the sparklines, and it buckets those for you. So it’s very easy to see what it takes to build that table. Additionally, at the top of the screen you see here, we have the request and response times for each individual query. So you know exactly how long each query takes and contributes to the performance within Analysis Workspace. So the next stop in the workflow journey that we’re on is segmentation. Segmentation is one of the most powerful things that we offer in Adobe Analytics. It gives you the ability to slice and dice your data down to the visitor, visit, and hit level and find really unique audiences to analyze further and find insights for. So if you’re not applying segmentation today, we’re going to hit on some tips and easy ways that you can work segmentation into your workflows, as well as a few more advanced ways and types of segments that you can build to take your analysis even further. Now, a common question we get is if you can create temporary segments on the fly in Workspace, ones that are not saved to your left rail that don’t clutter up what you have available there. And the answer is absolutely yes. The easiest way to layer in segments to your workflow is to use the panel drop zone at the top. You can drag and drop any set of components over from the left rail and drop them up top. So here we’ve dragged over mobile phone and tablet and created a segment on the fly to analyze our data by. So this is a really easy, quick way to start adding segmentation into your workflow. You can remove that segment by clicking the X and that segment will be gone out of your analysis. So it’s also a very temporary way to slice and dice as you’re having more business questions come to mind when you’re analyzing. You can even take this further by leveraging dropdowns. Dropdowns we released a couple of years ago, but they are by far one of the most used things in our product. If you drag over many items and hold shift instead of creating a segment, you’ll create a dropdown filter. And a dropdown filter is a way to slice and dice the data across many different items that you’ve brought in. So here we’ve created a device type dropdown, a marketing channel dropdown. And as you can see, you can have many dropdowns working together at once. They just they just function as an and criteria. So if you’re looking to slice and dice your data on the fly, definitely build in dropdown filters. These also are a great thing to build in if you’re going to share a project out to end users and you want them to be able to dig in a little bit further. This gives them a way to interact safely within the project so they don’t even need to learn how to use that left rail. Now, a couple of segments that I wanted to highlight and kind of get you thinking about are things that we added earlier this year. So one is the criteria of equals any of. If you’ve ever had a need to create a segment for a list of values, equals any of is something you should absolutely be using. It is more performant than our contains or contains any of criteria. And it’s also more specific. So if you’re looking for a specific set of device types or marketing channels in this use case here, you can add in equals any of into a single criteria with a comma delimiter. So further simplifying what you saw on the screen here, where we had four different channels listed out across four different criteria. Now you can have a nice, simple approach to building this segment. Another thing that we added to segmentation that has been super valuable for customers that we’ve worked with is the distinct count operators. Distinct count allows you to segment on a number of unique items within any dimension. A couple examples that I wanted to share to just get you thinking is first, the number of products within any particular hit. So if you’re looking to understand orders where multiple products were purchased to maybe find a segment of more loyal visitors or more engaged visitors, you can create a distinct count segment based on product. Now, what you’re seeing on the screen here being demonstrated is a good best practice. Any time you build a segment, you want to make sure that you verify the segment is working and returning the expected results. And you can very easily do that by creating different variations of this distinct count product segment. Now, a second example I wanted to give for distinct count is you can also count the number of marketing channels that any visitor has. You’ll change your container to visitor, bring over the marketing channel dimension, use a greater than distinct count operator, and then choose how many marketing channels you want the segment to have. So here we’re creating a segment for visitors that experienced more than two marketing channels. This is a great way to understand those visitors that are interacting with many marketing touch points. And you can further dive into these visitors using the flow visualization and just kind of understand, well, what is the sequence and the order of those marketing touch points and how is it influencing their behavior on your website or your app? So the next stop in our workflow journey is attribution. And when I’m talking about attribution, it’s kind of twofold. One, it’s looking at the full picture of data. The fact that you have data from the website and the app, bringing that together to ensure you’re looking at that full digital data picture. And the other side of attribution is being able to analyze all of that data and the various touch points customers have with your brand so you can understand the impact of each touch point on your key success metric. So let’s take a look at some attribution tips and tricks. The very first one I wanted to share is make sure you’re analyzing the full set of data in your project. And you can do this by leveraging