Getting Started with Adobe Journey Optimizer Analysis in Customer Journey Analytics

Transcript

Good morning and good afternoon to all. Joining will be getting started in the next couple of minutes. And just a heads up, today’s session will be led by Rodrigo or asha. He’ll be discussing getting started with agile analytics in T.J… We’re going to wait just another minute for attendees to continue filtering in. Thanks for being here. And we’ll get started. And while we have you here, we are going to just highlight a couple upcoming webinars that are in the next actual couple of weeks. I’ll be copying in some registration links. You’ll see we’ve got some great webinars across various solutions and capabilities as well as targeted for different audiences and I will be pasting in that registration for each of these in the next month. Again, just a reminder, we’ll be getting started very shortly. And meantime, sending along some links to upcoming webinars. Thank you very much. We’ll get started in just about 45 seconds. So we have a good amount of our attendees in the room. Thank you, everyone, for joining. All right. I’ll begin with just a quick opener here. Hello, everyone. Good morning. And thank you very much for joining today’s session, which is going to be focused on getting started with Adobe Journey Optimizer Analytics in J. My name is Christos Takis. I will be your host. I’m a part of the ultimate success team where we focus on helping Adobe customers get the most value out of their Adobe Solutions. I’m going to go ahead and kick things off and just give some context around this webinar. First and foremost, thank you all so much for your time and attendance. Just to note that this session is going to be recorded and a link to this recording will be sent out to everyone who registered. This is a live webinar, but it’s in the listen only format, but it’s very much so intended to be interactive in that as we go through content in today’s session, feel free to share any questions in the chat and the Q&A by our team is aiming to answer as much as possible. And in addition, we’ve reserved some time to discuss questions that are surfaced at the end of the session. Note that if there are any questions that don’t get answered during the session, we’ll do our best to take time to take note of those and follow up where we can. Additionally, we’ll be sharing out a survey at the end of the presentation that we’d love your participation in to help shape future programs. That being said, I want to introduce our speaker. I’m joined by Rodrigo Radia. Rodrigo is a principal customer success architect who has spent the last five years in Adobe’s consulting org, where he’s been focused on Agile and Adobe campaigns. He brings a lot of very rich hands on experience working with Agile as well as other products within the experience cloud. He’s also has a lot background in consulting, so we’re very excited to hear from Rodrigo today and we’ll go ahead and get started. I’ll hand it over to Rodrigo. Thank you very much for this. So here we the customer success architect based in in in Spain. And I’m here today to try to give an overview of the value that we can get from from Agile while integrating with Adobe with our customer journey analytics. Right. For Agenda one, we just had a webinar introduction. We had a brief session on why a deal a year better together, right? And why it’s not replacing at all some features and insight. And then a couple of practical use cases on how to configure and bring Agile data into customer journey analytics and a very simple use case that I think could be useful for for most of the customers that have both products, which is intended to show how Journey Optimizer tags can be used within CGA to produce and compare reports in. And afterwards we will have, I mean as time permits will have a summary in Q&A and then will close out and show me a possible link, as Crystal just mentioned. So let’s get started and let’s start with the what’s why are these tools better together? The first and not the would be related to that but the one just to introduce a feature that has been released in the previous month, I think if I’m not wrong that it’s the new on the new report. Okay in available in agile that our build based on on C day technology right. So these reports are in limited availability. Okay. And that probably I mean but they are fully functional and they will be replaced at certain point what they become in general available. And it’s more or less similar information, a little more advanced than we used to have in in a deal. Okay. And those are just mainly more or less the same metrics, but you have the ability to build there and they are using already the C data technology behind the scenes. Okay. Even those will be available not and it’s not needed that you have a a license. It will be available for everyone. Okay. And at some point when those reports get get generally available, the old out of the box reports in deal will be will be it will just disappear. And this one, I mean, this new reports will be the only the only way to see the reports. Okay. So if you own a activating this reports will create a new specific connection in and a data view and CD8 with all the needed metrics and data sets just with one click and that and those connections and data views, if you will own CG, then you can later on extend that new metrics and create custom reports and analytics that you can also jump from a geo user interface to CGA in order to to deep dive and customize the metrics to rate share and all those nice features. That CD has come any way. Even if CGA says it’s not there, you will be able to see that all those reports seen in radial UI and be able to to filter by by dominance campaigns and be able to do some, some more advanced analysis than we used to be able before with this