Getting Started with Adobe Content Analytics

Transcript

Hi, everyone. My name is Danielle Doolin. I’ll be kicking things off today. Welcome to our webinar on getting started with content analytics. Let’s kind of walk through what the agenda is going to look like today. We’re going to go through some of the objectives of the meeting, the session. We’re going to discuss why content analytics matters, why it’s present in today’s day and age, why AI matters when you’re looking at content analytics amongst other things, and then how easy it is to get started with implementation. That’s really one thing we wanted to stress today is the fact that it’s easy to get stood up with content analytics once you have the right tools in place and to be able to start to see your content performance data so you can make more impactful decisions. It’s going to be a lot more about kind of live reporting and demos today versus a lot of slide ware. So we’re excited to showcase for you some of the capabilities and how to get started with content analytics. We’ll then go into some resources that are available and open it up to questions at the end. I’ll also mention in the corner, you’ll see a list of resources. This is just some additional areas that you can go to get more information on content analytics if you’re interested. So to get started with the presenters for today, we have my counterpart on the product management side, Director of Product Management, Priti Bhutani. She’s going to be walking through content analytics reporting. We also have our special guest, Joe Christopher from BlastEx. He’s the Chief AI Officer. Also check out his BlastEx blog. He actually created a blog for us when we launched last year. We announced it last year and have some other exciting announcements for content analytics this year. So make sure to check out those resources. For the webinar today, what we’ve really identified with content analytics and what we’ve started to do with analytics in general at Adobe is to provide it to democratize work, kind of internal teams within your organization. So it’s not just about the analysts being able to leverage this data. We want the content performance data to be able to reach the hands that need it. So for content creators, they’ll be able to now understand the content they’re creating, how it’s really performing, how it’s making an impact, and really how to optimize that content to make it stand out more and more engaging to customers and end users. Then we have content marketers. They’re creating content across different channels. They need to understand maybe content is resonating on mobile devices and less so on web devices. How do you kind of adjust your creative delivery and creative activation plans in order to really effectively serve the right content for your end users? And the analysts are able to bring forth this reporting to democratize this data to the different teams and help them to be able to understand how impactful content performance data is in that content supply chain workflow.

So after today’s session, what we really want to just encourage that you are able to kind of dig into content analytics a little bit more, understand how easy it is to get stood up and launched with content analytics, be able to understand how you can effectively connect content performance data with engagement and behaviors of your end users, and then deliver really impactful asset level insights and performance data to your management team. So really being able to leverage that content performance data to make it stand out more So now I’m going to go ahead and turn it over to Joe from Blastex.

Thank you, Danielle. And hello, hello, everyone. So today, we’re going to start with a clear understanding of why content analytics matters. And I’ll speak to this from my perspective of being at Blastex Consulting. So, you know, at Blastex, we work with organizations on their content and analytics strategies. And this diagram, you know, really captures, you know, what, how it works at a high level, right. And we’re going to get into the details of this here shortly. This diagram right here that we’re showing though captures what we see the content lifecycle has become incredibly complex. Planning, creative collaboration, scale production, publishing, at every stage, and different teams are working in different tools with different data. So let’s on this slide, let’s look at the analysts on the right side. In most organizations that we work with, the analyst is one of the most disconnected people in the content lifecycle. They don’t know what to measure because nobody upstream has connected the business question to the content strategy. They’re building reports with the context without the context to make them actionable. And content marketers are in a similar spot. AI is accelerating how fast, you know, teams can generate variations, and that volume just keeps going up. But there’s no real time feedback loop telling them what’s working and what is not working. And the creators have zero visibility into how their work performs. So you end up with this fragmented workflow where everyone is producing content, but nobody has the clarity to improve it with confidence. And this is exactly the gap that content analytics is designed to close, connecting content performance data back to the people who need it and across the entire lifecycle. And that’s what we’re going to walk through today. So while over 80% of marketers agree that creative drives performance, fewer than 60% share a clear definition of what great creative actually is. And that’s a problem we see all the time. So if teams can’t agree on what good looks like, they can’t measure it or optimize towards it. And without that shared definition, decisions default to intuition and opinion and not data.

Content analytics addresses this by surfacing content attributes that are tied to performance, which gives teams a shared data-driven language for what’s actually working. So Adobe frames the content supply chain around five building blocks. We have creation and production, asset management, workflow and planning, delivery and activation, and reporting and insights.

