Win the AI-Powered Shopping Era | Retail Keynote, Dreamforce 2026

Please welcome Vice President, Solution Engineering, Salesforce, Michelle Collins.

All right.

Welcome, everyone. Thank you for being here for our retail keynote. We are so excited to have you here with us today. We have to start this presentation with a thank you. Thank you for being here on the last day of Dreamforce. Thank you for being here after a night of Dreamfest with Usher and Gwen Stefani. And thank you for being here in your biggest season of the year in retail. We are so glad to have all of you here in the room with us and watching online. But the biggest reason that I'm excited about this keynote is because Salesforce is doubling down on retail. In fact, you're going to see more innovation coming out of our commerce portfolio than you've seen in the last 5 years. And we've made incredible acquisitions in this space with Simulate, Contentful, Fin, and many others that we're going to bring to life for you for retail, and many of you are already using today. And lastly, we have so many agents that we're packaging up to get them to speed to market to you faster. So you can get these in front of your consumers, your associates, and start using this technology now. And it's more important now than ever before because in the state of retail, you've got a balancing act right now, right? We see that macro headwinds are changing the way that you're going to market. You've got margin pressure that's been compounding for years. You've got consumer headwinds where consumers are shifting their focus to purchase based on value. And you have AI changing the way that consumers are experiencing your brands, right? Browsing in LLMs and making sure your brands are showing up in the right place in the right way. And this is challenging all of us to think differently. How are we all going to shape the future of retail? And how do we take these challenges and turn them into a competitive advantage? That's what we're gonna talk about today is how do we make sure that your brand shows up in the answer of the LLM? How do we make sure that you have that complete context everywhere and that you're aggregating it across channels and meeting your consumers as they move across experiences with your brand? How do we anticipate the customer's needs before they need them and then respond to them in real time? This is what we're gonna talk about today. And the key question that we want all of you to think about as we move throughout this keynote is, What does it take to win in this AI-powered era? And that's where Salesforce comes into play. We are ready to help every single one of you take your retail brand to the future. And you may have seen this slide before throughout Dreamforce. Here it is again, but we're going to talk about it a little bit different lens for you now. We're going to talk about it in the lens of retail, starting at the top and working down. When we talk about AI force, we're really talking about how do you activate across any surface in any channel, whether that be Claude, Slack, Coworker, or email, SMS, voice, any channel of your choice, any surface that you want to engage with your consumers on or your associates? And then within those channels, how do you activate any agent anytime? This could be a personal shopper agent. This could be an order servicing agent. But you have the opportunity to activate those agents to any channel at any time and use the power of the context This is where your business processes are defined, where your workflow lives. This could be something like creating a campaign or setting up your rules for loyalty or when you're going to escalate a case. All those business processes live in marketing and commerce and service applications that your teams use every day, and all of those need to be grounded in rich context and data across your enterprise. And that's where Data 360 comes into play. That's where we're activating the data. Raising it up to the applications, to the agents, and to your consumers and your associates on any channel as they move throughout their journey with your brand. So we're going to bring this to life to you today through a demo and a series of demos and customer conversations where we're going to unify every dimension of this journey. We're going to start with data and the fact that real-time data and context is grounding every experience that will come after. So as we'll talk about marketing, commerce, service, and loyalty, we'll have this data foundation that will be the baseline that everything works off of going forward. And so let's go ahead and jump into that now. Let's jump into the first section, which is going to be all about data. And all of you, many of you in the room are experimenting with AI. And the number one thing that we've heard from everyone experimenting is an agent is only as smart as the context that it is given. So how do we activate the data that you've invested in for years across your foundation of your enterprise? How do we activate that data and get it into the hands of your consumers and your associates in the channel and experience of their choice. And no better people to talk about that than our incredible customers from Lowe's. I am so excited to welcome to the stage Amanda Bailey and Abhishek Nandi from Lowe's. Please give them a round of applause.

Hi, good to see you.

Hi.