multiple report suites. This is probably the number one thing that was asked for when workspace was built. And we’re super excited that we were able to add it this year. The report suite picker is now a panel level object. So each panel can reference a different report suite. For example, here we’re bringing our web and our app data into a single project. Now, the data is not combined, but it is available in these separate panels. If you want to apply a single suite, you can always do that by right clicking and applying that report suite to all panels at once. Now, the blue outline around the panel indicates the active panel in the project. And that report suite is what will dictate the components available in the left rail. So always keep an eye on that. You also have the report suite at the top of the left rail listed there so you know exactly what those components are based on. Now, when looking at the full data picture, you can do this with panels that are top down, or you can shrink those and look at the panels side by side. So always remember that the canvas in workspace is completely customizable. Both of the visualizations within a panel, but also the panels themselves. So you can get a very clear view if you’re trying to compare and contrast data sets across different suites. Now, I mentioned multiple report suites is panel specific. So within analytics, you do not have the ability today to combine web and app data on the fly in reporting. There are a couple options if that is what you’re looking for, though. You can do that through implementation by implementing multi-suite tagging. Or in customer journey analytics, which is the new cross-channel product that we’ve been bringing to market, you can add your report suites together there into a single view. In customer journey analytics, we call those data views. And by far and away, one of the biggest use cases we’ve seen so far for customer journey analytics is the ability to combine that cross-channel data or that cross-report suite data into a single view. Once combined, you have all the power of analysis workspace that you’ve come to know and love to analyze those cross-channel touch points. So if that’s interesting to you guys and that’s something that you’re looking for for your business, I’ll share some resources on customer journey analytics at the end. Now, the second side of attribution that I wanted to talk about is attribution modeling itself. Attribution modeling is built directly into analysis workspace and is available from any metric row that supports attribution. So you can modify the attribution model under the column settings, choose from a set of 10 plus models and choose your look back window as well. Look back window can look back over the full visitor lifespan, the visit or the session, or you can set a custom look back window to look back over a various set of days. Now, this is available for any package that has attribution IQ, which is all current packages. If your package does not have attribution IQ, you still do have some of these capabilities available. You’ll have first, last, linear and participation models available to you in the calculated metric builder. So definitely take advantage of modifying your attribution model on the fly to get a different lens and view on the data you’re looking at. We made a couple improvements to attribution IQ this year as well that I wanted to highlight. First, we’ve added a custom look back window. We added some out of the box options of 14, 30, 60 and 90 days. Or you can set a completely custom window if you want. So if you want a 22 day look back window, you can absolutely do that. This custom look back window can be combined with any of the models. So here we’ve shown participation. You could do this for U-shaped or first or last touch as well. And the advantage of a custom look back window is that it looks at the conversion point and then goes backwards from there. So if you had previously been using a visitor or visit look back window and experienced quite a bit of none values returning, the tip here is to start leveraging these custom look back windows because of the way they position themselves off of the conversion. You’re going to see attribution to a lot more relevant data points and rows in your table. And we cover that quite a bit more in our documentation if you want to dive in after this session. Now, another thing that we added for our ultimate package is the ability to do algorithmic modeling. Algorithmic modeling is not rules based like our other models. It’s data science based. So it looks across all the different touch points, leverages our algorithmic modeling and assigns out credit to the various touch points based on what that model yields, rather than a rules based model where there’s a specific weight that is assigned to the starter player and closer touch points. Now, this is something that was only previously available in Data Workbench. So really thrilled to be able to bring this to Analysis Workspace and close one of those gaps that we’ve had with some of our legacy products. Now, I mentioned earlier where customer journey analytics comes in and how it becomes even stronger with multiple report suites available. Same goes for attribution IQ. Attribution IQ and analytics today looks at all the digital touch points that you have coming into Adobe Analytics. Customer journey analytics takes that idea even further by allowing attribution IQ to work across all of the cross channel touch points that you’ve brought together. So if you have a view where you have digital married together with point of sale and maybe call center, attribution IQ will continue to work across all of those touch points and find the different weights that it should assign to those various touch points. The same goes for segmentation as