report. With the old style kept still, a lot of the previous feature announcements that was then. We will continue with the story of integrating integrating agency for deeper analysis by integrating ADR and CDR and RTC to be also apply as well as as an activation platform and that’s life that’s available for you. We closed loop on marketing actions to collect, activate, measure and learn right then we would start to cycle again to get the data activated that measure the performance and then run it again. The objectives is to continue to look through these devices to learn and improve the execution and measure in every cycle. This way we will improve our results over time as we gain experience with the solution and understand better the way our customers interact with our brand. Both digital and CD. Our applications that sit on top of that experience platform. So the applications have most of its data shared between among them. In this case, do generates data in unified profile store that also travels to experienced platform data like PDA rich data directly from App Data Lake, and then can get all the reports and all the visualizations from that data like did doesn’t meet unified profile but only from from Italy. So there is data we joliette can bring execution instantly into app data lake and that is used to generate the ideal out of the box reports. Additional reports use the data to show detailed view of the ideal metrics whole journey and campaign execution. But the standard reports indicate you are not that customizable. You can generate new customer intelligence. You cannot define your own metrics. For example. So as both platforms sit on top of app, CGA is able to leverage the data generated by Agile and other data, ingest it into AP so you can combine the data generated by video. But any other data that is suggested into ATP and use more data sets for deeper analysis and take a holistic approach in this aspect. And ideal artist in Gazi to place will be in charge of activation actions and see they would handle the data analysis part. Now we will focus on the value added by day on top of Fadil implementation, and let’s start by compiling what we can do in a deal out of the box reports and what we could do by leverage. You can make the out-of-the-box reports. We can get reported insights like journey and campaign metrics. How many profiles enter exited the center? Abrams Counting How many profiles? Well, from each step, execution errors, etc. we can visualize channel metrics like for different channels, email system as suspicious as we can see, and metrics delivered bounces, errors, open clicks, and all this kind of information that is directly related to channel execution. And these reports allow us to answer questions such as how many people in their journey, how many messages that’s about in each campaign or in each activity, how many people clicked in the link or how many profiles took specific path in a job? Right. We see it. We can also build reports that deep dive into viewing data not just about the journey execution, but also commodity formation not generated by ideal, but by profiles activity such as what comes next, right? Like Bates visits flex time spent on Bates or mobile browsing behaviors. We can connect inside as well online on all client information, right? So we could ingest data like point of sale data purchases to avoid having isolated view of your business. Right? So to connect point of sales data and with mobile data, this way we can have the full online and also claim experience Tracking from the email sent through the click, then arriving into the website and getting the full browsing experience and see what’s the part of our funnel that behave better or if there are any problems, right? CDO So yeah, so it also allows deeper analysis by combining data from other parts, such as estimating the contribution of a marketing campaign to the sales attribution, creating more custom metrics specific and meaningful for your business than the out of the box metrics, which we can also then audiences back to HP to reengage with them and activate them back in activity or another donor or complaint in case, for example, we detect broken sales flow or something that is not working really well within our client experience. Ideal of out-of-the-box reports are available just up their Journey campaign is published to be able to check sooner what’s happened. Especially are it useful to monitor standouts and prepare recovery actions in case of errors? Right. So that’s usually the main use of a do out-of-the-box reports in, let’s say, the what we used to have in the last 24 hours and then the all time reports are useful to understand campaign performance, deliverability, performance. We have ability to focus in the specific time periods and we can deep dive. We have also the channel reports where we can see the comparison. It’s in different campaigns across time and and this kind of performance monitoring or performance reports are nice, but when we are CDR, we can have additional features like connecting data with additional datasets to understand the impact of the campaigns in both online and offline conversions. This will allow us to relay data that is not age generated but customer journey generated. Digging into additional insights, letting us go deeper in the analysis of our customers with other elements that are outside of Adobe Journey Optimizer, like the interaction web or in the mobile app, or even if we have some other interaction with our physical stores or similar, it would be something that could be doable. Through CGA. We can perform advanced scoreboard analysis to understand when seasonal and or special moment behaviors. This can be useful when comparing same or similar campaigns and or consecutive different periods of time or understanding return and churn rates. We can create