All right, so if any of you joined our Data 360 keynote yesterday, you have already heard from these incredible customers. We are so excited to have the opportunity to showcase them again. and hear about their data activation story. So I'm going to start, Abhishek, with a question for you, which is, over the years, you guys have had a data foundation in place, just like many of our retailers in the room, that you've invested in for years and decades, as we've been customers of Lowe's for years and decades. But you added Data360 to that foundation. And tell us about it. What value were you looking to unlock? Tell us about that data transformation for you.

Sure. So if you think about all the data that we have, we are predominantly a Google BigQuery shop. That's where we have grounded all of our enterprise data with governance, lineage, everything. But data is of no good use if you really can't activate on it, especially in the marketing context or the digital context or for any matter, right? That's where Data Cloud comes in. Now, the way we operated, right, it was already in the Salesforce ecosystem. We were using Marketing Cloud heavily, and it's been a while that we use that product, right? And it only made sense to bring in Data Cloud and Data 360, as we call it now, right, to go alongside with that and make it really easy for the team to activate the data.

Incredible. Yeah, you guys were early adopters. You've been on the journey from Data Cloud to now Data 360. We love the value that you're unlocking. And Amanda, I'd love to ask you from a business perspective with your consumers, with your associates, your enterprise, can you tell us a little bit about the value that you're unlocking with Data 360?

Yeah, absolutely. I mean, the biggest thing I like to think about is, you know, we know a lot about our customers. We have over 700 attributes in Data360 about our customers, and the real power is to take all that data and turn it into true customer understanding. And so, you know, customers don't wake up every day thinking, oh, I'm going to find personalized marketing today. You know, they wake up on a mission. They've got a project to do. They have a repair to complete. And so part of this is helping us better understand, you know, who that customer is, what they're trying to accomplish, and then it makes our job easier to deliver a relevant experience. So whether that's the content that we give them or the offer that we serve up, it's a valuable, relevant experience that they keep coming back to. And so that has allowed us to take personalization from just a marketing tactic and strategy to truly an enterprise growth engine. And so there's real power in how we've used I love that.

And I think it's not just great for Lowe's as an enterprise, it's better for your consumers. We all want real-time personalization. We want you to meet us in the moments that matter. And you guys are leading the way. So thank you both. Thank you for being here. Round of applause for Amanda and Abhishek. Appreciate both of you. All right. Well, thank you, Lowe's, for enlightening all of us on the data foundation. We all know it's a critical piece of everything that's going to come after in this story. And to move us on to our next section of the keynote, I'm gonna welcome to the stage my friend Kristen Bertane. Come on up.