well as our journey visualizations like fallout and flow. So the next step on our journey here is a visualization. Now, anytime we do a tips and tricks session, we always want to share visualization tips. We believe it’s very important to bring your data to life and share visualizations that will resonate with your users and your recipients of your projects. Don’t just share a data table to them and expect them to sift through the data and look for the insight. If you can add a visualization on top to really help tell that story. So let’s take a look at some ways that you can bring your projects to life by adding in visuals. The very first thing to think about is actually not graphs or charts or visuals at all. It’s your your tables themselves. There’s ways that you can use the column settings and tables to really simplify what you’re sharing to your recipients. So here we remove the background of our page velocity column. We also remove the percentage because that wasn’t relevant for this particular metric. And you can customize each column just like we did here. You can also adjust your column settings and the totals that are shown in those columns. So we now have two different totals, both the table total and the grand total or report suite total available. But sometimes it doesn’t make sense to show both of those or any totals at all. So you can always go to your column settings and turn off one total or all of them at once. Really just whatever makes the most sense for the story that you’re trying to tell. Additionally, conditional formatting is a really great way to convey a data story without having to even share numbers at all. For example, if you bring in any sort of time comparisons like year over year or month over month, as you see here, you can actually turn numbers and percentages off completely and turn on conditional formatting. What conditional formatting does is it overlays a heat map onto the column based on the underlying data. So without even sharing any numbers, you can convey to your project recipients if a number is good or bad compared to those prior ranges. So you can see here row 15, the stock value month over month is doing really well. We’ve got a bright green as opposed to row three individual bond, which has a red value, meaning the month over month isn’t as strong. So when possible, remove the numbers from the table and tell your story through visual color instead. Always look for opportunities to add visualizations and maybe even remove the data table altogether, carrying that idea that we just shared of removing numbers. One cool way we’ve seen this used is with cohort tables. Instead of showing the whole cohort table itself, maybe just add a line visualization and plot the average retention or churn line instead. So again, anytime you can reduce that kind of cognitive load on your recipients, you want to try to do that. And bringing in a visual and removing the table altogether is a great way to do that. And in case you missed it, the way that we hid that table was clicking the dot up where it says line. That’s where you manage the data sources and you can show and hide the source tables from there as well. So if you wanted to add that table back, you just go ahead and click that dot and you can say show data source and it will come back to the project. Now, when it comes to visualizations themselves, we’ve added quite a few helpful settings this year to help you kind of tell that so what story. In particular, with line visualizations, I want to highlight a few new settings. First is the ability to show min and max labels so that they’re always present on the line chart. The next is the ability to overlay a trend line. So if you have a visualization that’s kind of going up and down, you can overlay a trend line and see the exact trend that that data point is having. One of the last things is the ability to show or hide the X and Y axis. So if you really want to create a nice simplified visualization, you can turn off the axes, turn off the legend if you’d like, and turn on those min max labels. If you take a look kind of a minute ago in this video, you’ll see where we started and now you can see where we’ve ended. A nice simplified line visualization that’s telling us the story of this particular trend. It’s slightly trending up. We can see that with the trend line. Our max was 606,000. We can see that with the label. So a great way to make a visual really stand on its own and highlight some key data points to the recipients of your project. We mentioned trend lines. We’ve also added moving average to the trend line. So not only can you pick from some regression options, you can also add in a custom moving average as well. You’ll see here we’ve interacted with our periods and overlaid a 14 period moving trend line. Now I say periods because these trend lines will flex to whatever granularity you have for your chart here. So we’re looking at different hours of the day. So this is a 14 hour moving average trend line. If we change this chart to day or week, then it would be 14 day or 14 week. So really flexible ability to add in moving averages. This is something you previously would have had to go create a calculated metric for and you would need to create a metric for each type of average you wanted to do. Now it’s all on the fly. Easy to overlay right in your line visualization. So particularly for trends like this where you have a lot of oscillation up and down, highly recommend adding in a trend line to kind of put that data in perspective. Now one last tip I wanted to highlight is how to better organize your analysis strains of thought in a project. You can definitely organize by having different panels of data. But one thing that I’ve been doing a lot of in panels that I’ve been building is bringing in just a blank text