From Data to Action: Uncovering Valuable Insights

Hello everyone and welcome to the Skill Exchange for Adobe Analytics. I’m Iwona and the session we’ve designed for you today is focused on beginner users. So if you’re just starting your journey with Adobe Analytics and you’re not sure what to do in your first steps there, hopefully my session will be able to guide you through. Before we get into it, a few things about myself. I am a Senior Analytics Manager at IBM Consulting. I’m originally from Romania, but I’m currently living in London and I have a Master’s degree in Software Engineering. As far as for my hobbies, I’m into anything that has to do with creativity, baking, painting, anything to do with arts and crafts. If you’d like to keep in touch, I’ve included a QR code, so please feel free to add me on LinkedIn and we can have a chat there as well. Now let’s dig into it. On today’s agenda, we have a few topics. We’re going to discuss the Adobe Analytics basics. We’re going to then look at how to approach data analysis if you’re not sure where to start from. And then the last section is focusing on sharing and collaboration and different features that Adobe Analytics has and can enable you to do anything to do with self-serving analytics within your organization. We’re going to end the session with a Q&A, so please send over any questions you may have. Web analytics basics. I really believe it’s very important to understand your basic metrics as those will form the foundation of any reporting you’ll be doing in your job. I have picked the most commonly used one that I believe everyone will use on a day-to-day basis, so let’s just have a look at them. The first one is unique visitors. This usually represents the number of unique users that are on your website or application at a given time. This could be looking at an hour, a day, a week, a month. It can go as granular as you’d like. And usually a visitor will perform multiple visits on your website. So there’s always a correlation between these numbers. Usually the unique visitor’s number is lower or equal to the visits one. The visits number is a sequence of page views in a sitting, so you can imagine a user coming and spending some time on your website browsing. This metric is commonly used in reports that display the number of user sessions. That’s another common name for it. And again, it can look at different periods of time. But visits usually consists as well of multiple page views. So again, there’s a correlation where usually the visits number is lower or equal in some cases with the page views one. And lastly, speaking of page views, those are counted every time a user loads a page on your website, so every time that server call is sent. And it’s very important to note that track link calls are not counted as page views. Now let’s get familiar with an analytics workspace now that we’ve discussed some of the metrics. So let’s have a look at the landing page. This is where you’re going to be presented with a list of your projects, which you can search through, you can filter, or you can even organize in different folders. There are some options for you to customize this landing page and to filter your projects. In this specific example, we’re just going to start by creating a new blank project by clicking the top right button. You are going to be presented with some templates that you may have created for different types of reporting. But just for the sake of this example, we’re going to start from a blank workspace project. And please keep in mind, this is demo data just to show you some of the features. So you have a blank panel with a blank freeform table. And then on the left hand side, you have different sections for panel types such as blank panel or page summary. You’ll have a section for visualization types where you can select things like a donut chart or a bar chart depending on what you want to analyze. The next section is focusing on metrics. So you’re going to be able to see dimensions, metrics, segments, and this is where you can also create, search through them, or select date ranges. There is also an additional dictionary section. Now going back to the panel we’re working with, a very important feature that you should always check is a date range at the top of the panel. This is where you’re going to select the time period you want to analyze. In my example, I’m looking at a specific week’s data and you also have a menu where you can choose from presets like last week’s data, for example. This will ensure that every time you load this project, it will show you rolling data. So every time we’re going to be looking at last week’s data. Now let’s imagine we want to analyze unique visitors for that last week. I’m going to search for the metric on the left hand side and I’m going to drag it over to my table at the top. Now I can already see a daily breakdown of unique visitors and I can notice if there’s been any sort of increase in traffic or decrease throughout the week. I’m going to name my table daily unique visitors and then I’m going to also name the project and ensure I save all my work for future reference. Giving it a name. Now this table is showing me numbers for each day and it’s also showing me a total. I can customize the size of this widget and I can place them around the way I want. But let’s say I want to get a bit more detail about those unique visitors for last week and not just some numbers. I want to see how the data look like compared to other days or period of time. So I’m going to look for the key metric summary visualization which is one of the newer ones that Adobe has released and I’m just going to drag that over to my panel. Now I have a couple of options to include the metric I want to analyze. So I’m just going to look for unique visitors and then looking at the date range is the panel date range comparing to the week before. So