Thank you, Michelle. So as Michelle and the Lowe's team just established, Data360 is there to capture every customer signal the moment it happens. Your shopper abandons their cart, they purchase something, they abandon their browsing session, And Data360 is there to capture it. But now we've created a really good problem. That's a lot of signals that we're capturing from our customers. I think Lo said 700 attributes that they have about their customers. So now we've created a really good problem. What do we do with all of those signals? The challenge is reacting to them and responding to them in real time with the right offer, the right content, the right messaging while it matters most to the consumer and it's still relevant to them. So I want you to think about this today. What if you could determine your strategy ahead of time and then hand off the real-time execution? So we've shipped a wave of innovation and agents to help you do just that. So you're going to define the strategy, define the business outcome, and then let agents take care of the rest, reducing the burden on the marketer and making sure that their job is so much easier. We have a couple of examples up here that I'll talk about. So the marketing goals agent that you're going to see in the demo today, you define the business outcome like increasing conversions or increasing order revenue through your site, and then we're going to agentically build a campaign and an audience segment that's going to drive towards that goal. Also, great news for marketers. We've shipped a wave of innovation to help you work faster across content, segmentation, and journeys. And if you hadn't already heard in the marketing keynote, we've announced how AI Force will allow marketers to execute on campaigns from all of their different surfaces, be it Slack or Teams or Claude. Very exciting. And the great thing is we already have customers doing this in amazing ways. So take Under Armour, for example. They are using the power of Marketing Cloud and AI Force to automate 10 million email sends. And all of these are shaped by the individual shopper. They're not a blanket send. If you want to hear more about their story, take a look at the Marketing Cloud keynote on Salesforce+. And also, I just wanted to say thank you so much to the Under Armour team for everything that you've shared this week at Dreamforce. And congratulations on all your success. Let's give it up for Under Armour, everyone. All right, so we're going to get into the demo, and what you're about to see is really exciting. The first thing you're going to see in this first chapter of the demo is meeting your customer where they're actually shopping today in an LLM like ChatGPT. That's going to drive traffic to our retailer site, and then when our customer abandons their browse, gives us that first signal, we're going to see how we actually reengage them within the Marketing Goals agent. And now to help me drive the demo, please give it up for Mike Gross, my demo driver. Okay. So we're going to get started here with our customer Lauren Bailey in our fictional retailer Northern Trail Outfitters. They sell outdoor wear apparel and products. So we've got a pretty basic-looking profile of Lauren Bailey here. It's nothing super special, but I promise you we are going to come back to this throughout the demo and at the end, and you'll see how it evolves, because this is going to be the profile that stitches together all of the engagement touchpoints that Lauren has across sales, service, and marketing with NTO. And the value of this is that Data 360 is going to power all of the personalized touchpoints that you see today in all of these demos. Okay, Lauren Bailey is going to start her shopping journey today where so many of your customers are doing their product research. They're no longer directly going to sites. They're going to compare in ChatGPT. Lauren is shopping for a really specific scenario today. She's going to a camping music festival. She wants gear that's going to be really technical for her and withstand the weather, but it also has to be stylish as well too. So she's loving what she sees, and Commerce Cloud has surfaced a product carousel directly within ChatGPT, and that makes all of NTO's products actually discoverable to consumers when they're shopping today. So she's loving all of the products. She's going to click through into the site and start browsing. So now Lauren is going to be logged into the site. She's clicking around. She's browsing. But she's getting maybe a little bit overwhelmed. She's seeing a lot of products across a lot of categories, and she gets distracted, and she walks away. So now the challenge becomes, how do we keep customers like Lauren engaged in a really highly personalized way? And then how do we do that not just for Lauren, but for all of your customers at scale? So now we're going to see the experience of the marketer working in Marketing Cloud. So in the past, marketers have had to do a lot. They've had to pull external data from IT, go out to a number of different dashboards for insights on their campaigns, and sometimes in the worst-case scenario, also cross their fingers and hope for ROI. But in the new way of working within Marketing Cloud and the Marketing Goals Agent, we're going to flip that on its head. We're going to actually start with the end in mind. We're going to set our goal, which is to increase order revenue in this case, We're going to set the dates that the campaign will run, and then most importantly, we're also going to set what is our budget for the campaign so that we're always driving towards that and we're not overspending in places that we don't need to. Under the hood, a campaign agent is going to draft the brief for us, and a content agent is actually going to draft all of the content in there as well. In the optimisation tab here, the system is going to let us know if we're ever falling short of our goal and nudge the marketer to make a change. But here, setting an adaptive audience segment as well too is things where things are going to get really interesting. So we're going to have the marketer still describe their segments, but the system is going to continually score individuals in that campaign to make sure that they're still eligible for it and that we don't overspend on the marketing budget We optimise on our paid and owned media channels, and then also we don't over-message to your customers. So all of that is done in the system so that the marketer has less burden on them and so much of an easier job, and they can be a hero to the company and be a good steward of the budget. The output of all of this is going to be a really highly personalised email campaign that goes out to Lauren that's based on her abandoned browse session, and perfect for all of her festival needs. It feels really tailored and personalized to her. And we're going to do all of this at scale for all of your customers. And the best part is, as Lauren engages with this email content, as she clicks, as she opens it, all of that is going to be fed back into Marketing Cloud to train more agentic campaigns and make everything more personalized for Lauren. And now to continue on with our story, please welcome to the stage my friend and industry advisor, Martin Priest.