flow and flow diagrams to help understand the and broken sense flows. Try to dig into root cause analysis for those subset of behaviors to experimentation with video campaigns is possible. It’s a feature in radio. You can you can experiment in content and subject. Heather Come and try to evaluate which version performs better. And there is already some out-of-the-box reports in in a deal that can help you try understand, then insight can let you understand insights on the winning version based on the expected behavior, which is usually one of the agile out of the box performance metrics like the the one that is the most clicked or the most open. For example, in case of emails. However, in CD eight, there’s a new visualization called Experimentation, where the success metric could be defined as any of the metrics in your implementation, right? So instead of just getting which version have most clicks, you could also try to implement which version finally ended in converting more people, right? I mean, getting more purchases. You could even try to have other of six metrics to effectively measure the uplift, understand your experiment is completed, are not from a statistical point of view. This is a really good tool when trying when tying campaign experiments to revenue related metrics rather than just message to open clicks. We’ll see a very simple example in the demo we have later about an execution and before we might test is this. However, due to the lack of data I have to in my in my instance, I will see that the experiment is also not conclusive. And so this concludes the introductory section of the value provided by integrating and K together. And now we will move to the next section where we did dive into the configuration of the related datasets and how to set it up to let us see did use the data generated in I do in this section I want to show how would you listen to really behind the scenes with CG and what’s the configuration applied to enable the connection and the data of you? Right. This configuration can be done in another meeting way. There is one click way of doing this. There’s a configuration applied in the background that creates everything for you when you enable the new CDA based reports. In Agile, there’s a toggle and that creates the connection and the data view in the in a cage. Then this case is not visible to customers who are only entitled to a deal, but it is there for customers with a proper c d license. Along in this connection and data value to be customized to be properly enhanced with other implementation you may have instead. So this one click configuration creates automation automatically all the connection, dimensions, metrics and filters needed to leverage the default reports with a technology or a. Although most of the customers or most of new customers would enable the CDR based reports with the wobbly version, it may be the case that some other customers may have already a mature implementation of CTA and they would prefer to have the deal with specific connection to an existing connection and their view instead of re-implement in their setup on top of the automatically created date of your connection. This configuration to integrate EDL and CDR should be a one off effort and the steps are detailed in the official documentation. I will provide a highlight on how to do this with the demo, and I have to say that I mean, I’m no no CAA expert and it’s taken me a couple hours probably. So if you have any TDA expert in your organization, it should be a one off effort. And if you want to do more complex stuff, it may take you more time. But just for the agile, out-of-the-box reports, it should be quite easy. So the first part of the configuration is creating the connection with in place, adding all the ADA out of the box datasets like message feedback and tracking data datasets, agile, ideal entity and journalist that events. With these datasets, the default EDL reports can be mimicked into a CDR. Then if we are interested in other informations to combine, we can add additional datasets like profile data to filter users or demographics, loyalty or online browsing behavior to track online activity and purchases. Point of sales, call center to track offline activity like offline purchases, call centers, Statistics about satisfaction and satisfaction purchases made through call center, etc. All this information can be used together to generate better insights Suite Dr. into the introductory section. The slide various second step is to create the data views from the connection we just created in the first step. There are also instructions on creating the metrics, filters and calculated metrics to be able to mimic the out of the box idea reports that’s shown in the official documentation. But we can add also or own business related metrics. Okay. Adding additional calculated metrics like unique open rates, additional click through rates like CDO, R or unique CDO, R, create create derived metrics, metrics such as those that we will use in the next exercise. We should also add the additional metrics from the dataset that are not related to AL directly, But I mean, this other additional datasets can help us understand better our customers after we are done with creating how data view. And I think the additional metrics we could already start descending into our workspace and the next step is that I’m going to try to show a demo. Okay, So let me try to stop here and share my browser and what it is. Okay? That should. So I’m just going to keep a couple of insights on how all this configuration works behind the scenes. And so there’s a one click configuration, but I think it’s quite interesting, especially for a users to see how all this works, right? So I have here my test instance, and this is the connection I, I created, right? And here I have the ideal out of the