box, collapsing it and giving it a title. This is a way to kind of use text boxes as like section dividers. So if you have a panel with lots of information in it and you want to organize in a bit more, you can bring these over and divide up your work a little bit. They’re a great place to also just leave notes as you’re building out your panel as well. So not only have I been organizing my house during 2020, I’ve also been organizing my workspace projects and hope this is a great thing that you guys take away as well to start organizing projects at your business. Now, anytime you’re building out analysis, you’re finding insights. You don’t want that insight to just stay with you. You want to share it to your organization so that other people can learn from your analysis and start making data-driven decisions on their own as well. So with this last workflow step, we wanted to talk about some tips and tricks for sharing your workspace projects and sharing Adobe Analytics data out to your organization and the different mediums you have to do that. Before you start to share your project, you want to think about the recipients that you plan to share it with. What are their goals with the data that you’re sending them? What questions do they need to answer and how do they prefer to consume information? When we talk about analysis workspace, we tend to kind of lean into these three types of personas. Executives, analysts, and novices. Analysts are definitely where we see workspace shining the most. Analysts get in, they slice and dice the data, no problem finding the data they need. But they need to ensure that the insights they provide are researched completely before they share out to the organization. Novices and executives may come into the UI, but they may prefer to receive information a bit more instead, either as a PDF or a CSV or for executives in particular, maybe at the palm of their hand. So think about who you want to share your data with before you really even get into workspace and let that guide how you build your project and how you decide to ultimately share those insights out. So let’s talk about the various sharing options you have with analysis workspace today. First thing I want to hit on is under the share menu, the ability to share the project you’ve built. Now, this year we have expanded these sharing options quite a bit and we expanded them into three different roles. So before you could share a project to just a set of recipients, and now you can put those recipients into edit, duplicate and view roles. Edit is useful if you want to co-manage a project with your colleagues. Duplicate is useful if you want to share a project to users at your organization that may understand data and how to use workspace, but you really don’t want them altering your project in the end. And view is a role that is useful if you want to share a project to users that may be less familiar with data, with analysis workspace or just with Adobe Analytics generally. But you still want them to be able to come in and consume data and insights and interact in a very controlled way. So let’s take a quick look at what each of these experiences looks like. Now, the edit experience is just like as if you owned a project. So you can save over the project, you can modify the share recipients, and you can fully interact with everything you see in the project. So if you don’t own the project, but you’ve been made an editor, it’s as if you’re a co-owner with the person that created it in the first place. This has been extremely helpful for non-admins who have never been able to co-edit with their colleagues in the past. Now you’re able to collaborate with anyone at your organization. Now, the duplicate experience is a little bit more restricted. So the duplicators cannot save, they cannot modify the share recipients for a project, but they can otherwise fully interact in the project. The main difference here between them and editors is that duplicators can only save as to create their own kind of spin-offs of the project that was shared to them. They can never save their changes back to the original copy. So again, this is a great role to put users in that you feel confident making decisions off of data in Analysis Workspace, but you may not want them to influence the original projects that you created and shared with your organization. The last role is the View role, and it has the most visually different view of Workspace. People in the View role can’t save, save as, or share. They have very limited menu options. You’ll see it’s been simplified by a lot. They don’t have the left rail. It’s not even hidden off screen. It’s just not available at all. The main thing viewers can do is interact with the project through the drop-down filters that the editors have put into the project to begin with. And they can also hover over different parts of the tables and visualizations to get more information. But by and large, it is a view-only project where there’s only some controlled interactions that were curated by the editor available to these users. Now, deciding on which role to assign is a tough decision. I mean, you can always go and change it. It’s not a permanent thing. But here’s kind of how we think about it. The Edit role is great for those analysts. Maybe like if you have Level 1, Level 2, it’s like your Level 2 analyst. Those that are super savvy with the Adobe Analytics data structure, they know what’s in EVAR, what’s a prop, how to use those together with events. They know how to use Analysis Workspace. So they can completely co-manage a project with you safely. The Duplicate role is great for maybe those Level 1 analysts. They’re just getting their feet wet with Analysis Workspace. They generally understand your data. And you feel good about having them work within the