 Hey everyone, and welcome to this CUBE Conversation featuring RudderStack. I'm your host, Lisa Martin, and today we are very excited to be joined by Eric Amuega, VP of Marketing and Operations at RudderStack. Eric, thank you so much for joining us today. Thank you for having me, Lisa. Tell us a little bit about RudderStack. I see from the home page, RudderStack makes it easy to collect and send customer data to the tools and the teams that need it. But give us the real story here. Fantastic. So RudderStack is a warehouse native customer data platform that enables companies to collect, to unify and activate their customer data. Right? I'll talk more about what each of those pieces mean. Warehouse native, what that means is that instead of storing data, RudderStack actually turns a customer's data warehouse or data lake into a customer data platform and handles all of the collection, modeling, i.e. unification and the activation of that data. Nice. So talk to me about how RudderStack enables customers to handle things like event data specifically. As we all know, we create all these events with apps and things like that and we want this experience to be like real-time. How do you guys enable customers to deal with event data? Absolutely. So when the company started, the first product that we built was our event stream pipeline. And the event stream pipeline, what it does is that it collects customer data from your website or your app, right? So each interaction that a customer has with your web property, with your app property, is an event that's being created. Now, after collection, instead of us storing the data, which is what traditional CDPs do, we actually send that data in real-time to downstream destinations. So for example, if you want to activate real-time use cases, that data is collected and set in real-time to downstream destinations. So downstream destination could be an email marketing platform. It could be, you know, a Facebook ads platform. It could be any sort of downstream tool that business teams use to activate customer data. In addition to that, we also have the ability to send that data into your warehouse or data lake, whether it's AWS S3 or Redshift. And we expose that data for you to be able to model that data to build whether it's ML models, whether it's, you know, AI use cases, you can do all of that modeling in your warehouse and then use our reverse ETL pipeline to activate that data as well. So we have both the real-time pipeline to downstream destinations, but then also sending that data to your warehouse or data lake where you can do additional modeling and create more value from your customer data and activate it using our reverse ETL pipeline. Is that all done simultaneously? So the best way to think about it is there's both real-time and batch use cases. So real-time is done in real-time, right, with very limited lag. This batch use case is typically when you're sending data, for example, to a data warehouse. You probably don't want to do that in real-time. You can configure it down to five minutes, but for warehouse or data lake use cases, typically, you know, you're okay with a little bit of lag. You can configure it down to five, 15, 20, 30 minutes. And because when you're doing the modeling, chances are you want to run, you may want to run the data in batch, right? You're doing modeling overnight, creating those audiences, creating those user features, and then activating those downstream. But then there are also additional real-time use cases that don't involve just the event stream pipeline, sending data from web or app to downstream destinations. You may also want to activate real-time use cases, for example, personalization. When a customer is interacting with the app or website, already having pre-computed features, a feature could be something like a customer likelihood of churning, right? If you have a churn score and you have that customer who's been flagged as having a high likelihood to churn, you could actually send a notification in real-time. You know, maybe it's a 20% of coupon in real-time and the customer is engaging with the app. So that's an example of a real-time use case that Radistar can activate. Real-time is one of those interesting use cases, Eric, that we've been talking about a lot for the last few years. It's something that became no longer a nice to have for organizations in every industry as consumers, customers, business, personal, have this expectation that we're going to have a real-time experience. And to your point, it's going to be personalized. The expectation of personalization is only increasing. Talk a little bit more, expand on the personalization. You talked about a real-time use case, but enabling your customers to deliver that personalized experience, their customers are demanding 24 by 7. Exactly. And actually, I was looking at a study the other day by McKinsey. Basically, they looked at what value does personalization drive in terms of top line? It actually drives 15% to 25% in terms of revenue left. That's revenue per customer, right? And that goes to your point. Customers increasingly are looking for personalized experiences to engender greater brand loyalty. And it's something that you have to provide, especially if you're working with B2C-type brands. So that's the first thing. So this is an important use case that a lot of B2C companies need to unlock to win market share, to drive revenue growth. Specifically, how we enable that. We talked about the real-time activation using our event stream pipeline, sending that to downstream destinations. That's kind of like the most basic level, right? Where you're just collecting data and based off of customer activity within a certain browsing session, you can tailor that experience to the customer. But then when you start to think about the more advanced customers get, for example, they want