Salesforce, Inc. (CRM) Earnings Call Transcript & Summary

My name is Andrew Lee. I'm a Senior Product Marketing Manager here at Salesforce, focusing on our Salesforce Data Cloud product. I'm super excited to be here today to talk to you all about all things Data Cloud and how it can serve as the trusted data foundation for the entire business, helping you drive growth and build loyal customer relationships. Before I begin, if you've been to a Salesforce event in the past or been to one of our webinars or presentations, you'll recognize this slide, I'm sure everyone's favorite slide. But I just want to give a quick reminder that we are -- Salesforce is a publicly traded company. We might discuss some things today that are on the road map or a bit more forward-looking. So for any customers in attendance, please be sure to make all of your purchasing decisions based on products and services that are generally available in the market today. And now that we have that little housekeeping bit out of the way, just like to say a big thank you. Thanks for joining us. Thank you for your time today. Thank you for your business and partnership, collaboration. Thank you for inspiring us to innovate every single day. And today, I am joined by Sonia Lele, who is the Senior Director of Data Science here at Salesforce. We have a real treat for you today because Sonia is going to take us -- going to be taking a lot of the data and customer experience concepts that we talk about today, bringing them to life through the eyes of Salesforce and our own internal implementation of our products. As you can imagine, like many companies, Salesforce has customer data that lives in many different places, whether that's CRM, marketing engagement systems, ad tech, et cetera. Sonia will give us a deep dive into how Salesforce uses our Data Cloud product to harness all of that customer data, which then enables our teams to deliver more impactful marketing journeys, personalized marketing journeys, optimizing sales or service conversations, support cases and also just boosting productivity across all of our internal teams. So with that, let's go ahead and dive right in. I'm going to kick things off today, set the stage a little bit by posing a question. And that is what do all of these companies have in common, all of these logos that just popped up on the screen? The first thing you might think of is, oh, well, they're all popular technology companies, which is true, but the answer that we're looking for today is all of these companies have been able to harness their data to drive incredible growth, loyal customer relationships through, what we like to call, frictionless experiences. We all know that the most successful companies are data-driven. Data is what lets them be customer-centric. Personalized experiences create these frictionless experiences that we'll hear more about and also work smarter with AI, right? So augmenting and scaling the capabilities of their people and teams like they've never been able to before. And you've heard me say the word harness a bunch of times already today. If we think about the word harness here, this is easier said than done, right? Harnessing data to accomplish all of these things is difficult. It's hard to do for a variety of reasons, I have a few listed out here. The first one being data proliferation, it's an ongoing challenge. So the amount of data, whether that's structured or unstructured data that's generated by consumers like you and me, people and businesses is growing at speeds that we've never seen before. All of that data needs to be stored and managed in some way. And our research is also showing that 80% of modern enterprise data today is made up of unstructured data. And this is something that folks really haven't had to grapple with before, but think of things like a knowledge article, PDF file, maybe it's an audio file of a sales call. All of these things are being, like I said, generated at unprecedented speeds and we need somewhere to store and manage it and make it work for us in an impactful way. But more often than not, this data often lands in disconnected source systems. We call these data silos. It becomes difficult to access and maintain. At Salesforce we like to call this trapped data. So this is data that someone like a marketer or a sales or service rep can't readily access and use. And because of this, there are often dependencies on IT teams, analytics teams, more technical folks, dependencies on them to share or, I guess, extract that data first and then share it with these different business-facing teams to power different processes or workflows and all of this really creates these really sort of detrimental bottlenecks and can drastically reduce speed to market. For myself as a marketer, coming from a marketing lens, we think about marketing teams, it can take weeks just to get the data that they need to create a segment for a campaign, launch the campaign and then get the insights back that they need to understand if a campaign performed well. So being able to close that loop can take weeks. And then lastly, the last one that we have on the screen here, constantly changing data privacy regulations is a challenge, right? So this is your legislation like GDPR, CCPA, keep hearing about the demise of the third-party cookie, deprecation of third-party cookies. All of these things are really