project. But not quite saving to the project yet. And then lastly, the View role. This is something we’ve seen be really successful for executives and novices. They’re newer to Analysis Workspace, certainly newer to your Adobe Analytics data structure. But you want them to be able to get in there and interact a little bit. The View role is definitely a step up from a static PDF. But it’s not quite full-blown Analysis Workspace, which comes with its own different learning curves. So if you’re looking for ways to decide who gets what, this is a good rubric to start from. So each of the roles that are assigned can be viewed by that user in the landing page or the project manager. They can go see what role they have for each project. And you can have different roles. This isn’t a user-level setting. This is a project-by-project setting. So myself, even though I do know Analysis Workspace and probably could edit any project, sometimes I’m put in a View or a Duplicate role. And that’s completely fine. It gives me a more simplified way to work in that project. Now we highly recommend almost graduating your users up these roles. So if you start novices or executives in the Can View role, you can always move them up to Duplicate or Edit as they become more trained in using Analysis Workspace and your data. You can move groups up and you can also move individuals up as well. So there’s a couple different ways you can manage the recipients in your project. A great thing we also added that wasn’t available previously is that non-admins can also now share to groups that they are a member of. So that’s no longer an admin-only function, which really helps to make those non-admins have capabilities that they need to be co-editors of projects. One way that we encourage you guys to help train your users is show them the training tutorial. This is really going to help them become savvier in Analysis Workspace and become able to work on their own and answer their own questions. And then give you the confidence to move them up the roles to View, Duplicate, and Editor over time. The training tutorial is available from the new project modal. Once opened up, it’s a very detailed walkthrough of the Analysis Workspace UI, different terminology that we use, and then just how to construct business questions. So we start with step one, bringing over dimensions, step two, bringing over metrics, and we walk them through how to create their first analysis in Workspace. And it really sets a good foundation that they can then build on from there. Now for the executive persona in particular, but certainly any recipient, we also offer a mobile app. So it may not be realistic to think that everyone’s going to come into the UI. So we wanted to bring the data to where folks are, and that’s a phone. Everyone has a phone. Everyone has different apps. So we now offer an Adobe Analytics mobile app. Now, instead of creating a browser-based project, you choose creating a mobile scorecard instead at the beginning of your workflow. This is why it’s very important to think about who you’re going to share data to and how you want to share it before you get started. So you’ll see here we’re bringing in our different KPIs. Each KPI can also be broken down through our dimension drill-ins, and you can also layer on different filters as well. As the analyst here, you’re really building out this view that you want the mobile app user to see in the end. Now, what does this look like? The mobile app can be launched on your phone, and it uses biometric authentication, so Face ID. Once you open it up, you have the option to pick from the different mobile scorecards. And in the mobile scorecards, you’ll see all the KPIs that that analyst built in for you. You can modify the date range if you need to. You can apply the baked-in filters like mobile or tablet customers. And finally, you can click any of the metrics you see here to expose the different drill-ins that were prepared for you. So we’ve seen a really good adoption of the mobile app, especially across our executive persona. But if you haven’t created a mobile scorecard for your business yet, we highly recommend doing it. It’s a great way to ensure that anyone that needs data has it anytime, anywhere that they need it. So we’ve reached the end of our tips and tricks workflow journey that we’ve been on together today. I super appreciate the hour that you guys have spent with me today, and I hope everyone was able to take away at least one or two tips and tricks to improve your workflow. Now, to recap, we covered the workflow areas of analysis, building your tables in the most effective way that you can, using some of the new panels that we’ve put into Analysis Workspace to tackle some core use cases that you have. Segmentation, you know, layering in segments on the fly from our panel drop zone, as well as building some new creative segments, leveraging some interesting criteria we’ve built in. Attribution, both, you know, bringing together the full data picture, as well as analyzing the different touch points customers have with your brand. Visualizing data, so bringing it to life, not just sharing a bunch of numbers to your end recipients, really trying to tell a more visual story. And finally, sharing out the analysis and insights that you’ve got through Analysis Workspace, both as, you know, projects that can be delivered, you know, in PDFs and CSVs, bringing users into the UI and bringing data out to the mobile app so it’s accessible anytime, anywhere they need it. Now, I wanted to leave you with a few additional resources. In addition to the recording that you’re going to get of this session, we have a lot of valuable Spark pages and videos and various different resources available to you. The first is the Adobe Analytics YouTube channel. So we put up new videos