to calculate machine learning features in real-time. There's kind of a number of different steps. The first step is the data collection. There's a number of different data sources for customer data. We have talked about the event stream, which is data from your website or your app. That's event stream data. You also have SAS data that lives within other SAS tools that you're using, whether it's a CRM platform, for example, Salesforce or customer service platform, for example, Zendesk, right? So you have two types of data, behavioral data and relational data. You collect that into your data lake or your warehouse, that's step one data collection. The second one is unification, right? You need to resolve the identity of that data so that you can say this specific customer, this is their behavioral data, right? And this is the additional data we have on them. For example, how much they've spent on the platform, right? The last time they logged on to your app or your website. Once you have that rich customer profile, you can now calculate, you can enable features and calculation of those features. A feature could be what I talked about, revenue over the last six months, last time I used a logged in. You calculate those values, right? So now in your warehouse, you have a user, the entire customer journey from behavioral data and additional features that could either be deterministic or probabilistic. And then now you can unlock true personalization where you're using all of the data that you have on the customer to deliver the experience that they need to make that next purchase. So the warehouse native CDP you said helps data teams collect data, unify it, activate it. You talked about some of the key features. Talk a little bit about your warehouse native CDP in terms of differentiators versus competition. Absolutely. So I'll be the first to admit that the customer data platform landscape is extremely crowded and it's very difficult to figure out who's within the customer data space. The way we like to think about it right is like there's three primary types of customer data platforms. You've got your traditional customer data platforms. You've got a new entrance. It's called Composable Customer Data Platform. And you've got the warehouse native customer data platform of which we're the leading warehouse native CDP. And I'll talk about each of those. A package CDP typically, traditional CDP typically, they collect data, they store data in their own infrastructure. They hide it behind a black box and you have a UI that you engage to, for example, build audiences and then activate. So what ends up happening with a traditional CDP, you basically have a parallel architecture that's completely separate from where the rest of your customer data lives. That is your warehouse. That is your data lake. That's your Redshift. That's your S3. Now you have this parallel architecture. And what that does is that you essentially end up creating two problems. One, you've got data silos, right? You've got data sitting in your warehouse. And then you've got data sitting in SAS black box. That's a traditional CDP. That's the first problem. You've created data silos where you may have different versions of the same data. Second problem is that you're paying for storage twice, right? You're paying for storage in your own warehouse where you have your data anywhere that you're using to run your business. And then you're paying another SAS platform to store your data as well, right? And then obviously you're locked into whatever features the traditional CDP has. If they don't have real-time activation, you just have to use whatever features that they have. That's the traditional CDPs. They've been around since 2012. They're about the second one is composable CDP. This is a new model, which is kind of, you know, cobbled together a number of different SAS solutions to kind of build the CDP, right? So you have somebody doing a vendor for event stream. You have a vendor doing extract, transform load into your warehouse. And then you have one of the traditional, one of the reverse ETL players activating that data. The challenge with that is that you actually have three different vendors that you have to manage. The data schemas are different depending on the data source. And you're having to do a ton of work to integrate all of those tools together to ensure that you have end-to-end data that you trust. That can be extremely costly from a resource standpoint. Rattastack is a warehouse-native CDP, right? And the question our founders asked themselves when they were building Rattastack back in 2019, 2020, is if you were to build a best-in-class CDP with the evolution technology since the CDP, the first CDP tools were put in place, how would you do it? The first answer was obvious, right? You've got a proliferation of data warehouse and data leaks. You know, there's the decoupling of storage and compute. They have a ton of security features and they've kind of become the gravity where all of operational data is moving to. So the question was how the challenge for our founders was can you build a CDP on top of a customer's warehouse to avoid the data silo issue to reduce the cost from having to store data into different places, and hence Rattastack was born, right? So that is the warehouse CDP. Now, the advantage that we have is that we actually have the end-to-end feature set. We've got the data collection with our, what are considered best-in-class events to pipeline for behavioral data. We have our ETL pipeline, specifically focused on customer data. That's data collection. We also have a product in beta right now called profiles that enables the unification so unifying all of your behavioral data and transactional data or relational data in the warehouse. And then we also have reverse ETL to activate that data. So it's kind of the end-to-end story