So you’ll notice that reporting and insights is highlighted here, and that’s where content analytics lives. And it’s the place that ties everything together. Without this insights layer, the other four blocks are operating without a feedback loop. And so you’re creating and managing and delivering content, but you’re not learning and optimizing from it. Content analytics closes that loop by connecting what you produce back to how it actually performs.

So what is content analytics? Content analytics enables brands to measure the creative assets and experiences across platforms, making content more measurable as we’ve talked about. And I think of this as a crawl, walk, run model for content measurement. And you start with the lowest hanging fruit content usage on the lower left of this slide. And a question you might have that you can answer would be, how many impressions has my creative served? Am I under or over utilizing an asset? And then next is content engagement, which has questions like which content has the user actually seen and is it effective? Are they clicking? What’s the click through rate for certain attributes? Then you move up to content and journey. And this is where it gets powerful. Now you’re tying content to conversions, not just engagement. So are they converting? Do they come back to purchase? And then the top of the model is the content and personalization. So brands that can tie content to the customer and their journey can use AI to surface which themes and attributes are resonating with specific audiences on specific channels. And that progression from what’s getting impressions to are we delivering the right content to the right audience? That’s the value arrow in the middle. And Content Analytics gives you the data foundation to move up this model.

So let’s talk briefly about what data you get with Adobe Content Analytics. Content Analytics has two building blocks that put structure to previously unstructured content across experiences and assets and experiences the text on a given page, the content text, and that’s identified by unique experience ID based on the URL and the version. And then an asset is any image on that page. And even if that same image appears at different URLs across your website, it gets grouped under a single perception ID to consolidate that together. And AI is automatically generating the attributes for both of these. For assets, that’s the visual stuff on the right of the slide, the objects, the colors, the photography and then for experiences, it’s the text attributes on the left, readability, word count, tone, density, persuasion strategy. On the metrics side in CGA, in the workspace, you get views, clicks, click through rates. Plus you can use calculated metrics like asset revenue share views per visitor, different things like that. Prithi later on, we’ll be showing what this looks like inside CGA workspace a bit later in our webinar. So now let’s shift gears into how to get Content Analytics set up. Let’s look at how this actually works under the hood. Here shortly, I’ll show you the steps which are very simple, especially if you’ve already implemented CGA with the AAP web SDK. But high level, this is going to leverage the AAP web SDK via an extension to capture this data. And this data gets sent to an AP data set. There’s processing that identifies the content, derives attributes via AI. I know I said a lot there, but I’m going to show this in action here shortly. Okay, so let’s see how easy this is. The setup of Content Analytics is done via a guided onboarding wizard in CGA. And there’s three things to take away before we jump into the live demo of the setup. First, it’s built on that AAP web SDK. Second, it works with any CMS or digital asset management platform. And third, AI handles the heavy lifting on metadata and attribution that we discussed earlier. So let’s switch over to the live environment. And I’ll walk you through the setup and show you how easy this really is.

All right, so we are now in the CGA workspace. This is just a sandbox environment that I’ll be showing and doing a live demo of how to implement Content Analytics. So we’re going to go up to Data Management inside CGA, and we’re going to go to Content Analytics Configuration.

As you can see here, once this loads, you’re going to see a few existing configurations I have, but today we’ll walk through a new one from scratch. So to do that, we’re going to click on Create Configuration at the top right. And this is that guided onboarding wizard that I mentioned. And while there’s five sections on the left here that we’re going to go through, this is much simpler than meets the eye. So let’s just dive in. So first, we need to populate a name for this. So for this one, we’re just going to use a fictional brand called Bright Step as a fictional shoe retailer. All right. And then from there, we’re going to select the data view that this needs to be bound to. So this is the data view in CGA that this new configuration is going to be mapped to. And so for this one, we’re going to select a data view, and we’re going to select the Bright Step retail view.