forcing brands to rethink how they're acquiring, engaging and retaining the customers that they've already spent a lot of money on acquiring. So a lot of things to kind of focus on from a strategy standpoint. And I think the big takeaway here, the main takeaway here is that companies that aren't able to harness their data in AI, there's that word again, harness, if you can't do this, companies won't survive, right? So how do you make sure that your company is one of the success stories. We talked about these different data silos, these disconnected source systems trap data. If we could centralize data, that might solve the problem. But it's only the first step, really. So centralizing data is the first step, but it's not enough. Marketers like myself or any team in an organization that needs to harness data and be data-driven, they need to be able to do 2 things, right? They need to be able to access and action that data, unified customer profile data in order to drive that growth and increase that customer lifetime value that we have alluded to. And that's why myself, Sonia and all of Salesforce are very excited to introduce the Data Cloud product. Data Cloud, we like to say is the trusted data foundation for the entire business, build for customer relationships not just marketing, not just CDP use cases. It's not just about bringing data together, for bringing the whole org together around customers and being customer-centric and data-driven from that angle. And all of this is accomplished through an operational customer profile, real-time data access for marketers and other frontline teams and as well as seamless activation to any channel. So if we take a second to really break these down one by one, starting with an operational customer profile, what do we mean by this term? Might be a new term for some folks. But unlike a lot of data solutions that you might see out in the market, Salesforce allows you to segment and action on your data, so you can target the right audiences on the right channel with the right message at the right time, and operational customer profile enables you to harness all your data from any source across the business and lets you gain insight into customers and increase their lifetime values. You can think about it as maybe a single source of truth for your customer, a single view of the customer, but accessible, actionable, this profile can travel across your suite of apps and teams. It's all shared data, powering common processes. An example of this could be ingesting connecting e-mail engagement data from maybe your marketing engagement system, service data from a solution like Service Cloud and then audio files, that unstructured data, like we talked about earlier from external systems via Amazon S3 buckets or MuleSoft, tying all of this engagement data to a single individual. The second benefit here is real-time data access by marketers and others, really about increasing that speed to market that we talked about earlier while you have that trusted data that you need to create audiences, power AI and analyze performance at your fingertips, right? So in just a few clicks, right, Data Cloud lets you analyze historical marketing performance insights, then you could take these and then use them to create a new, let's say, a high lifetime value segment using either generative AI or a drag-and-drop visual builder, no need for SQL here, and then ultimately being able to target that segments with one-to-one personalized experiences, whether that's on web, mobile, SMS, et cetera, which is kind of a good segue into the last benefit pillar here. We have seamless activation to any channel, right? So this is being able to personalize every touch point across the customer relationship with AI recommendations, decisioning, journey automation. So using all that customer data that you have from across the business, powering these types of recommendations, powering that AI, maybe it's for a new product launch that you have, and you can use web behavior or past purchase data, using it to feed your different AI processes, really understanding the best time to deliver, let's say, an automated journey via e-mail or mobile push, et cetera. Maybe you're giving a personalized discounts on these new products, items that you are launching and based on everything that you know about the customer and when they're most likely to engage. And if we think back to the beginning of today's session, we talked about how the Amazons, the Googles, the Netflix of the world are known for providing frictionless experiences. And that really means every single interaction is impactful and relevant, meets the customer where they are, where they want to be, also at a time that's convenient for them. I believe a company like Tesla, you can go about the whole buying cycle for a new Model Y or whatever it might be without even talking to a human, if you choose not to. So that's what we really mean by frictionless here. So in these types of scenarios, every single moment counts across marketing, sales, service, commerce teams, et cetera, the entire organization, like we talked about, coming together around these experiences. But I think the question then becomes how do you make every moment count? How do you start kind of actioning or thinking about a strategy to do this? And it really comes down to 3 things or a 3-pronged strategy, so to speak. And