that’s already selected. I can also apply segments but for this analysis I’m going to keep it simple. Now the widget is loading. It’s going to show me the total number of unique visitors for that week and the percentage change compared to the week before. So I’m already getting some insights that the week performed better and there has been an increase. I can also see a trend line on how the numbers look like throughout the week and I can choose to emphasize the widget on the number rather than the percentage change. And that’s it. I’m already getting insight on how the number is performing over time. Now let’s discuss how we can approach data analysis. Let’s say you want to start analyzing your data but you’re not really sure where to start from. The way I like to explain this concept to new joiners in my team is that they should focus on four main questions because they are covering most areas in analytics. So starting by asking who are your users? This will be looking at what the demographics of your users are, what their preferences are, what devices they’re using and it will enable you to segment them in the future on different types of users. Next where are they coming from? This will be looking at metrics like marketing channels, referring domains, previous pages and it will help you understand which of your channels are performing best. Now the next question, what are they doing on your website or application? This is looking at the behavior. Are they clicking on specific components? Are they just landing on the homepage and then bouncing off going to a different website? Are they actually completing a form? This should tell you if the specific areas you’ve designed for their digital journey are used the way you intended them to. And the last question, which is also the most important one, are they converting? Now conversion can mean different things to different industries, different websites and platforms. It could mean placing an order, it could mean filling out the form, but that will tell you if your business is succeeding or not. So it’s very important to actually get to that question as well. Now we’ve discussed how to start analyzing, but now you can start building your measurement frame as well. You can align those KPIs to your business objectives. You can start gathering input from different stakeholders and understand their needs and their objectives as well to be able to start prioritizing those KPIs. And you can understand their impact and their relevance in the whole ecosystem. And I like to think about those KPIs on three different layers. There’s your top level KPIs, which everyone will use no matter what industry you’re in or what your product is doing. Things like the ones we’ve already covered, page views, visitors, visits, they’re quite standard. Then the next level, the next layer looks at goals measurements. So understanding if your users are converting towards your goals, things like leads generated, orders placed, or the revenue that comes from your digital space. And then the additional layer is the actual behavioral analysis where you understand which components your users are interacting with, whether they can find the right content, they can use the navigation correctly, whether they’re starting forms and completing them. Now we’ve discussed the measurement frame and you can go into customizing your tracking setup, which I believe it’s very important to be as tailored as possible to your system’s unique needs and data points. And in order to do that, it’s very important to be able to maintain an open communication with your development team and demonstrate to them the value of their efforts, because in that way they can make sure your tracking setup is as efficient as possible and they can improve it to the point where you get very rich data out of it. The last point here, it’s about error tracking, which might not be as common, but I believe it’s very important to have that in place if you want to understand technical issues and provide granular insight to your teams to be able to investigate and find a resolution before your customers face those issues on your platform. We’ve discussed the tracking setup, we’ve got the KPIs, but now we’re looking at the data and we want to make sure we’re adding context to it as well. Now let’s discuss about adding context to the data. I’m going to go through a couple of concepts that I believe are very important to take into consideration no matter what industry you’re working with or what your product is designed to do. The first one is actually knowing your industry and how different external factors can impact your numbers. As we’ve seen during the past couple of years, there’s different events or decisions made by government, and these can impact your numbers even if you don’t realize that in the moment. Knowing your industry is very key. Paying attention to trends and patterns. This refers to knowing the average of your numbers, to being familiar with those, and being able to easily spot when something is changing throughout the week, throughout the month. If you always know what they should look like, it’s very easy to point out when something is going wrong. Setting up alerts for your main KPIs to be one step ahead of your team. That’s also a very useful feature that Adobe Analytics offers, and I believe one of my colleagues will have a session dedicated to that in the next couple of hours as well. Historical data is your friend. This is also related to the trends and patterns you’re seeing in your data. Making sure you understand how seasonality can impact your numbers. For example, if it’s a holiday season, maybe it’s a bit more quiet, or depending on your industry, maybe that’s when your numbers increase. It’s just good to know that that’s not something to panic about, and historically it’s been the same in the previous years. Estimating the cookie