And this is important that when we’re selecting the data view here, just a note that this is going to update this data view that I’m selecting with to provision the Content Analytics metrics and dimensions that are part of this. Okay, so we’ve now selected the data view. And then next, we’re going to enable the collection of experiences, which is that text content on the website. And so there’s a setting here that exposes the configuration to input a domain regular expression and the query parameters lists to narrow which content should be included and so forth and how your content is structured. But so for our demo today, we’ll put in a domain regular expression to match our main store and website of a fictional site. So I’m just going to copy that and paste this in. Okay, so we have a regular expression here that matches a fictional domain that doesn’t actually exist. And for the query parameters, I’m going to type in the two that I want here, which is maybe we have a category parameter, and we have a collection parameter. Okay, so that’s the first step for you know, the experiences. Next, we move to data collection in the guided setup here. And we’re going to choose an existing Adobe tags property. So for this one, we’re going to choose our right step demo Adobe tags property. And the next step in the data collection in the setup here is to specify the pages to include or exclude again using regex. And here I’m going to paste in a negative look ahead regex that will exclude our checkout account and cart pages. So bear with me while I copy this. So this is a negative look ahead, you know, in regular expression format when you have that question mark and then the exclamation that starts to build the exclusion by doing so. And then we for assets, we’ll put in a negative look ahead regex to exclude our logo because the logo exists across all pages. So maybe we don’t want that to be included as an asset. So again, bright step logo dot PNG, we’re just going to exclude that. Now, once we have this configured, the last and final section of the wizard is to optionally configure header overrides. So if you have bot protection in place on your website, this is where you would configure the headers so that the Adobe bot that needs to reach your site to analyze content and assets and do all the AI magic behind that so that it’s able to do so. So you would click on configure header overrides, put in a header name and header value. But this is an optional step depending on what you’re doing here. So at this point, you know, before we click the implement button on the top right, there’s a helpful note at the bottom here that says exactly that what the wizard is going to do behind the scenes. I’m actually going to show you a lot of these things as we step through the demo today. But let’s go ahead and click on implement. And we’re going to click on Continue. Oh, we got a network error. This is the joy of a live demo, but I am prepared. We’ll click it one more time to see if it works. Nope. All right, I am prepared. We’ll show you the backup of what this looks like because I have an existing implementation. So bear with me. So over the first thing we’re going to do after you click implement, and that works, right, we’re going to go into a EP. And you know, for this, we’re going to look at the changes that happened. So for this, let me change my sandbox. And we’re going to go into a different sandbox here and we’re going to go into schemas. So in our schema list, you’ll see that what the wizard actually created behind the scenes is three different schemas. You have the experience attribute schema, the asset attribute schema, and the content analytics events schema. So these are all schemas. The content analytics one is a time series schema. Now, the other thing that the wizard did is it created data sets. So if we go to data sets here, and I already had it filtered for content, you’ll see a new data set that was created for content analytics for this bright step data set. But you’ll also see two more data sets, the asset attributes and experience attributes. Again, the wizard created all this for you. So this is what happens over in AP. Now, if we go look at the data collection screen, in the platform, you’ll see first that under data streams, we have a data stream that was automatically provisioned for content analytics events. And if we go into this one, you’ll see that it automatically set up a brand new data stream to receive the data from a content analytics perspective that will be mapped to Adobe tags that we’ll show here shortly. And if we edit this, go into this, you’ll see that it’s mapped to the data set that was automatically created. So the data streams automatically configured for you. Then if we go over to Adobe tags that are as our tag management tool, and we go into the into the property here, we’re going to see that an extension was added for Adobe content analytics. So this is automatically created for you. And then it also provisioned a rule called instantiate ACA library, Adobe content analytics library. We click into this rule, you’ll see that when the library core library loaded loads, it takes an action of instantiating content tracking. So this will all sit inside of Adobe tags, and you’ll need to publish this change to get this active and start collecting data and so forth. So then finally, if we go back over to CGA here, and we go into connections, you’ll see that in our in our demo here, and we get the right one. Here we go. Inside the connection, it automatically added in the data sets for content analytics, the asset attributes and experience those are lookup event data sets. And then content analytics here is an event data set that uses the same identity map that that has been configured. Now, the final piece that I want to show you the setup here is that in the actual data view configuration, if we go into the bright step, retail here, and actually, this might be the wrong one, let me filter. Opponents. The other thing that happened here is that we have metrics and dimensions that have been added in automatically into our to be available for our data view. So if we just filter this for content analytics, you’ll see that it shows all the metrics and dimensions that are available inside that data set. And so you’ll see things like experience clicks, experience views, asset clicks, on the dimension side, asset photography styles, lots of great information to help you with your analysis, and so forth. So you know, to summarize the actual setup of the, you know, the flow of content analytics, if you already have CGA implemented with that AEP Web SDK, it can be done in just a few simple steps because thankfully, Adobe does that heavy lifting for you by creating the schemas, the data sets, the extension configuration in Adobe tags. It also sets up the connection changes in CGA and provisions the dimensions and metrics in the CGA data view for you. So very simple to set up. And now I’ll hand things over to Prithvi to show us what this looks like in Workspace Reporting.