with data collection, unification, and activation, which are really the three components that you need for a customer data platform. Right, and you did a great job really describing the differences between traditional CDPs and what Rattastack has built. It's almost as if they deconstructed the traditional CDP and understood some of the challenges. So you must have a favorite customer story. You run marketing that really articulates the value. And some of the things I was hearing as you were describing the differentiators is there's got to be massive cost savings, ROI, maybe TCO reduction. Tell me your favorite customer story that you really think articulates to Rattastack's clear value. Absolutely. I would say, and this one is not public yet, but we work with a leading shoe retailer. And initially when we engage with them, they had a team of engineers that were working to build connectors for their ad platform. So essentially figuring out how to ingest data from Facebook ads, from Instagram ads into the warehouse doing modeling on that data to identify, to model the performance and analyze the performance of the ad campaigns. And then based off of that sending data out back to the ad platforms to run new campaigns. And we had an interesting model with them. We essentially had a POC, a proof of concept where we had a number of different yardsticks that we had to hit. And one of the things they wanted to do, they wanted to prove actual value from the product. So it's a paid POC over a period of three months to prove the value of Rattastack. And they ran the POC for three months and they saw a 400% increase in return on ad spend by using Rattastack. And the reason we're able to unlock that is, one, we had native out-of-the-box connectors for most of the major ad platforms like Facebook ads, Google ads, ETC. We exposed the data for them to model, for modeling and analytics in their warehouse. And then we also had the reverse CTL pipeline to actually send the results of those campaigns down to the ad platforms. And so you have that essential loop in terms of we're running a set of ad campaigns we are pulling that data into the warehouse. We're doing identity resolution in terms of who isn't interacting with those apps, unifying that data and then figuring out the ROI and then investing in the best performing ads. So that was just a simple story where in three months we're able to, one, unlock, basically enable them to get data flowing across the entire data stack, which is something that a team of five engineers have struggled with over a year to do. We're able to do that in three months. And then two, time to value, actually demonstrating the fact that we could have a positive ROI in a very short time period. So 400% return on outspend is an incredible start. That's nothing to scoff at, especially in this environment when there's a lot of pressure on ad budgets, being able to unlock significant return on outspend is really invaluable. Wow, what a substantial customer story. I'm sure that was one of many that you had that really shows the value and how you're able to show customers ROI that quickly, especially in today's climate, incredibly important. So you guys are a young company founded back in 2019 and 2020 or so, what's next? Here we are almost halfway through calendar year 23. What are some of the things that we can expect if we keep our eyes on Matterstack? Absolutely. So I think the vision has been clear from the very beginning. As we talked about the data activation lifecycle, the goal is to continue to beef up each and every step of that. We have a best in class data collection with an event stream pipeline. We have our unification product called profiles that is in close beta right now. And then we have a reverse ETL product. Once all of those three are live over the next few months, you essentially have the end-to-end warehouse native customer data platform that will enable our customers to unlock similar value and even more value similar to the case study that I mentioned earlier. So you'll continue to see continued innovation across the data activation lifecycle. And then there's two areas where we're going to focus on from a feature standpoint. One is data compliance. You're seeing increasing activity from regulators, not just in EMEA, everybody talks about EMEA, but also in the United States, California with CCPA. So we have an exciting set of data compliance features that we will be launching, that we have in place and we'll be launching and beefing up over the balance of the year. And data governance as well. So when you look at customer data is extremely important. It's the most valuable data that a company has. But then it's not, it shouldn't be, you shouldn't be hovering up all the data that you can get. You need to be intentional about it, that you're collecting the data that you want in the specifications that you want that data in to ensure that you can unlock value from that data downstream. So the three areas to look forward to over the next six months. One is excited about our unification product profiles. So more to come on that. And then we're going to have data compliance features and data governance features that will unlock our next phase of growth. And really as we continue to grow with large enterprises, these are features that increasingly becoming non-negotiable, that a lot of other CDPs cannot provide. Awesome. Eric, we better let you go. You must be really busy. We so appreciate you coming on theCUBE, talking about Ridershack's warehouse-native CDP. You did a great job of articulating what it is, its value, how it's different. We so appreciate you for coming on theCUBE. Eric, thank you. It's a pleasure talking with you. Absolutely. Thank you so much, Lisa. Appreciate your time as well. My pleasure. Keep it right here for more action on theCUBE, your leader in hybrid tech event coverage.