that's, one being your data strategy, so it's connecting, harmonizing, unifying your data, your AI strategy, so analyzing, predicting, generating outputs with AI, augmenting your people and your teams like we talked about before. And finally, your activation strategy, right? So it's enriching, activating data to personalize every touch point, reaching your customers in a frictionless way based on what we know about them. And Data Cloud can really serve as the foundation for all 3 of these things across the entire customer relationship. And we can use this diagram that we have up on the screen here as a sort of mental model or road map for how everything comes together. Everything starts with data. So we start on the left-hand side of the diagram here, everything that we're trying to do here when it comes to creating these frictionless experiences starts with data and it starts with maybe anonymous data. If a person is clicking on an ad or browsing your website, every single kind of subsequent engagement from here helps you build and gather more information about that person, eventually helping them to convert, become a customer and then keeping them around for the long haul as well, too. So different types of data that you might start to capture and unify, identity data, e-mails, phone numbers, past purchase, past events or behaviors, real-time data like someone visiting your website, watching a video. Again, unstructured data is a major focus for a lot of folks these days, so think of audio files of the sales call that we talked about, and then transactional, operational business data, think inventory or like very granular engagement data like clicks from a mobile app or something like that. The takeaway here is you want to be able to stitch together all of this anonymous and known data so that you have that complete context of every customer's journey with a brand, and then you can take action based on that. And this is where we can start to help, too, with Data Cloud. So we make it easily to connect all of this data from Salesforce apps, but also external data lakes, other external sources, unifying it into that operational customer profile that we talked about before. And once you've unified all of your data, you can move into the middle section of our visual here, starting to leverage AI to make sense of all that data and then help you make every moment across the journey more impactful and also help your teams become and stay more productive. So that starts with gathering insights that you need to make decisions faster, inform your strategies, these insights can take shape through different data visualizations in the platform, AI-powered recommendations, all grounded in that customer, that trusted customer data that you've gathered and unified within the data cloud. And then from here, you can start to plan campaigns faster, more intelligently, you just have to generate content whether it's an image that will live in an e-mail, body copy, subject lines, all that kind of good stuff, now made faster and smarter by AI. And a few important things I want to mention here as it relates to AI and our AI innovations. The Data Cloud user is in the loop at every step of the way. There's always a human in the loop. We don't use your customer data to train our AI models and we have a 0 retention policy. So your data is safe and secure. And then you may be familiar with some of our predictive AI capabilities as well, too, in addition to generative. So it could be features like engagement scoring, send time optimization, next best action, next best offer, product recommendations, all of this baked into the platform as well. So now we can talk about activation. If we move to the far right of the diagram, thanks to Data Cloud being an organic product natively built on CRM you can seamlessly activate to all of our Salesforce apps, activating that data, using AI to personalize moments across whatever channel it might be, your usual marketing suspects, e-mail, ads, mobile, web, advertising, but also across social platforms, loyalty programs, sales, communications, service cases, et cetera. And then to the extreme far right and down across the bottom here, we can start to collect performance, customer data, revenue data, things like that, bringing it back into the platform, closing that loop so that we can understand how our activations are doing, how our customers are responding, using it to optimize our strategies going forward. Okay. Truly, that end-to-end solution from a data perspective, but I want to switch gears a little bit here. I hope these ideas have all resonated with you so far. But the power of Data Cloud really shines through when we talk about real-world business use cases and the value that comes about from there. So I want to dig into the so what now, how does Data Cloud solve for different jobs to be done? What value does it bring me and my business? We're going to take a look at this through the eyes of Customer Zero, which is Salesforce itself. And Sonia, who's joining me today, leads our teams that are building out and implementing these products and capabilities and to create these experiences for our own customers and doing this across the entire customer life cycle, right, sales, service and more, not just marketing. But enough from me. With that, I'll hand things off to Sonia to tell us about Salesforce on Salesforce.