control impact on your number. Most of us nowadays use cookie control tools that allow users to opt out of tracking, which means we’re not seeing a full picture of our numbers. But trying to compare our data sources with other data from different systems can help us estimate how many users are rejecting cookies and still painting that full picture. And the last but not least, of course, integrating other data sources, maybe from your media agencies, from the search team, just to understand the full customer behavior before they come to your platform and maybe if they leave the platform to go somewhere else. And now that we’ve discussed all these concepts, let’s get into action and see how we would analyze something. So let’s say I’m trying to look into the conversion rate of my website. I’m just going to go into Adobe Analytics from that blank panel, and I’m going to drag the conversion rate over in the table and just have a look at the date range and select the full month of June. Apply the changes and we can see a conversion rate of 7.31%. But that’s not telling me a lot about the month of June. So I want to start looking into more granular numbers and maybe visualize the conversion rate as a graph line to see if there’s been any kind of spikes in traffic, any dips in conversion rate and what possibly caused them. After I’ve done that, I can already get more insight out of it. So there’s been a decrease in conversion rate on the 11th and then on the 19th and 21st of June. Now, this is good because I know something changed and the conversion rate got to around 5.20%. But I want to take this further and understand what caused it, if there’s been something wrong with the traffic, if there’s been a technical issue. So I’m going to try to isolate my analysis to those couple of days where the decrease happened. I can see the conversion rate is calculated by dividing online orders by visits. So I definitely need to look at those metrics as well. So what I’m going to do is duplicate my table so I can add more metrics to analyze. And I’m just going to hide the graph source for now. In my table, I’m going to include the visits and online orders as well to see how those metrics perform throughout those specific days. Now, from the left-hand side, if I go at the day dimension, I can select specific items. So those days that I’ve mentioned from the 19th to 22nd of June. If you keep your command or control keys pressed on the keyboard, you can select multiple items and drag them over to your table. I’m just going to do that. And then I’m going to drag them over at the top of the table to filter out my numbers by day. Once the data is loaded, I can already see that there’s been a gradual increase in visits over those days. You can see from the 19th to 20th of June, there has been a considerable increase in traffic. However, it has not reflected into online orders. So that means that I did get more traffic to my website, more visits, but they did not actually translate into more online orders, which could be something expected if maybe I had a campaign that was meant to bring awareness and consideration. And it was not particularly designed to increase the orders number. But that unfortunately had an impact on my conversion rate because it did not increase the orders. So now it’s the point where I’ll try to understand what happened. This is where I would engage different stakeholders to understand any sort of activity that happened on the website that could have bumped up my visits and maybe I wasn’t aware of. I can also take my investigation in workspace further by looking at the sources of traffic. So I’m just going to take the marketing channels and break down visits by them to understand where this traffic has come from, what caused the increase. Once I break down the visits and then ordered the rows by the day where the increase happened, I can already see that the unspecified channel is bringing most traffic for that day, which is something I wouldn’t expect. Unspecified means that Adobe wasn’t able to identify which channel has brought most traffic to the website. This could be caused by a technical issue. It could be caused by not using the right URL parameters in my campaign maybe. And this is why it’s important to work with different teams to understand what activity happened. Now, if you look on the second row at email, you can also observe a slight increase in email traffic. So perhaps there has been an email sent out where the links were not tracked correctly and the traffic was not attributed to email. But that’s where you need to continue your investigation. Now let’s talk about sharing and collaboration. The ultimate goal is to help your team self-serve, reach data maturity, empower them by supporting their work through data and insight. So here are a couple of features that you can use in Adobe Analytics to achieve that. You can share your projects with other workspace users. You can schedule reports on a daily, weekly, monthly basis, depending on their needs. You can also create templates based on different types of personas or requests. So for example, you may want to have reports for campaigns or for product analysis. You can use text boxes within workspace to add definitions, context, and disclaimers. So let’s say there’s been a technical issue and you want users to be aware of what happened during that day, or you may want to add definitions for marketing channels to ensure that everyone is on the same page when analyzing those. The last but not least is to have a dedicated space for your team that includes learning materials and best practices. This will help you make sure there is consistency in reporting and everyone is looking at the same metrics and understanding them correctly. Now let’s have a look at how you can use those features in workspace analysis. If you look under the share menu, you can select sharing your