box datasets like Journey Step events. Okay, all the configuration is shown in the official documentation, right? Like the how to configure the personal ID and depends on your configuration. You will have to import new data or just the dataset that feels. So we have the journey step events, we have the our entity dataset, which is useful to convert the journey and campaign unique ideas to a human readable names to be able to to filter effectively and give more flexibility to our customers. So the key configuration here is how to configure the the key is the matching key is the person alias, but all that is displaying step by step in the official documentation. And then we have also some other datasets. This one is the regular dataset that tracks clicks for for you most. Okay. And we have also I mean, I have also some more data for other data sets for CRM. So this one is profile for CRM. This one would allow us to do in a filtering or breakdown based on, for example, demographics or legislative levels. If we have that kind of information, then we can go to the data views. And in the data views we have to add our components. Okay. It’s very interesting here because the information we have in the in the regular or or out of the box implementation, it’s nice, but we can create so many other metrics and components that we that are not really useful for for our metrics and create a complete set of I mean, give a new dimension on what we do with ideal reporting, right? We can use also that the right fields and example you have here, the email clicks, the email opens, we have the bouncers. So we’ve all also some calculated metrics or metrics that have some some additional configuration here. And whenever we have all this connected, we can just go into, I think this one and start dragging and dropping the metrics. I think this one is about just events, but it’s not related to the ideal. But I have here some ADL report again. So this is about the the feature we are going to see next, how to use tags and names. Okay. In freeform tables, this is information about which kind of actions. Okay. And flow diagram. We have also the new experimentation report and all this is information coming from the out of the box of Agile data sets, right? So this one is the one I took previously. One of the experiments is inconclusive because it has run for less than seven days. Well, this is my test environment and I try to generate just some data between different treatments and try to compare which version had more clicks. Okay. And what’s the leave gives you some static statistical metrics for confidence and so on, trying to give you insights on, okay, this AP test experiment. Okay, this this one has more opens than the other one. But is it conclusive? I mean, can we say from a statistical point of view that this person behaved much better than the other one, and if so, with which level of confidence? So this is the kind of well, not the kind of stuff we can do with with the CDA using age data, but just a small set of sample reports that we could see. Okay, now I’m going to share again my presentation and continue with the second use case. But my okay, so now we’ll go with the second use case in this webinar, which is the use of age or that’s within CDA reports. So first what our thanks in a deal right. Thanks are properties that can be applied to journalists and campaigns can be used to filter them in the main journey or combine this right so applying some time to a journey and then I can filter for, okay, give me all the judges that have this done. So that is the, let’s say, the out of the box usage of this of this feature. But I implemented recently for another customer one one feature and I think it’s it was insightful to be shown in this in this webinar and so talks are in defining two levels right that name and then tag category. You can define also tags without category and those will be grouped in the Uncategorized category that should be applied to a journey or a campaign before it’s published. Okay? Otherwise S.J. won’t be able to report on them. This happens because the ideal entity dataset records getting just that whenever the journey of the campaign is published. So this is a very important point. The Dutch, these use case or this feature will only work correctly if we target the journeys or campaigns before we publish them. So tags are fixed properties that cannot adapt to the business definition, so they are free text you can just write whatever you want and apply to a campaign or region. You can use them in the way you prefer. And from my perspective, from my experience, the most impactful exercise is to choose and group the right categories and that to be deployed on your environment, because that is what determines the usability of the future. So if you do, if you create tabs and categories that are really useful to your business, then your reports or your usage in CGA would be better if that grouping done was not that good, then this feature loses a little bit of its of its mean of its meaning. So my recommendation is to spend time in planning what kind of categories and that names I’m going to to create. There are some that category is second A can tell because I’ve used the in previous customers for example, there are time bound categories, light years systems, months of the year or day of the week. Those may allow comparison of seasonality. Comparisons are also some that categories could be activation type like for example activation think like newsletter or transactional campaigns, special offers, informational campaigns, security, etc. Sponsor category is also something that that is useful. I have seen in customers that are from media and entertainment and detained in the vertical that we may have. We may have some email campaigns coming from sponsorship, right? So we as a company have some