Amazing. Thank you, Andrew. Hi, everyone. I'm Sonia Lele. I'm a Senior Director in the Marketing Decision Science Group, our internal marketing analytics group at Salesforce. And today, I'd love to talk to you about how we have implemented the Data Cloud, which we call it internally as our truth profile. I'm here at Salesforce and activated a whole bunch of use cases across sales and marketing teams and that I will talk about in more detail on the upcoming slides. All right, so what is our ultimate goal here? As analytics team, as a marketing group, we do want to build a 360 view of our Salesforce customer. We want to keep the customer at the center of everything we do. We want to make it easier for our customers to try, onboard and renew our products. We want to make it easier for our sales and marketing teams to drive more leads and revenue, and we're using this as an opportunity to be able to showcase for Salesforce on Salesforce. So when we build our truth profile, we have to be thinking of our customer journey. So I'm just going to give you an example here of a hypothetical Salesforce customer journey. Let's assume we have a customer called Samantha and let's assume Samantha is in the market for the Data Cloud product. She does a Google search for Data Cloud, see on Google, she's searching for Data Cloud, which takes her to our website. She does some preliminary research on the Salesforce Data Cloud and then maybe wants to get onto Trailhead, Salesforce's learning platform, and take a trail on Data Cloud on Trailhead. Let's say she's taken a trail on Trailhead, has come back to salesforce.com and engaged with a chatbot on our site to ask a few more questions about the product. And eventually, she's moving forward in her purchase journey and has decided to purchase the Data Cloud. Samantha is then in the Data Cloud implementation phase where she is in contact with the customer success group and maybe becomes active on the Salesforce admin community. And so what we're seeing here is that all of these touch points of Samantha's are being recorded in different systems of record. And the Data Cloud is really where we will be able to harmonize all this data as it will allow for the identity matching to happen across systems, so we have one single source of truth of Samantha. And this is what we've been able to do in our Data Cloud, which we call our truth profile. So this is what our Data Cloud or truth profile looks like at a glance. What you see in the middle of it here is a picture of the Data Cloud. Sorry, I'll just go back here. And what you see on the left-hand side is all of our sample customer, Samantha's touch points or what we consider data signals. So her completed trail, her chat with the chatbot on the website, maybe she's watched a demo video on Salesforce+. She's interacted with customer service. She's attended an event like Dreamforce. So these are all Samantha's touch points or signals and further on to the left, as we progress from left to right, we look at all the enterprise-wide data sources that capture these signals. And so what you see in the middle there is the Data Cloud. What you see is from the left is all the data that we have built into the Data Cloud and at the bottom, you see the different types of audience models we have built in there, such as enterprise, lead score and the product interest score. And on the right-hand side are the various activations we have launched across sales and marketing, such as the activations to our sales teams via Slack, some e-mail marketing activations as well as a website personalization campaigns that I'll spend some more time talking about. So again, at a glance, our truth profile, like this is how we think of it in numbers. And today, we have about 82-plus data strains that come into the Data Cloud where we work with about 3 billion records with a lot of the matching across data sources and identity resolution. We came up with about 208 million total profiles and then when we complete the identity resolution process, we have about 105 million unique profiles to work with. So I'll start off with some of the use cases that we have moved forward based on all the work that we've done in Data Cloud. The first one is based on a customer segment we built in the Data Cloud using attributes from our CRM system, which we call internally as Org62 as well as our website data. This use case originated from an insight we got from our sales teams, where our sales teams flagged the gap in their knowledge of what their customers do when they visit our website. So we took this problem and identified a strong data signal, such as maybe a visitation to a pricing page, that could be of value to a salesperson to take follow-up action on. So we built this dynamic segment in Data Cloud, which identified all contacts in the Salesforce and account list, we married that with website analytics data to create a segment of the contacts that have visited our website and specifically to a product pricing page in the last 24 hours. And we then built an alert mechanism via Slack to send our account executives an alert every time a key contact in their account visited a pricing page on our website. So we were able to scale this program. We started small with about 1,000 account executives. We've scaled it over time to about 4,500 account executives across the company. We've got some great feedback from our top sales reps who find this Data Cloud-generated activation extremely helpful in helping them close deals. And here's an example quote we got from one of our senior