Thank you, Kristen. So what Kristen just showed you was marketers being able to reengage with shoppers wherever they are and in AI channels that you don't control. And it's those AI channels that shoppers are increasingly using to search and discover your products. We've seen 200% growth year over year in customers that are using AI channels as the starting point in all of their shopping journeys. This means that shoppers are arriving more considered and potentially misinformed than ever before. So it's becoming critical to understand and capture their intent and action on it in the moment. And with 41% of shoppers using natural language search on branded websites, they're openly giving you those intent signals But you have to be ready to action on them because getting mentioned upstream isn't the goal, conversion is. And so this isn't a shift in preferred channels, it's a structural change in how customers expect to be able to shop. We need discovery to move at the speed of conversation. You need to be able to capture that customer intent and conversation when they arrive. Now there's 2 sides to this: your doors and everywhere else. And if product discovery is happening more and more in channels that you don't control, then it's our job as retailers to make sure that we don't just show up there, but that our product and our brand is represented the way that we intend, and that we're ready to bring that customer into our property with compelling information. And then we can capture their intent through natural language search or 2-way conversation to action that in real time, turn it into conversion. Because getting mentioned upstream isn't the goal, driving conversion is. So let's say we do everything right. A shopper does an LLM search, our brand and our products show up the way we intend, and they choose us. Where does that choice get tested? In the store, digital or physical, that's where trust has always been made or broken. And so we need to be able to deliver on that promise, right? 81% of shopping journeys begin online, while 84% of them are still completed in store. If there's one thing that every generation can agree on, from Generation X to Gen Z, it's that brick-and-mortar is still the number one preferred channel by everybody. So we need to be able to connect everything that brought the shopper there, that got them to come in the door or click through, and deliver on that promise. And we've heard you. After announcing doubling our investment in commerce, we're announcing the largest commerce release in 5 years. And so connecting this journey has never been more critical, from LLM search to checkout, be it on your storefront, your brick and mortar, your app, or other channel than ever before. And how we're enabling you to do this is through things like Agentic Commerce channels that allow you to syndicate your product catalog to traditional social or these new AI channels to become visible. Palmada is an AEO tool from Contentful that allows you to monitor and optimize your brand and product health in these new AI channels. And then Shopper Agent Carter and Agenticom Research allow you to understand customer intent and deliver on a 2-way conversation driving personalization on the site. And the newly announced yesterday Store Associate Agent allows you to bring these agentic capabilities from the storefront into brick and mortar, enabling the store associate to meet your customers with these new capabilities. Now, let's go back to that shopping journey with Lauren and take everything that I've been talking about and let's bring it to life. So I'd love to invite my friend Anna Levine onto the stage and Mike behind the booth to take you through it.

Thanks, Martin. Let's go back to our story. Our customer Lauren loved the personalized email from NTO. She's now at home preparing for her trip. She's very excited about the festival, but she just checked the forecast and it's gonna rain a lot. She needs a new tent, so she goes to the NTO website and she types, I need a tent for hot and rainy weather. With traditional keyword search, a query like this might give her no results, or results that are not relevant for her. But agentic commerce search understands what she really needs in natural language. A tent that can stay cool but also can handle heavy rain. So she quickly finds the right tent and adds it to the cart. But now she needs boots for the festival, but not just any boots. She wants something that can work for the weather and also feel like her style. So for this, she goes to the shopper agent for help, and the agent recommends boots that match her style in colors she's bought before and that are available in her usual size. But how can the agent give these recommendations that are so personalized? This is because she's logged in and Teon knows her from past interactions, and it uses all this context for very personalized recommendations because it knows her past purchases, her returns, her style, her preferences. And did you notice all of this was preselected for her? And the site dynamically changed to show the same products the agent was recommending for her to make it easy for Lauren to find the right boots and add them to the cart. So before checkout, she sees that the agent recommends a very nice raincoat and she really likes it. But it's a very technical piece, so she wants to try it on first. She decides to buy everything online and pick up in store. So she checks out, and when she walks into the store, the offline and online experiences come together. On the point-of-sale console, the store associate can search a unified real-time view of her customer, with data like her loyalty status, past purchases, her preferences, real-time recommendations. So all this information is context for the store to continue Lauren's experience seamlessly. So Lauren tries on the boots and the jacket. She really likes the jacket, but the boots are a bit too small. The size she needs is not available in the store. But that's not a problem because the store associate starts an omnichannel exchange to find the right size for her and send the boots to her home. This way, the store doesn't lose the sale and this— and Lauren can get her boots without having to start over. From search to store, every step stays connected. And with every interaction, NTO learns more about Lauren and makes the journey with the brand more relevant and personalized for her. Back to you, Martin.