project with workspace users and you are presented with a menu that allows to give them different types of access to your project. It could be editing the original if you want to collaborate on the same project with them. It could be editing the copy if you want them to copy your project and work on their own, or it could be just read only writes if you just want them to see the report. You can also share a link with them. Just keep in mind that gives them editing copyrights as well. The last thing I would like to show you is a bit of a template and something I would expect to see if someone was to build a overview workspace to show me the performance of a website during last week. So just a few simple things that you can add and help maybe the execs in your company understand the performance of the website. As you can see at the top, I’ve added the main KPIs that we’ve discussed about, the key metric summary and how it performed against last week’s data. So things like the visits, the unique visitors, the online orders and the conversion rate. Those will be the high level KPIs that everyone cares about. Then I’ve added a traffic trend line just to understand how those look like throughout the week and whether there’s been an increase on a particular day. And last but not least, I added some sources of traffic, things like marketing channels and referring domains. And of course, how the conversion rate looks like for each channel, because that helps me understand which of them performs best. That’s everything I wanted to cover today. A few takeaways before we go into the Q&A session. It’s important to know your basics, know your foundation metrics, because you’re going to work with them all the time and you need to be familiar with them. You have to add context to your data. The same numbers won’t mean the same thing for different industry or different products. So context, it’s what helps you understand and generate insight for your teams. And last but not least, sharing and collaborating is very important. It’s enabling your teams to be data driven, to use insight to make decisions. So just use all those features that can get you to there. That’s it for me. And I believe now we’re going to go to the Q&A session. Thank you very much.

That was great. Thank you for joining us, Ioana. Thank you so much for having me. It’s always a great opportunity to be able to talk about the things I’m most passionate about. We are happy to have you. Let’s jump into some of our Q&A. As mentioned, if you have any questions in the audience, please drop them in the chat and we’ll get to as many as we can. With that, our first question is from Pritpal. What is the difference between visitors and unique visitors? That is a great question. And I think throughout my answers, I’ll keep referencing back to the experience leak. You can find all the definitions there and they’re probably going to be better than what I’m going to explain here. But yeah, keep that in mind. So with unique visitors, that’s the actual naming that you’ll find in Adobe Analytics. And by their definition, users are usually unique. So that’s the metric you’ll be using in Adobe Analytics. However, you can break it down further if you want to see returning users or new users. So you can make the difference between users who have come to your platform for the first time maybe, or maybe they’re returning for a second or third time. But generally the terminology of users or unique users is the same. Great. That is very helpful. We have a follow up question that’s a bit similar. What parameters are captured to determine a unique visitor? So within the implementation, I think there is a couple of different ways to do that. It can be done either through a unique ID that it’s used to kind of recognize a user, or you could do it through cookie information that’s being set in the user’s browser. Again, the documentation should show you the different ways and the naming of those cookies as well, which is super helpful. But it’s very important to know that sometimes users can clear their cache or clear their cookies, which means they can be recognized as a different user after a certain period of time. So it’s good to take that into consideration as it can happen that the user throughout time is recognized as multiple users. That is a very helpful call out and reminder for our audience. Our next question is around dwell time and what that is exactly. Sure. So dwell time refers to the time that the user is actually spending on your website or application. And in Adobe Analytics, you can have a couple of different metrics to use. You can either understand how much time they’re spending on a specific page, or you can associate the metrics to specific dimensions. So you can maybe try to analyze how much time they’re interacting with different components or sections of your website. You can also see the entire time spent within a visit. So you can get to calculating averages and just knowing usually the interaction with your website platform, if it has a lower time or a longer period of time. So plenty to dive in. But again, Experience Link will be your best friend in explaining those in detail. Awesome. So basically just how long people are hanging out on your website. Exactly. Our next question is from Raul. What is a track link? That’s a very good question. I think I’ve referenced that in terms of page views. So track link refers to those exit links or maybe download links that you have on your website. It was just a reference that those are not tracked as a new page load or a new page view, but they are tracked as clicks to either external platforms or to download specific documents. And there is a specific way to implement that in Adobe Analytics as well. Awesome. It sounds like those are very helpful and seem to be pretty necessary in your reporting then. Our next question is coming in from