sales leaders which says, "Getting these notifications is extremely helpful. Knowing which of our customers are looking up pricing is great intelligence and certainly a buying signal." And this is what the Slack message specifically looked like. It is personalized to the account executive, it says, "Hi there. [ Hanks ]. The CMO of Northern Trail Outfitters," which is one of the top contacts identified in his account list, "visited the pricing page on Marketing Cloud in the last 24 hours." And so again, these type of alerts have been well received. They have high value, and it's an example of our impactful program we were able to drive from the Data Cloud. So our second use case I'll talk about is e-mail activation. And so for e-mails, we have 3 campaigns that we executed from Data Cloud segments over the past year. All these segments are what we call or what we consider internally as high-value segments. The first segment, we call the Salesforce+ [ World Cup ] e-mail segment, it was intended to drive high engagement from users who have visited and signed up to attend a virtual event or who have watched a video Salesforce+ which is our streaming platform that was launched during the pandemic where people can actually view our events virtually as well as a lot of streaming content. The second segment was based on a high-value attribute in our CRM system called a high license utilization account. So the goal here was to get contacts in those companies that have high license utilization of a Salesforce product, and allow them to go in and be able to self-serve to renew or increase their license usage. And so this was a second e-mail program that was activated with Data Cloud signals. And the third segment for e-mail was built on -- sorry, was built on an advanced audience model called the product interest score, where we were able to use several data points across systems in the Data Cloud, to use machine learning and AI techniques to generate a product interest score and then use our dynamic send engine to pick the right product messaging for users based on this product interest score. So these are 3 high-value Data Cloud-based activations for e-mail that we have executed in the past year, and we're actively measuring and testing them for continuous learning and optimization. I'll share here another illustration of our customer journey and our identity-based marketing strategy via what we call our unique ID or Trailblazer ID, also known as TBID, which is kind of the core tactic of our subscription-based marketing strategy. So this is an example of how our customer journey for Samantha plays out. Imagine she's searched for a video demo of Data Cloud, which has brought her to our website and say Salesforce+ specifically to watch a demo video. If she is a first-time and an unknown visitor, Samantha will be prompted to sign up for a TBID or a Trailblazer ID to watch this Data Cloud video that she has come to view. So this is a simple 2-step process where she signs up using her Google ID. And as she signs up, we also request of her some supplemental information such as what is her role type and what industry she's from. She is then authenticated and connected, and we now have this TBID subscription data that flows into the Data Cloud and gets modeled into the data model object. A Data Cloud segment is built, identifying in the case of the Salesforce+ e-mail, it identifies new subscribers over the past 2 days. In this case, we see that the size of the segment is about 12,900 people and then we activate this segment. The segment is activated via the data extension on Marketing Cloud to the standard MC Connector from Data Cloud. And once it's in Marketing Cloud, we can then build the e-mail journey and activate the e-mail. So this is an example of how we go about assigning a TBID or a subscription-based ID to a user and then using that TBID to do the identity resolution in Data Cloud, building segments after that and then activating those segments using Marketing Cloud for e-mail journeys and personalized e-mails. So the last use case I'll talk about is our website personalization use case. And so Salesforce.com, if you look at our website today, it has gone through a lot of digital transformation over the last 2 or 3 years. The entire website is now rebuilt to have sections of the site, which we call blades that are easily swappable and capable of pulling in dynamic and personalized content from a personalization engine like Marketing Cloud Personalization. And so in one of our first website personalization tests, we created macro segments in the Data Cloud based on our top B2B-like firmographic attributes such as by industry and we built 12 segments for our top industries like financial services, health care and life sciences, consumer retail goods, et cetera. So this is just an example of the types of personalization campaigns that we've launched on our website. And when the algorithms work at the back end of our Marketing Cloud Personalization, we have the ability to set these algorithms based on the type of use case and type of goals and KPIs we're trying to hit. But this is just some screenshots of what personalized content actually looks like on our website, when these algorithms are in action. What we see is for these industry-based segments that we pulled in from Data Cloud, even if there's an unknown visitor who's not associated with the TBID, if that unknown visitor has a match with the Data Cloud segment and we identify them as being someone from the financial services industry or health and life