Thank you so much, Anne. Mike, that was amazing. And I would love to invite Chad Best, Senior Vice President of Customer Experience and Operations from Lovesac, to the stage. Chad, thank you so much for joining us today. So I'd love to start at the beginning of sort of where I was talking was, so as a retail leader that is at the forefront of these changes today with consumers, who's literally on the pulse of experience, what changes are you and your team seeing with customer behavior today?

Yeah, I would say quite a few actually. It used to be in our stores that customers would would come in learning about the brand, learning about our products. We would go through a typical demo process, but that's not really the case anymore today. Customers come to us really educated about the brand and the product. Their questions are much more refined in what it is that they're looking for, and in some cases, they've even made some early decisions about fabric choices or size, etc. And so they're pretty deep in the decision set. We also have seen just some pretty dramatic increases in e-commerce traffic, as you mentioned earlier. But to the point of stores still being important, brick and mortar, it is still a really critical part of our customer shopping journey. We're seeing double-digit increases in conversion as a result.

And these double-digit increases in conversion, these are not necessarily actions that you're taking, but like a shift in consumer behavior and education when they're arriving in the stores?

Yeah, absolutely. We're seeing customers when they come to us, as I mentioned, pretty far down the decision set. However, buying furniture is a considered purchase and does take some time. So we use Salesforce POS to capture information about the customer. And then we continue to follow up with them throughout their journey, answering questions and to help them convert. One of the things that we're really working on right now is unifying that experience end-to-end so that we can provide visibility not just to our store associates, but to our CRM team, our customer service team, etc., so that we're all able to help that customer wherever they are in their journey.

So you're using this you know, this number one preferred channel by everybody, Salesforce POS, to collect all of this customer information, feed it back to the customer intelligence layer, and using it to sort of manage the customer through this journey for this high-considered purchase that may not always be closed in one store visit or even one channel visit.

Yeah, that's absolutely true. We can, we can meet the customer where they are as they have questions because we have visibility to where they were in their journey prior. So if they leave our store and go to the website, we now can see that, we understand what they're doing, and we're able to pick up exactly where they are and help them convert.

Can you take us back to the beginning and maybe tell everybody sort of like what was the impetus for this change? What started this journey?

Yeah, absolutely. You know, for us, we were capturing on the store side all of this incredible data from our customers. We spend on average at least 30 minutes with our customers, sometimes quite a bit more, and we were understanding what their budget was. Were they interested in financing? What colors did they like? What size did they need? And we had all of this incredible information, but we weren't putting it in a system that anyone else could access. And so other than that store associate continuing the journey, the customer was just kind of left to start over in any other channel that they they might visit, or potentially even on their next store visit as well. At the same time that I was trying to figure out how to unlock quotes to drive conversion and get this data, my CRM counterpart was really trying to figure out how do we unify a bunch of platforms and figure out how to personalize communication. So we joined forces, and now we've started a multi-cloud project with Salesforce as our partner. And really looking to just transform the end-to-end experience.

Yeah, with the ultimate goal of connecting all that customer data and making it seamless.

Absolutely.

Chad, thank you much for taking the time and sharing everything with your industry. Thank you, everybody. Now, I'd like to invite onto the stage my friend and colleague, Celeste Richardson, industry advisor for Salesforce.

All right.