Thank you, Joe. Let me just get to sharing my screen. Good morning, everyone. Thank you for joining us today. And let’s take a look at what this looks like on the reporting front. So Joe talked about how you can see all this various data set up from a tracking perspective and how it will automatically flow into the system using the wizard. So as long as you kind of know the domain that you want to implement this in and some basic rules that you would want to set up and what you want to include and exclude in terms of collection, the setup is really quick and automated. And the tags are automatically deployed and it just needs approvals on your side to make sure everything is working. The data flows all the way over to a data view. And here in Workspace now you can start using that data view and that data for analysis. So here I have a fictional outdoor clothing store type of data here available. And I have this project already created. There is a content analytics template available as well that you can use, which auto creates all of these tables and visualizations for you based on some questions that we think teams commonly ask about content performance. So you don’t even have to create this manually. You have a template available that you can use to get started as well. So first up here, I’m just showing assets by conversion. Like we talked about before, assets are atomic images that are on your web or other channels that we support data. So these are the specific images that I have. And here the conversion metric that I’ve taken is a classic retail metric of orders. I’ll talk a little bit about what kind of conversion metrics we can support, but just know that any funnel conversion metric that you’re tracking already in your CJA data can be connected back to the content that you are measuring. So for B2C if it’s orders and add to carts and subscriptions, and for B2B if you have things like signups or form fills or request for demo type of metrics that you have on your website that you’re collecting, you would be able to tie back content to those metrics. So here I’m connecting it back to total orders. I do have attribution applied on it to make sure that I’m attributing the last touch image back to total orders. You have available a lot of out-of-the-box attribution models in this workspace, and so you can choose to pick any of them that’s available out of the box. So here I have my top performing asset IDs, you know, buy, orders. Each of them do have a thumbnail right next to it just so you can click into this little info icon and view a larger image because sometimes the asset ID that comes in is just a URL like this. And not everybody knows the thousands of URLs that’s deployed on your website. So we do have available a thumbnail that you can look into, and when you click on the info icon you’ll see that there’s some information available here. Obviously, you know, how many impressions have we seen of this asset of all times? So since we started collecting the data, how many impressions have we seen just of this asset? How many experiences does this exist in? So Joe talked about the concept of experiences where, you know, it’s a collection of images and text and CTA. So how many of those did this asset specifically appear in? What we saw. And then when was the first time we saw this image and what was the most recent impression? And then the attributes. So these are the Gen AI derived attributes for this specific image that are also available here as a quick visual. So this content model is just a really quick way for you to be able to see what are some basic metadata that you need to know about your asset if you’re interested in digging into data. So let’s keep going. Now I’m interested in asset decay. Like we all know assets have a freshness to it, a timeline to it, right? At what point, how long can assets really be live and still give us positive returns? It’s something that a lot of teams are interested in. So what I’ve built out here as a table is just a comparison of asset clicks compared to from February to March, since I have whole month data there. Right? And so I’m trying to see, hey, which of these actually declined when I compare those two time periods versus which of these had a positive lift from a CTR perspective? So asset views and asset CTRs are medics that you’ll just get as part of the data collection. It’s not something that you have to build out, but I have built a time comparison here. So very quickly I can see, hey, these three assets, month over month, there is a decline in CTR. So maybe something that I need to consider looking into. These are also trending towards a decline. They’re pretty much on the border. So that’s something for me to address as well. So depending on how much content is refreshed on your properties, you might want to consider a time comparison accordingly. For some it is weeks, for others it is months. So you can compare and contrast and see at what point are you starting to see negative returns on your assets. So you get a sense of how frequently you have to refresh it or how long it can live giving you a positive impact on your CTR or conversions. So again, I chose to do this with CTR, but you could totally do it with conversions or any of your other conversion metric too.