sciences industry, they will now start to see industry-relevant content on the website. So this is, again, some examples of how we're setting this up and what this looks like live on our website. And so for example, in this case, we're showing how we've set up campaigns to always have a control group and a personalized group. So in this case, this is an example of a control group, which is seeing static content. And so what you see over there under research, video series, webinar is a static page and then we compare this with our personalized group, which we are now enabling personalization on this page, which is the personalized page. And you'll see like based on, again, industry, the content on the right-hand side rail under guide article is now personalized based on algorithms running in the back end to the user that's come to the site. And so we're continuously testing control and test groups to see how much the personalized experience is improving the lift in our KPIs, things like engagement rates, but then also further down in the funnel are things like improved volume of leads, quality of leads and then lead conversion. And this is just a screenshot of a earlier iteration of our website, again, which shows personalized content running in the wild on the right-hand side rail where, again, these articles over here are pulling dynamic personalized content. And in every case of these, we've seen a lift in our personalized group, which I'll talk about in the end, when we -- on how we measure these campaigns and activations. So when we talk about personalization specifically, there are 3 essential components that needed to come together for our personalization strategies to work from Data Cloud segments. One is content, which is all the content on our website needs to be integrated, standardized, it needs to be tagged and curated in a way that makes it ready to be personalizable. Profiles, which is understanding who you are speaking to, spending time doing an audience discovery of your high-value customers. This is where we spend a lot of time in Data Cloud is really identifying our high-value segments. And then Algorithms, which is the decisioning engine that will determine what content to show the person at the right time. So I'll just touch on these briefly. But essentially, like content is all about getting this content ready to be personalized and there's elements such as open graph where every page needs to be correctly tagged with open graph elements to make sure the page can show perfectly when picked up by social media. And again, taxonomy and having consistent taxonomies across not just salesforce.com, but also our other sites like trailhead.com was extremely critical to being able to personalize on our homepage using content from all over the site as well as Trailhead. Secondly, tagging, like I said, is a big part of this. So tagging content improves the quality of the recommendations from the recommendation engine. It is also used to build the affinities of the profile in Marketing Cloud Personalization. And so tagging and getting this content ready with open graph, et cetera, was a big part of the work that we did in enabling personalization at scale. And then finally, like we did on segmentation, which is the third piece is where we do a lot of work in the Data Cloud. We do things like audience sizing and audience mapping. And so in this case, it was an example of a data visualization where we can do all this in Tableau, but we had essentially modeled a group of customers for next-best product. And we can use this data visualization in Tableau to size how big this segment is and what the opportunity is for marketing. So we have brought on website behavior to see this high-value segment has visited our website in the past 12 months, to see if we could activate a personalization campaign to the segment when they come to our website. So just like audience sizing, we also do an exercise called audience mapping. Also in Tableau, we have different ways of visualizing this data, so we can find the audience opportunity and plan the channel and messaging for this segment. For example, if we look at this visualization where we've got a volume of traffic to our website against revenue associated with that segment, we see that small business is a segment that we could target on the website as it has a high dollar value, and it also has a large footprint on the site. And then finally, as we build these Data Cloud segments for custom activation, we also had some off-the-shelf segments that we built for marketers across Salesforce Marketing who can go in and start visualizing some of this segmentation in the Data Cloud. And so what you see over here is some off-the-shelf visualizations that we have built for our marketers. You see contacts by industry distribution, so a marketer can go in and see what is the volume of high-tech people that they can use to potentially do an industry-focused campaign on. In the middle, you also see like contacts by job level, if they want to prepare a marketing campaign for -- based on role type, specifically to managers or to VPs or to CTOs then again, you're just looking at the volume of people in that segment, gives them ideation for what type of marketing campaign they want to run in what channel, using what type of messaging. Similarly, we have like contacts of high product interest score, we have buyer segments, which are key decision-makers with high product interest score. So lots of different visualization here on allowing marketers to really see what we're doing