every month covering new tips and tricks, as well as new innovations. So be sure to subscribe and listen for new videos all the time. We also have our Adobe Analytics release Spark page at adobe.ly slash aareleases. If you’re not following along with this page, definitely check it out. We have the last 12 months of innovations always published, so you’re not missing anything that’s being added to the product. We also layer in the documentation and the videos that we have for each of these individual features. If you’re seeing our releases come out and you’re hoping for something to be added to Analysis Workspace, like, man, I wish they put this into the product or that into the product, let us know. Go to adobe.ly slash aaforum. This is our experience league ideas area. You can add in your ideas, you can add in your use cases and have a conversation with product folks like myself and try to influence what we’re doing in the Adobe Analytics roadmap. I mentioned customer journey analytics before. If you’re interested in learning more about how we’re evolving Analysis Workspace to sit on top of the Adobe Experience platform and satisfy cross channel use cases, not just digital use cases like we have previously, check this out at adobe.ly slash aacja. All right, that wraps up all the tips and tricks that I had to share with you guys. I really appreciate all the time that you spent here with me today. I hope everyone was able to learn one or two new things that help make your workflow more efficient, more effective in Analysis Workspace and Adobe Analytics generally. With that, I think we’re going to open it up for some Q&A. So looking forward to hearing from you guys. Thanks. Thank you, Jen. Those are some great tips and tricks. Now, I’ve been watching and I see a lot of questions coming in on the chat. So let’s bring on a live expert to answer your questions. I want to bring to our show Solutions Consultant, Joe Morris from Sydney, Australia, to answer them for you. Joe, please join me on this virtual stage.

Thanks, Badshah. And thanks to Jen for what is an awesome session. I think I’ve even picked up lots of tips and tricks from it. So it’s a really great way to look at using the tool. So obviously, as Badshah said, I’m Joe Morris, a Solutions Consultant here in a very rainy and overcast Sydney today, which is very different to what it was yesterday. But I’ve been a Solutions Consultant here at Adobe for a little while now. And prior to that, I was client side on within Sydney here as well as back in London. So I’ve been working on Adobe Analytics for quite some time now. I’m here today to talk to you, talk through some of the questions that have gone through within the session. And hopefully be able to provide some value to them. And I’m just going to jump straight in. So I suppose one of the questions here with the most votes is, is Adobe Customer Journey able to connect from people who come from our ads, Facebook or Google, directly to my company mobile apps? So for this one, I’m assuming you’re talking about Customer Journey Analytics, which Jen mentioned. So Customer Journey Analytics sits on top of our Adobe Experience platform. And you can find lots more information on that on the Adobe Experience League. But the Adobe Experience platform basically enables the functionality to bring lots of different data from lots of different sources and standardize them in what we call our Experience Data Model. And Customer Journey Analytics sits on top of that. So when it comes to your Facebook and Google advertising data, you can take ad log data from these platforms and bring them back into the Adobe Experience platform for you to be able to analyze end to end customer journey within Customer Journey Analytics. But also if we think about how someone goes from an ad to your website or your app and then interacts with those touch points, that data also comes into Adobe Analytics. So when that Adobe Analytics data comes back from your web and app engagement, that comes back into Adobe Experience platform as well for you to leverage within Customer Journey Analytics. So long story short, yes, it can definitely be shown within Customer Journey Analytics. Question two, can a fallout look across sessions? So with regards to a fallout chart, it has the functionality to be able to look at both a visitor and a visit level. So if we look at a visitor level, that enables you to look at subsequent sessions where the, I suppose, the user has the same Adobe Experience Cloud ID. So hypothetically, I might have been on the website today, but I also might have been on the website three days ago. That, if you looked at it from a visitor level, would look at me as a visitor and look at my subsequent visit. If you wanted to look at from a visit perspective, you could also do that, but that only looks at the isolated visit. So you can look at both and that can all be done within the settings of the fallout chart. The next one is, can we also assign edit and view roles for segments and calculated metrics, please? So for this one, unfortunately, you can’t do specific view rights for segments and calculated metrics. What you can do, though, is when you create them, you can create them, but not share them with any other assigned group or user group or, I suppose, or user. So when that comes to a calculated metric or a segment, you can create one in isolation and simply not share that out to any other user group. So when it comes to the individual wanting to use those calculated metrics, you can do that within your workspaces, but it’s isolated to you as a user rather than being a user group or wider for the entire business to use. So rather than being a view permission as such, similar to the work analytics workspace, this is more to do with how you actually share those segments out.