sponsors and we have to advertise and to our customers on behalf of those sponsors. So then creating a sponsor category is something that is really useful because we can then create media reports based on that particular sponsor or report to our management in aggregate it just to check the performance or do seasonal comparison on the performance related to to those sponsorship categories or see which sponsor behaved better or compared to last year, what kind of campaigns behave better. So what else? Okay. We’re able to also recommend to to prevent attacks by department or team or region. Right. We have a several teams in our organization like regional or department. We could have that category categories like okay, this campaign belongs to the security or this is from the global marketing team or this is from us marketing team or email marketing team or something like that could be could be also useful in terms to compare to measure the performance of some major news campaigns depending on the region. Okay. So the first of these is really simple. Okay. They in the implementation inside deal, we have to head up the the tag categories of names and then apply the touch to the journalists or campaigns. Once that is done, that would be enough for for the additional part, of course, we have to remember to apply those categories and names. I mean, the all the facts Before the Journey campaign is published all the way that that will be available in CGA. After that, the implementation is is just in a day sorry insignia and it’s really simple. It’s that by default the tags are a field that comes in, in an array in the agile entity dataset, but those by default are not that useful because tags can be applied in a zero to end relationship. Okay, you know, zero to many. So we can apply a journey or a campaign. The amount of tax we want and the targets are also contains several values. So the implementation in CDA is not. I wouldn’t say that is easy, but once you understand the concept, it’s quite quick and it’s also a one off effort and so we have to create three derived fields, one for that category, one for bank name and one for both. Combine in the right field definition. It’s a two step is to look up in the transfer into the translation table feel entity record dataset and then regular expression replace. Probably. There are many other ways to do this implementation, but I just chose this one I’m no fool expert in and I found this that that’s just one. Once we have that done and journeys I mean and data is coming to the CGA we can just use it right So we always we’re able to to filter or group the results or the metrics by name, right? But now we can use some other filtering based on whatever tax or categories we want. And now I’m going to go back to the demo and try to show a set. Okay, Here it is. Okay. And for example, let’s see. And then you have this funny say, should and let’s create a three for table and say, okay, I want the message to deliver and your nick clicks. Okay. And and then I can add the dimensions of okay, let’s see your tag category and I want to replace this one that way and it will show us if I go here to last 90 days, where is where I update the stick. Okay. So here we can report on the categories. Those are the categories I have in my in my instance. Let me just see if I could show them here in my tags I have here my categories. So campaign type sponsors, team, Uncategorized year and something. Right? And then inside sponsors, for example, I could have campaigns for Allianz for I don’t know. I mean, I just figured out some some names and then apply the objects. So this is the configuration done in in in a deal thing I have here in my list. The journeys if it wants to load. Well anyway, I have here another window. So here is where I have the tags this campaign has. Okay those that I have applied. And now apart from just being able to filter here, okay, we should be able in CD8 to break down by campaign type A. Okay, now let’s see. Okay. Joe, did that name good journey tag name for sponsors. So I think is a quite simple use case because the implementation itself, I mean, what it takes then is just the the amount of time you spend figuring out which tags you want to use. But then the implementation in ADL and in CDR is really straightforward. And you can use it for creating aggregated data, filtering in and and so on. And up to here. Well, I think sort of I think I had a bigger one, but well, it’s more or less the same since information is shown here that we are. So we can look and see how the different campaigns fare, different year, how it perform, and then we can break down again by name or by any other dimension or a metric we may be interested into. And up to here, the my presentation. And now I’m going to share again window. Where is it Here. Okay. And just us summary. We have the key points here for S.A. helps you to dig deeper into the only behavior items that interest across the datasets and the power that allows you combining different data from different data sources to get more meaningful metrics. And here have also been the links to the out of the box configuration for the out of the box integration now and then over back to Christmas. Awesome. Thank you so much, Rodrigo. And we’ve been getting some questions in the the chat I want to tell before we dive into the Q&A portion of our event, there is going to be a link to a quick two question poll just generally to get your feedback and to help shape sessions. You’ll see that coming through momentarily. And let’s jump right into that some of the questions for Rigo. So first is our why is something that we can configure in CGA is is something automatically calculated by the tool and presumably the tool being a Joe is oral. Why something that now in a day all we want we won’t be calculated our way, but that’s something we