in Data Cloud and take action on these segments. Finally, we have very, I would say, robust ways of how we want to measure these personalization campaigns. In almost every personalization campaign we ran on our website, we saw a pretty massive lift in our engagement metrics. And so in this case, you see our personalized group, within about 2 weeks of running this campaign, we saw 8% click-through rate compared to our control group that was seeing static content, which was getting an average of about a 1% or 2% click rate, 1.9%. We also have a measurement plan on all the different types of metrics we want to monitor when we run these personalization campaigns. We typically set it up like this with the do-no-harm metrics. Do no harm just shows that when we have all these technologies and products working in the back end, we want to make sure that it's not causing any harm to page load time or any of the core like site metrics in terms of the page rendering and so on. So we look at things like our core offers, our form completes just to make sure everything is working correctly, and we have 3 post type of measurement for do no harm for pre, before the campaign is launched, post for after the campaign is launched. But then we also, during the campaign, we set it up as a control and test group. Then we have our test and optimize metrics, which is a large set of metrics we look at, including things like click-through rates, chat initializations, click-through to start trails, click-throughs to completed trails on Trailhead. And then we have a lot of like measurement and ongoing reporting metrics, which we use for optimizing these campaigns. So the purpose of this measurement and learning agenda is to show that we're looking at metrics across -- through many different lenses to continuously test and optimize various variables on these campaigns. And finally, I'll say that all these use cases from working on Data Cloud and activating these use cases at scale across sales and marketing teams, what we've learned is the truth profile -- I'm going to start with the middle one, which is really starting small and focusing on value. So for example, we had 82 different data streams coming into the Data Cloud. But if we really look at like 2 or 3 of the most important data strains, like our CRM data or our website analytics data, we were able to move use cases forward that focused on creating segments, based on top contacts and people who've also visited our website. And so it was very helpful to start small, one use case at a time and focus on the value and measurement and the KPIs that you can show for each use case success. The first one is really about building the whole Data Cloud project like a product and so having your jobs to be done, creating systems so that you can work in an agile fashion, having road maps on clear use cases and release cycles, et cetera. So building it like a product, starting small and focusing on value one use case at a time and really focusing the use case on quality over quantity. So it's not about how many data sources you have in the Data Cloud, but it's really the quality of the data sources. And so these, I would say, are the biggest learnings that we got from this experience in our digital transformation and Data Cloud journey. So with that, I will pass it back to Andrew.

Awesome. Thank you so much, Sonia. It's -- I -- as a Salesforce employee, I'm pretty familiar with the story, but I think learn something new every time I hear about all the awesome work that you and the teams are doing, so appreciate that. And for our audience, I realize we've given a lot of information today, a lot of things to think about and hopefully, a lot of little pieces of inspiration as well, too. But -- so there's obviously a lot more where this came from, a lot more where you can learn a little bit more about our products and how we're using them. So before we wrap up, I just wanted to leave you all with a few next steps in case you're interested in learning more about the Data Cloud or any of the other products that we are also implementing and we are also Customer Zero for here at Salesforce. So the first one is just get in touch with us, right? So we have a few different links on here, sfdc.co/contact if you want to connect with our account teams, have those conversations around maybe scheduling a workshop or a demo, whatever that might be in terms of what your use cases and goals are. We have a whole video series about Salesforce using Salesforce. So if you visit sfdc.co/datacloudsfonsf, you'll actually find a special video episode on Salesforce+ that goes over some of the things that Sonia covered today in terms of how we're using Data Cloud to grow our business along with other tips and tricks on implementation readiness, our own implementation highlights, best practices and more. And maybe some of you will even get a Trailblazer ID for the first time. So that's pretty cool. And then lastly, if you're looking for a Data Cloud demo with some additional use cases, visit our website, sfdc.co, Data Cloud Marketing Demo to take you straight to it. Tons of resources, tons of additional videos, use cases, whatever you might need to kind of continue on your Data Cloud journey. And that brings us to the end of the session, really flew by, but I hope you found it informative. And both Sonia and I look forward to keeping in touch with you as you progress on your own data and customer experience journeys. So with that, hope you have a great rest of your week. Thanks so much, everyone, and we'll talk soon.