Will these sessions be available after the training? Yes, they will be recorded. There has been recording after the session will be available. The table builder has limitations of breakdown. Can we expect to an increased view? So with regards to the table builder itself, it does provide the opportunity for you to actually show what that view is. So when you say enable table builder on the call to action, it basically stops creating that table as Jen showed. And then by doing that, you can essentially decide what that view or how many rows that you want to be able to see. With regards to an increased view on top of that, that’s probably something that we need to speak to the product guys about, but you can add tickets into the Adobe Analytics section within the experience league to see if anything can be done from a roadmap perspective. But yeah, definitely the table builder does enable you to show the multiple rows. There is also the ability to export out that data, which Jen showed, that 50,000 rows. That’s also a good way of looking at it if potentially you wanted to look at the big set of data as well.

So the question here is how to exclude non-required hits and visits, not matching the condition from the visitor level segment and get only the required data. So for this one, there’s actually two ways of doing this. And it depends on the use case. So you can either look at, I suppose, excluding the data before it even gets to Adobe Analytics. But for this one, I get the sense that it’s more about excluding the data from within your specific workspace once it’s actually in Analytics. So to do that, there’s a few different ways you can exclude it. So you can exclude by IP. You can exclude by what we call bots. You can exclude by Vista. And this can, so Vista, you’d have to speak to your, I suppose, your CSM or account executive to work through that one. But with regards to excluding by IP and excluding by bots, you can do that within the admin settings within Adobe Analytics. So you can definitely exclude specific hits to ensure that your analysis and your workspace is tarnished with any unrequired data.

So what is better to use to track campaign variables? Should we use EVAR or PROP? We don’t want the value to be overridden since we have many campaigns. That’s a really good question. And to be honest, it’s kind of use case dependent, which is always not a very good answer to give and normally what people don’t want to hear. I suppose when it comes to thinking about which one you actually want to use from tracking your campaign, you ultimately want to decide what the end result you want to see. What data do you want to see and what do you want to get out of that from an insight perspective? From my experience from client side, I always tended to go down the EVAR route, but that was always use case specific and I suppose a little bit of bias on my side of things. But it’s definitely use case specific, but all I would say is think about ultimately what you want to get out as the end result and what data you want to be able to get out of it.

So I found that segments are not working properly with metrics like page velocity. Is this a bug? So anything to do with bugs can be found within normally within the experience league or normally within the forums. Me personally, I haven’t heard of any bugs that are in there to affect any segments not working properly with metrics. With page velocity, what I would say though is probably reach out to customer support and they’ll definitely be able to have a look into that for you and see if there’s anything that can be done.

I’m just going to sort see if there’s anything new coming in. Hi, we’re dealing with a lot of implementation of mobile app analytics and we’ve been facing issues with verification of those implementations, especially with Android OS while using traffic sniffers. On this one, I suppose where I would come to it is I would definitely recommend using the Adobe debugger within Adobe Analytics. That will certainly help show you how the implementation is working and if there’s anything that is showing up from a red flag perspective. Obviously, without knowing the specific difficulties that you’re seeing, it’s obviously difficult to answer, but I would definitely recommend the Adobe debugger. That’s something that I’ve used a lot in the past and something that I know quite a lot of clients that I’m speaking to are seeing a lot of value from. Hopefully that might give a good indication of how that can work.