can do within CGA right? Because you can tie, you can tie the product lookup or when it definitely something you need to, you can, you can perform okay within CGA but in deal you’re sending I mean the information generated by deal is okay, I’m sending messages, I’m sending push notifications in our offers, but that’s the information generated AGL Then you have to extract from the purchases dataset, the the revenue and that’s used to calculate the ROI. So it’s not, it’s not directly calculated by ideal, but could be calculated within CGA using the ideal data. In this way probably you will be understanding that you will be able to understand the attribution more appropriately. Right? Right. So it’s, you know, not necessarily something front and center, but much like any other marketing or personalization activities, you can use proxies or other success metrics that may be indicators of success. So let’s say even if you’re not a revenue generating part of the business or maybe you’re more focused on B2B, there’s there’s proxies in place where you could use to to get to understanding the impact. And that can be done in CGA because we have that ability to look across various touchpoints. Okay, great. Next question was around GEO data in CGA and wondering if that is something that is updated in real time. How close to real time? It’s not real time directly because the way the information happens, it’s cool. So it generates the data near real time that flows into streaming connectors, near real time to be able to react on, you know, clicks, email opens, but then CGA, CGA rates from data lake So it go the information flows to unified profile down to the data lake And then CGA will adjust the data from the data. LAKE So it won’t be real time, but I think in my past projects I’ve observed something like close to one hour delay at much. At the most, Yeah. Yeah, a most. Okay, yeah, maybe I that may be published like the actual, the detail. What is the expected. Well I think that would be probably in, in sea data you guardrails. I’m not 100% sure because I’m not an expert on all of those regardless but I see that is tied not to radio itself but to the rate of data ingestion said in CGA. That makes you do. That makes sense, right? Okay, we’ve got a couple more. Let’s see. This was from Marina. Marina has been experimenting with the default data view in the last few days. However, she had difficulty finding journey entrance metrics. Most are related to tracking channels. Does she need to manually set up those types of metrics? Should and shouldn’t be necessary? What I think it could be happening, I’m just guessing. Okay, because I really don’t know exactly what’s up. I know that is actually happening, but any deal, if you use the aggregated reports, okay, if you go to administration, I mean, generally reports in the left trend there you have the aggregated data, but Jardiance was metric. You’d be available in each of the reports specific. Journalist So if you navigate to the reports from the journey, you should have a specific metrics related to the journey entrance. That’s what it should be. Okay, great. And she followed up with a question around the documentation about default metrics in CGA data views that were shown on a slide from earlier I had sent across, and I think everybody should be able to see it. Just a link out to the default out of the box values for connections for data views and CGA in relation yo so hopefully that that addresses and is comprehensive with with what is expected there. Yes. I mean there’s also you can always I mean there are definitions of the data needed to mimic the ideal report. But anyway you can always more stuff right. If you understand the agile data dictionary you can create your own dimensions and metrics and the type more accurately into the into the data. Yeah, great. All right. So I think we’ve got one more question, and I, I will try to phrase this. What are the resource and permissions required since our team will need to use AJO and CGA capabilities. And presumably what we’re talking about here is what are the access levels or permissions levels required for someone to essentially do what you demoed? Okay. For the initial configuration, I think that shouldn’t be a problem because usually that that one off configuration in creating the data views and metrics that’s usually performed by an administrator and it’s I mean is then just once okay so those are probably administration because you would need creating connections and managed state of use. But once that set up, then you could just set up the level of access that you want in, get to access another view and access to certain components. That would be enough. And then in Nigel for the I mean that so related to the first use case that was bringing data into of of ADA into into CGA for the second use case in ADA, the only permission needed is be able to manage journeys. Okay. So be able to create and publish journeys that would apply just to, to apply the tags. And in CGA it would be administrative for once because you have to modify a component. Okay. But it would be just also a one off effort. And then with regular reporting permissions should be in a great. Great. That actually concludes the questions that we received. Rodrigo, really, really appreciate you taking the time and walking us through the demos. I learned a lot in regards to tags and and best practices and strategies around that. And overall, the exciting capabilities between Ajo and T.J. with that being said, I want to thank everyone for taking the time to join the session today. We hope to have you again on future webinar and please make sure to respond to the the the link in our chat, which is it’s a quick poll that again, is very helpful for us moving forward. Thank you everyone very much and look forward to seeing on the next one. Thank you.