Kendra, who’s asking, how do you set up error tracking? This is one of my favorite topics, if I’m being honest. As much as I love analyzing what users are doing on a website, I love to see error tracking because it flags every time something’s going wrong with a website or something’s technically incorrect if it’s been deployed to production and something happened. This is very much dependent on your implementation. So what I would suggest here is to work with your development team and implement what makes most sense for you. If you have a bunch of APIs in your website and you want to have some sort of tracking in Adobe Analytics that tells you if those APIs are failing, I would recommend you working with them and trying to implement that and getting it into your reports. If you just want to track 404 page views or if you want to know when an app crashes, similarly I would try to make it as custom as possible and I would try to capture in the error messaging some specific codes or some specific messages that then can help your team identify what’s wrong. Maybe you have some sort of dictionary that different error codes mean specific issues and that kind of reduces the time to resolution, which can be quite helpful, especially in a live production incident. Awesome. That’s really helpful. And I think that note on making sure you’re documenting what you’re setting up for your error tracking is probably just as important. So a very helpful tip for our audience. Our next question is coming in from Alyssa. How can you add channels to the marketing channel dimension? A great question. I think you need to have admin rights to be able to do that within your Adobe Analytics admin console. You should have some settings specifically for marketing channel and how they’re defined and how they kind of overwrite each other, which is quite handy. I think there’s another session today that focuses as well on marketing channels, so it might be worth staying and hearing that as well. And yeah, you’ve got a quite easy format to do that. You can create channels based on your needs, on the sort of type of media that you use for your platform, for promoting your platform. And you can also, if you have marketing channels that should have less of an impact and they should be override. Let’s say you have direct channels, so users accessing your website through a bookmark or through a direct link, and then you have search engine or paid search. You might want paid search to override the direct channel. So you’ve got these options in the admin console and it’s quite easy to use. But to Alyssa’s point, it’s very important to document what your marketing channels mean because I think different teams understand different things from marketing channels. So it’s quite handy to have that and share it with people. Awesome. That’s great. And kind of relates to Trevor’s session talking about democratizing data, making sure you’re making that easy for everyone in your organization to understand. Our next question is around defining metrics and how did you define metrics like online orders? I think it’s something that, again, you should do based on your setup and how custom and how you want the logic behind it to work. When it comes to e-commerce, of course, you have a set of metrics that you can use and you can set up your products and all the orders and everything within Adobe Analytics. And it’s quite easy to follow if you, again, go to Experience League and find the definitions there. So, yeah, with e-commerce platform, I think those very specific standard metrics and I mentioned that you’re using like product order, basket size, revenue, they’re quite standard. So I would just try to follow the documentation. Awesome. That’s very helpful. Experience League is a great go-to resource at that documentation. I know we also had a question from Joseph who asked where he could read more about error tracking and I would say Experience League is probably the good place for that as well. I think so. I’m planning to actually write a blog post there about that because I care so much about errors. So watch that space. Yeah. And you can check that out on the Adobe Analytics community. All right. Our next question is from Mohamed. How do we get sample data into Analysis Workspace? Are there any labs or practice data available from Adobe? I think that depends very much on your organization and the sort of contract or the sort of partnership you have. There are available labs, as far as I know. There are some sandboxes maybe you can access to practice. Of course, with sample data is a bit difficult because it’s quite limited. It’s not necessarily real time always. It’s good practice, but you have to keep that in mind. A little caveat. As for getting sample data, I think there are ways to do that. I haven’t done it previously, but I think if you work with your admin team, they might be able to upload and insert some of that data for you to kind of practice with. Awesome. Very helpful. Very helpful. And always helpful to get some practice in when you’re taking a look at all your reporting. Our next question is from Anna. Can a goal or objective be set up for a certain metric so I can see an actual result, but also compare against my company target? For example, having a conversion rate that’s above 5%. Yes, I think there are ways to do that. I would say when you go into Adobe Analytics, there are all sorts of configurations you can access in Workspace by creating your own metrics. When you create those metrics, you’re able to add constant variables. Let’s say you have a threshold of 150 for a specific metric. You could create a calculated metric that takes a certain value if that threshold is being surpassed or not. There are ways to do that, but what I would recommend, and again, I think we have a session around alerts today, is to set up an alert for your metric. If your goal is that