Another question that we get asked is like, hey, are there assets on my website that are just overutilized? Are they appearing too much, too little? I created a calculator metric here called order efficiency. This is just orders per thousand views. So this is a calculator metric that I created just to get a sense of how these assets are performing against that. So again, very quickly I’m able to see how the performance is against order efficiency as well. And if I want, I can always dig in and say, let’s say I’m interested in this one. And I want to get a sense of like, hey, how many experiences does it actually appear in? These are the experiences. I can very quickly click on that little breakdown icon there and it’ll break it down and show me what are the experiences that it actually showed up in. So this is in all time we saw 40 experiences that it showed up in for this time range. Specifically, I’m seeing that it showed up in 19 separate experiences. So again, very quickly I’m able to take a look at what are the different experiences that they showed up in as well. And maybe this is a moment for me to take a look at all of these pages and say, does this image really need to appear in all of these? Can I reduce that exposure so that it is not utilized too much or users are not seeing it too frequently on my website? So that’s something for me to take a look at as well. This is also a really good way if you have introduced some new icons or new messaging in your website and there’s a time to take it down, right? It’s good for you to kind of use this breakdown method to also say, well, hey, how many experiences is this still live in? Are there stragglers that are left out that I probably need to consider? So it’s a good quick way to do a check on any changes in your website that have to happen. I like scatter plots to look at kind of outliers. So this is just a classic CTR versus my order efficiency calculator metric. I’m just looking for any assets that might have given me high order efficiency, maybe with a high CTR or medium CTR as well. So that’s something that I get to look at. There’s not too many crazy outliers here given it’s a demo data, but you might start seeing as you run this on your data as well. You’ll see in most of these visualizations that they come with thumbnails as well. Just so again, you’re not just looking at those asset IDs, but you actually have a visual reference back to your assets and experiences as well.

So I’m looking at content fatigue now. Again, a good way for me to see that is like, hey, how many experiences does this really appear in? Is there a point at which it is too many? I have one here that’s showing up in 20 different experiences and order efficiency is decently high. So I kind of want to see, well, what are those experiences that it showed up in? And again, I’m just going to quickly do that breakdown and it automatically pulls up all of the different experiences here as well that I can see. I do have some mobile data coming into this as well since we support mobile. So that’s what you were seeing there. Now let’s go to attributes. Now are those generative AI derived elements. There are textual descriptions of your assets or experiences. For assets, your attributes are going to be more about the visual content and for experiences, your attributes are going to be about the text content. So I have these as the top converting attributes. You do have asset attributes as a dimension that comes in and it is what we call a mixed dimension. The first part here that you see before the colon is the category. So this is asset scenes and then the second part after the colon here, what you see is the actual value. So there is a ramp, there is a waterfall, there’s a road. These are different asset attributes and these are the top ones that show up against order efficiency for me. I also have tags and asset objects. Now what I’ve done here is I’m trying to look at what kind of asset tags and asset objects perform or how they form against two of my most important segments that I have. So all this while we were looking at just how assets are performing against conversions but now I’m starting to look at personalization opportunities. So if you remember the value hierarchy that Joe was talking about at the top is, you know, are there personalization opportunities that I can take advantage of because of this understanding I have of not just what content is performing well but what about that content is resonating with which part of my segment to drive a certain success metric. So here I have point of sale orders and high value customers as two of my segments that I’m interested in, right? And straight out of the gate, especially with this chart, I can see that there’s a slight difference in kind of what works for these two groups specifically. So the orange bar is one of my segments, the point of sale orders, and then, you know, the purple is my high value customers. So I can see very quickly that there are some that they’re closely tagged on. So photography seems to commonly work across these, any of my assets that are tagged with that. But if I take into, you know, snowboarding, that seems to work really well with my point of sale customers. But for my high value customers, customers who purchase over a certain amount very regularly, that is not a high performing tag in and of itself. And similarly, I can also look for objects. So there’s a huge difference that I can see in how bonfire images are resonating with my two different segments as well. So again, bonfire seems to work really well for one segment but not so great for another. And so I might be able to find what are different elements that resonate with these groups and look for personalization opportunities there. So if I am running a re-engagement campaign next, or if I’m going to target them with some sort of an offer, I know what kind of attributes work for them. And so I can talk to my creative team and say, hey, can we get some new creatives spun up or look at existing creatives that we have that have not been used, that actually have these attributes in them and be able to use it. Now, I have just taken three out of almost 40 attributes that you would get as part of content analytics, tags, objects, and scenes. There’s a ton more in terms of color palette and photography style and whether there are people in the images or buildings in the images or animals in the images. There’s a lot more different categories that are available and each have values behind them as well that are identified when we see them. And so you can pick and choose those that work really well for your industry and what kind of images that you generally have in your data set. So that kind of brings to close my demo that I wanted to show, specifically just focused on workspace and what kind of analysis that you will be able to see with this data that comes through.