Thank you, Martin and Chad. What an incredible story. Thank you for sharing. Love, Lovesac brand. Now I have the pleasure of taking us on our final stage of our journey today, which is all about customer service. And everything that we've covered so far is about anticipating the shopper, personalization showing up the moment that they're inspired, delivering a seamless commerce experience. But customer service has largely been the exception. It's reactive by design, waiting for the customer to call us, tell us their problem, and then we go and fix it. But that changes right here because the same real-time signal that powers personalization and intent now allows customer service to get ahead of the problem, to recognize customers the moment that they call, understand their preferences, their recent activity, their loyalty status. And the, the cost of doing this is not just operational. This is very much an emotional thing. 83% of customers say that they are more loyal to a brand when they are— sorry, they are more loyal to a brand when they receive proactive service. And this doesn't have to be full of big grand gestures either. This can be relatively simple, notifying them of a shipping delay or recognizing and prioritizing customers the moment they reach out and Showing them who they truly are to your brand. That's the real unlock here. It's about having a signal that's already in your data and having an agent that knows what to do with it. And what's really special is with all of this comes together when we talk about everything that we've talked about today, because it's not just about being efficient and opening and closing a service case. It's about what did we learn from that moment and what did that moment teach the rest of the business? So a return case that just got closed can send a signal to the marketing and to the commerce agents to prioritize a new fit option. Can also let merchants know if we're starting to see a return pattern that's flagging a fit or a quality issue so that we can get ahead of those. And this takes a best-in-class platform to deliver. Hopefully you had a chance to check out the Service Keynote yesterday. If you didn't, I'm going to highlight a few of my favorite features. Obviously, you've heard a lot about the Casey help agent this week, but I'm going to spend a little time talking about the Agentforce Contact Center, because this is something that you have been asking for. You don't want to manage a patchwork of integrations. And so we're bringing all of your telephony, your chat, and your CRM onto a single platform to help reduce total cost of ownership. We have Agentforce Voice that you're going to hear in a couple of minutes, where it answers the phone like a human would. And we have self-learning agents that allow you to leverage every single signal to continue to better the model so that you're not having to go back and retrain anything. Now, I'm going to invite my friends up to give a demo, but what you're about to see is a seamless handoff from a service agent. We're going to hold on to all the context that we have from Lauren's earlier journey so that she's not going to have to repeat herself. She's not going to have to explain things again. We're going to have all of that context held together. All right. I'm going to turn it over to my friend Kanavis and Noah, our amazing demo driver.

Let's give it up again for my amazing demo driver, Noah. Thanks for driving for me. Okay, so as Celeste just mentioned, we're gonna pick up exactly where we saw Lauren last leave off. That first begins with the SMS message that we see she receives indicating that her order will not be arriving on time. It's not the end of the world. We've all been there, we've all done that. But the beautiful thing about this particular interaction, Lauren has the ability to respond directly within the thread, and the agent is going to be able to assist her. So as you notice, a human did not jump in. It did not intervene. Instead, we have the order servicing agent hard at work in the background. It's retrieving all of Lauren's order details, and on top of that, it's also checking the live inventory at nearby store locations and preparing that delivery to shift over to an in-store pickup on her behalf. Translating what this means, Lauren's delighted. It means that customer retention is going up for her. She has a jam-packed weekend. She has a lot to do. This is going to save her an extra step. So now we— let's say Lauren's jam-packed, right? She has to get ready for the festival. She's packing. She has the ability to request that the agent go ahead and give her a call. Let's go and pick up that phone call, Noah.

Hey, this is NTO customer care calling you back. Is this Lauren?

Yeah, the one and only.

Great. Continuing from where we left off. I found the Ballard rain boots in a size 7 at our NTO Crater Lake location, about 2 miles from zip code 97604. Would you like me to order these for pickup or schedule same-day delivery?

Uh, pickup would be amazing, but here's the thing. I'm on my way to the festival tomorrow. I could maybe pick them up on the way, but are they even going to be ready?

Awesome. Same-day pickup orders are generally ready within 1 hour. Want me to place the order?

Yeah, that would be great.

You got it, Lauren. I've placed the order. You will receive a notification once your order is ready for pickup. Would you like me to cancel your existing order?

Uh, yes please. Go ahead and save me that step.

No problem. Working on it. I've canceled your original order and sent a request to UPS to reroute the delivery back to our warehouse. You should receive a refund within the next 3 to 5 business days back to your original payment method. Anything else I can help you with?