important and you need to know how it does, I would recommend setting up an alert so you can see if it fails behind or if it’s actually reaching the point you are expecting. It’s quite handy and it’s something that you can easily kind of action on if something bad happens. So yeah, two main takeaways. Have a look at calculated metrics and play around with those and set up some alerts. And then in my presentation, I did a bit of an overview of the key summary metric. Again, that’s something quite handy because it shows you a red and a green depending on what’s happening with that metric in a specific period of time. So there are quite a lot of ways to track those. Awesome. Very helpful. And as you mentioned, we have Gatai who will be joining us later who will be talking about alerts and we also have some sessions talking about visualizations. I know you just mentioned the color coding can be really helpful when you’re trying to narrow in on data in your reports. All right. So we have a question from Ricardo saying, I created a project for landing page A and now I want to create a copy to analyze landing page B. How do I create a copy and work on the report for landing page B without affecting the original project? That’s a great question. You have options in Analysis Workspace to duplicate your reports and maybe apply different filters. I think that would be a great way to try to approach it if you create maybe a standard workspace just generally for analyzing the landing page and then you can easily apply a filter at the top of the panels or different segments to be able to do that for different pages. What a bit of a trick that you can also do within the same workspace if you have segments or calculated metrics for that page, you could create a dropdown so you can select different pages within the panel and you can do that by on the left hand side selecting the values you’re interested in. So let’s say page A, page B and then dragging them at the top of the panel, which creates a bit of a dropdown. But if you want to keep them separate in separate workspaces, you can duplicate that and you’ll have a copy of it and you can make any changes you want. So it’s quite handy, but maybe start from like a template which you can then apply to multiple things. Awesome. And very helpful tips to have. Our next question ties back to some of those visits and unique visitors we talked about earlier. What if the same user is getting into your web application through mobile and through desktop? How does Adobe Analytics recognize that? That’s a very, very good question. It’s quite difficult to track that. I know there is some identity stitching coming in like CJ especially is focusing on trying to bridge those gaps between different apps and websites, which currently in Adobe Analytics can be a bit difficult. You can do that if you have, let’s say, some sort of login state for your users. So if they log in in an app and on the website, you’ll probably have some sort of anonymized ID that helps you recognize them. But I think currently in Adobe Analytics, that’s a bit of a challenge. So I would look towards CJ maybe. Great. Very helpful. And we have a question coming in around visualizations and visualizing data. Do we have the option to show the data visually and in a way that we could use for presentation purposes? Yeah, absolutely. You can use all those graphs or bar charts or anything you use in the workspace. You can either download it or you can simply screenshot it in presentations. If you mean interactive, present it in an interactive way in a PowerPoint, I don’t think that’s necessarily possible. But I’ve done it multiple times where I’ve just taken the result graph or any type of visualization I was using and I put it in a presentation. So you can definitely do that. If you’re not happy with that, you can download the data from the table and just create your own visualization in that format as well. Awesome. Very helpful. And I know we also have some great sessions later that’ll talk a little bit about transforming your data for effective storytelling using visualizations and some great playbooks to kind of get you started on Experience League as well. We have time for one more question. And this is a good one. Where would you recommend to start if you’d like to better understand how to interpret your data and understand better when the data is showing an action that’s needed? That’s a really good question. I think within our community, within Experience League, within all of these spaces where Adobe Champions and Adobe Experts are kind of interacting, they usually share a lot of good stories of how they approach that. So it’s definitely a good starting point. I also think just by going through the learning paths that you can access for Adobe Analytics, you can go through a lot of scenarios and examples of how it’s being used in real life. So it’s quite handy, especially if you don’t have prior experience. I would start by doing the training and then simply ask questions on the forums on Experience League. Everyone’s more than happy to help you. Just because they’re so passionate about data, they will just answer your questions immediately. That’s how I started. I didn’t necessarily know what I was doing in my day one, but just by asking the experts, going through the videos multiple times and you just experiment, try and failure. That’s my best advice. And once you get a sense of your data and how your users interact, you’ll be able to start spotting this immediately and it will come in very naturally. Awesome. That’s a great answer and truly the power of the community, which is why we do events like this. That is all the time we have for questions today, but thank you, Ioana, again for joining and for sharing all of your insights today. Thank you so much. It’s been such a fun experience and hopefully get to do this again. Thanks everyone.