Does cross-channel analysis is really available? Is this still in development? If yes, ETA please. Cross-channel analysis. I think this is referring again back to customer journey analytics, which Jen showed. This is, I suppose, the next evolution of Adobe Analytics that we’re seeing within Adobe. I like to think about it as it’s almost like Adobe Analytics, but on steroids because it simply enables the ability to leverage all that data that sits within the Adobe experience platform and start analyzing that across scale. Obviously, within customer journey analytics, that’s essentially where it enables you to look at data that’s coming in from your email channels, i.e. those that are engaging with your emails, those that are potentially transacting from offline sources, those that are coming in via your web and app behavior, and any advertising data as well. All of this data is coming in, forming that really rich profile and those really rich data sets. That’s where you can start using customer journey analytics to start analyzing cross-channel, but also end-to-end customer journeys. A lot of our clients are seeing huge amounts of benefit of being able to basically analyze on all of their digital behavior, but also matching it to their offline data without necessarily having to use the Adobe Analytics customer attributes. Also leveraging things like call center data to be able to showcase NPS scores over time. That’s where I suppose customer journey analytics will enable the full cross-channel analysis capabilities. With regards to is it in development or is it available, it’s available now. I would definitely reach out to your account executive or CSM and see whether that’s something from a use case perspective that would be beneficial for you. We’re definitely seeing over here in Australia, it’s definitely one of our, I suppose, one of the products that we’re talking about most with clients at the moment, just because of those additional capabilities it’s providing. My aim is to see across my website for peak engagement time as per users local time rather than visits across the world all being merged together in admins local time. For this one, what I would recommend is going to the admin within Adobe Analytics and looking at the report suites and seeing what time zone that’s being analyzed within that report suite. What you can do is potentially look at having an additional report suite with potentially looking at IST time instead of maybe the time that you’re analyzing at this moment in time. That can all be looked at within the admin section of Adobe Analytics. So I’d definitely have a look there and see if anything can be done on that front. Just seeing the newest ones coming in. Okay, site section captures correctly on browser console but shows it wrong in reporting. What could be the reason? That is a very good question. And to be honest, I’m not entirely sure. Which again is not normally the answer you’d want to hear. What I would do is, obviously again, the Adobe Debugger is a perfect tool to start looking at potentially if there’s something that’s going on from an implementation perspective or potentially looking at if there’s anything from a metric perspective that’s not quite working. So I’d definitely look at the debugger and see if there’s anything there that’s jumping out. Otherwise, as I say, reach out to customer support and they’ll definitely be able to have a look at that for you. Okay.

Just going through the questions.

Okay, so there is a few more that’s come in now. So one of the questions that’s been asked is with regards to exporting the data out. What’s the difference between using the API to export data versus the data feed? I suppose that ultimately comes down to the user within Adobe Analytics. The frequency you want to get that data out. The scale of that data potentially is it a large scale amount of clickstream data or whether it’s simply a small report. What I would say though is most people find that if they’re a user within the tool itself, they’ll use the capabilities within the UI of Adobe Analytics rather than using the API. Normally because Adobe Analytics or most users in Adobe Analytics tend to be more marketers or those who are using the tool day to day rather than the more technical users that potentially might want to go down the API route. They’ll use cases and I suppose the frequency and scale that you want to get that data out of. And then final question is, is there anything that should be done while looking at calculated metrics or segments that haven’t been updated in a few years? So one thing that we’d look at there and there’s a great article online with the seven steps to cleaning segments. But ultimately what you want to do is look at what segments are ultimately being used within the tool. What we’d recommend is before doing anything is update the naming convention with do not use within there and basically tag it as a do not use segment and just make sure that from then people are actively not using that, not using the segment. And then ensuring that anyone moving forward doesn’t use that segment and therefore over a period of time you can see that that’s the case and then you can start deleting it. But ultimately I wouldn’t start deleting it without renaming that naming convention to ensure that people don’t accidentally or unknowingly want to use a segment that’s not available.

And cool. So that is all the questions that we’ve got today. Thank you so much for the questions. They’ve been brilliant. And it’s been a blast to be here. All I’d say is if you want any more information on Adobe Analytics, please do reach out on the Adobe Experience League. There’s huge amounts of information there. And also, if you have any questions or want to reach out to me personally, you can contact me on LinkedIn. Just search for Joe Morris and you can find me there. Thank you very much.