No, that was all. Thanks so much for your assistance. Now, while that was amazing, here's the thing that actually matters the most. What you're seeing is a unified profile for Lauren, Data360, and it's going to have a summation of everything that we've seen up until this point. So whether it's the marketing journey that we saw Lauren first embark on at the beginning of the presentation, If it's a commerce interaction, whether that's in-store or on the web, and even some of those service touchpoints that we just saw, each and every one of those interactions is going to help populate this view that you're seeing. It's going to allow it to blossom. It's going to allow each and every time that Lauren interacts with this brand, she's going to receive an additional layer of hyper-personalization, ensuring that she walks away happy with all of those touchpoints. Furthermore, taking it another step, She has friends, she has family. She's going to communicate with them verbally via word of mouth. This was an amazing experience. So it's going to help boost your customer acquisition as well. So if we take a few steps back and just summarize everything that we've just walked through, we followed the journey of one customer on one journey and looked at one unified profile. But I invite you to imagine this experience for your brand and the implications with your customer base.

Thank you.

Thank you. And with that, I invite Celeste back to the stage.

All right.

What an incredible demo. And I love that we got to see where we started and where we ended with Lauren's profile. Okay, now I want to brag on my friends at Crocs for a minute because they are doing some incredible work. They are a powerhouse brand. They are not just a classic log anymore. That is hard to say. But they are a leader in agentic commerce, and they actually were experiencing a similar problem that I know many of you are, that they were fielding hundreds of thousands of customer inquiries, repeatable, avoidable questions like, where is my order, every single month. And over 80% of those were going to a live agent. That was you know, not efficient for the business and also not the best customer experience. And so they have built Rivet. It is a super agent that not only is handling all of those repetitive questions for the customer autonomously, but it also can offer shopper agent that is helping the customer discover new products, even helped me discover these cute sandals that I'm wearing today. So to talk a little bit more about the story, I want to invite Feliz Papich up to the stage. She's the SVP of Digital for Crocs. Felice, thanks so much for joining us today.

Hi.

So for those who haven't shopped Crocs recently, which I know they're all going to go out and get on their phones right after this, give us a little bit of backdrop of what— where Crocs is at in your journey and what made this change in digital engagement so important for the brand.

Yeah, we've actually been We've been, I'd say, really ambitious in the digital environment for a long time. So we went right into marketplaces without hesitation. We leaned into TikTok. And I think the next evolution for us is taking that to the step of truly figuring out how to deliver against intent. I think every brand's struggling with that. We have grown. Obviously, we're not just the classic clog. We have a big sandals business. We have our Jibbitz, which is personalization, I'd say, above and beyond any other footwear brand out there. So we felt like an agent was the perfect way for us to really speak to the customer, showcase everything, right? Whether you're going to Dreamforce and you need the perfect shoes, or if you're going on a beach vacation with your whole family, what does that look like? How do you get clogs, sandals, the right beach Jibbitz to go with that? go with it, and all of that was just a really great moment for us to kind of bring that long-tail conversation, really make it about the consumer. And then we also felt like, why not make this a super agent that can also handle the other aspect of the journey that really delivers that lifetime value? We, we call our fans Croc Nation, and we want to keep Croc Nation really happy, so we kind of tie it all together.

Yeah, so that was a big decision to have an orchestrator agent that's kind of understanding the customer intent and If they want help shopping, finding the right Jibbitz for their item, or if they actually need help and getting them to the right place. Talk a little bit about that decision-making process. How did you kind of get to this level of we need to orchestrate this and you want one surface for the customer to engage with you on?

Yeah, yeah, you do see a lot of brands doing both, right? You see a lot of brands out there that are doing the shopper agent and then they're keeping their customer service agents separate, and we really felt like why make the customer work for this. Like, if we can find a solution, we can deliver against that, then it's seamless. There's never a question of where they go or which spot of the website they have to find this. They just automatically see Rivet. They know that's their trusted source. They can come in and they can either ask, you know, about a discovery moment for them with the brand, or they can ask, where's my order? Where's my refund? They can even ask about our shipping policies. shipping timelines, all those things that are really critical. And then it can be handed right off to an agent if that can't be answered in real time, which is the greatest moment I think as a brand is when you really have found that opportunity to reduce friction. I think we always talk about it.

Yeah.

It's always a conversation. But I think most of the time we don't actually reduce that moment and make it very easy, which was really where we wanted to be is is just ease for the customer at the end of the day.

Yeah, and that makes total sense. And when you talk about reducing friction, you had to do some hard work, like making sure that Commerce and Service Cloud are integrated together so that you're able to capture those signals and build on that customer context. Can you talk a little bit about the technical process that went into this?