I want to talk a little bit about success metrics overall as well. So you can connect content analytics to any part of that conversion funnel. So again, whether that is in your awareness stage or your engagement or your consideration stage, if you’re collecting those metrics, if they’re coming in as metrics into your data, you will be able to attribute content and experience against that kind of performance. And you can also pick and choose what kind of attribution model that you want to use to address those metrics. And lastly here, I just want to show what’s coming up soon as well. We do have content analytics available right now for web. The reporting that you saw is classic analysis workspace reporting. We have reporting enhancements that are coming that make it a little bit more easier for personas who may not be as familiar with analysis workspace. You will see AI automatically generating and helping you identify some of these insights a little bit more. So that’s coming soon. We do have support for mobile app as well. That’s in beta now. That’s a little bit of the data that you saw coming through as well. We have some of our customers testing this out in a beta form and we will be launching this as generally available very soon as well. Paid media measurement, we know a big amount of the budget for content creation, content production does go into paid media, specifically social channels. And so we will start supporting analysis against those social channels as well, starting with meta and Google. So we will have that coming in the beta if that is starting very soon. So if you’re interested, please do reach out to your account teams. And then lastly, short form video measurements. So again, video content production is expensive and a lot of customers do want to understand how video is performing against their ROI of this video production that I’m spending on. And so for short form videos, specifically those that you would see in the social media channels, we would be supporting those in content analytics as well. And so these are all coming up really soon as we’re working through those on our roadmap. I will now hand it over to Danielle. Great. Thank you so much, Preeti and Joe. That was excellent. Okay. So I’m going to talk a little bit about some events and resources that are coming up. If you are attending Summit, we have a couple of events that I’d like to highlight here. On Monday, there’s going to be a hands-on lab. So just like you just saw from Joe and Preeti, they’re going to be walking through how to get started, get implemented with content analytics and other things. So definitely recommend that if you scan the QR code with your phone, it’ll take you directly to the page, the registration page on your Summit site to sign up for it. The other event we have is on Wednesday. So we’re going to be doing a content analytics session with Wyndham Hotels. They are a customer speaker that will be talking about how they’re doing the mobile app beta, what kind of insights they’re deriving from content analytics. It’s going to be really interesting. So if you’re in attendance at Summit, scan these QR codes. We’d really love to see you there in attendance to learn more about content analytics. And if you’re not going to Summit, we have plenty of resources available as well. You can reach out to your account teams if you want more information, but we also have some other resources available here. As I mentioned earlier, BlastEx has an excellent blog that was created for content analytics, talking about AI and measurement. We also have a use case guide on adobe.com where you can understand use cases for content analytics. There’s experience league where you can find all of the information you want. And then also we have our YouTube channel. So if you’re familiar with our YouTube analytics channel, there are specific videos for content analytics that you can check out, easy to digest snippets. So with that, I think we have a few minutes left. I know there are a couple of questions in the chat pod, more specific to the implementation and onboarding part. Do we want to take questions or? Yeah, we can maybe wrap up with a couple. So I think there’s a question on is this page include or exclude referral URL? I’m assuming maybe this was a question in reference to one of the data collection pages, the setup pages. If you do want to exclude a particular referral URL that’s coming into your website, you can include that as well. So I think you would include that as a query param and then it would automatically just exclude that.

And then the second question is on can we use the existing data stream and tags properly while configuring content analytics? We do have to create a new data stream to configure content analytics and it will use the existing Web SDK instance that you have. I know we talked a lot about Web SDK, but just as another side note, if you do use another tag manager, DIY or another commercially available tags manager, we do plan on supporting that as well. In about a month, that is something that we will be releasing as well. This is not limited just to Web SDK usage.

That’s it. Great. Thank you so much for your time. Thank you. Thank you both, Joe and Prithvi. That was excellent. Really informative. Thank you. Thank you. Thank you for joining.