Yeah, we learned a lot. So I'd say that it really is about the connectivity of your architecture and your data. It, it, it can't be ignored. It's super boring and it's not the most exciting part of the conversation, but it, it made a lot of sense for us to think about not only what was Rivet going to know about our customer, what was Rivet also gonna know about our brand and how do we create those environments in which it knew it was the best possible agent to deliver against product recommendations, against what they needed to know about their order, and we had to connect all of those things. So, you know, obviously with the super agent, you have the orchestration layer. You then have kind of the shopper agent that's connected to onsite search. You have it connected to your content management system. You have it connected to all these little pieces that were in the past fragmented. And then you have the customer service side of the fence that's connected into your OMS. It's connected down into the actual Service Cloud, right? But it really becomes important, the stitching, all of that. And then also, where is that data and how does Rivet get access to that data? Because without that, again, the quality isn't there. You can create an interface, but if you're not actually delivering against expectation, then unfortunately for the consumer, that's where you fall short yet again, right? So it was an adventure for us and a big learning, but, you know, couldn't be more proud of what's out there to our consumers.

Well, you've moved incredibly fast on this, and not to mention you've also launched this in voice, right? So you have consistency across multiple channels.

Yep.

All right, talk a little bit about just like what's been so surprising about this and maybe where you're excited about where it goes next.

Yeah, yeah, yeah.

We—

yeah, voice obviously was a really quick follow for us, but kind of bringing it to the point of we didn't just want to kind of do, you know, I think a lot of us in this room talk about the concept of an MVP, right? And we get sidetracked that the consumer doesn't think about it that way, and the consumer really needs the whole experience for the value to be there. So we brought voice in really quickly, and with our partners, you know, with Salesforce and with RoseTree, we were able to kind of take all the learnings we did upfront for the web experience and pull that into the voice and move super quick. You know, and so for us, I think the next iteration is is true, true personalization and really bringing Rivet into a shopping companion. Right now, Rivet is great at answering questions, and we want Rivet to be able to predict what the consumer needs. We want Rivet to be able to understand what you're looking for. If you're looking for these types of sandals, what else might you need? We think it is the perfect opportunity to bring Jibbitz to life.

Yes.

We really think that we have struggled. All 7 years I've been with Crocs, I have not figured out how to do Jibbitz online. And I think a shopper agent might finally be that moment for us that takes thousands of SKUs that are meant to be personal to each person and truly bring those into one environment, add to cart, get the consumer out, and also know that about them in the future.

That is incredible. And we are going to be waiting anxiously to see this innovation and transformation. Thank you for your partnership. Honestly, you have been a longstanding partner and just a see the level of innovation and how you're continuing to push forward is just so exciting. So thank you. Thank you to the entire Crocs team.

Thank you.

All right, I'm going to close us out because we started this hour asking, what is it going to take to win the AI-powered shopping era? Not just survive it. We want to win this era. We saw Lauren go on a journey. We saw her in new discovery channels. We personalized to her across every single channel. She had seamless transactions. She never had to repeat herself on what she bought, when she bought it, because the system already knew. That's the answer to our question. When your data, your workflows, and your agents work together as one, you have a single foundation that that is informing every next action, staying ahead of the customer and delivering the experience that they so much desire. Now, this is not goodbye. This is see you later because we still have sessions this afternoon. We hope that you can join us for. We will also be at Retail AI Club next, next week, and we will be at ShopTalk in 2 weeks. We hope to see you there. And of course, we'll see you all in January at NRF. We look forward to continuing the conversation. Now, for the most exciting part of our show today, we have a giveaway. Please check under your seats for a golden ticket. There are 15 lucky winners out there. So please check it out. We are giving you some swag to the Salesforce store. Meet us at the demo booth if you are a lucky winner.

Woo-hoo!

We just want to say thank you so much for spending the last hour with us, spending your time with us at Dreamforce. We know it's a precious time in retail. Thank you to our demo drivers, our amazing speakers from Lowe's, Crocs, and Lovesac, and to all of our Salesforce speakers. Thank you all.

Thank you.