The Innovation Playbook: Three Global Business Leaders on AI, Data and Brand Growth

Welcome to AdSe Week's podcast about the business of marketing, media, and creativity. In this episode, we're sharing a conversation recorded live at one of AdWeek's events featuring leaders from across the industry discussing the ideas shaping brands right now. Let's get into it. Thank you everybody for being here. We're so excited to jump into this great session, future of innovation. My name is Mark Stimber. I'm the senior media reporter here at Adri. We're joined on stage with three amazing marketers from three very different companies all sort of united around the same subject which is going to be what are you doing to innovate your firm. So as is sort of the case with these future of franchises I want to start with a little bit of a Q&A or like a a hot question at the beginning real quick answer. We'll start with you Edwin. What's sort of one consumer trend that you're seeing that's shaping how at Uber you all are thinking about innovation? A few weeks ago, several hundred Jenzers deleted their social media app. They're calling it the abstinence movement with the focus on A, not AV. Uh, all in the name of recapturing attention. Nate, >> for me, it's the level of engagement that folks are taking before they make a purchase. You know, they want to be able to fully experience the product. They want to see it in their homes, you know, whether virtually or in real life. If they want to ask a friend, whether it's AI or a real friend, they want to do that all instantaneously before pulling the trigger on a per on a purchase. >> Okay, Cheryl. >> For me, it has to be agentic commerce. you know, this whole dream of having an assistant. Everybody would have an assistant that you could give a tasks to and it could go all the way to purchase and uh you know, so this is certainly a time of innovation, new ways to discover and drive commerce and you need uh a way to trust that you know that commerce that purchase as well ensure that you have the same sort of experience you have in a store or an e-commerce today. So uh a complete reshaping of commerce. >> When Cheryl you talked a little bit about Ed Mastercard, the the thinking around AI is more of an orchestration I think was the set you used. What are you thinking internally as it pertains to AI? How are you rolling that out both sort of in consumerf facing and internal capacity? >> Yeah, I mean we look at it in two ways. You look at it and what is the productivity that we could you know change processes and systems so that we can accelerate our innovation even faster and then you look at the you know the innovations themselves. What can AI bring? And just to share an example when we focus on innovation and looking at process change one that comes together is our AI card design studio that we just launched. And this is a system like we you know we design cards every single day for our customers. We've now taken all that creativity um into a system that enables our issuers to design their own cards and then ultimately soon early next year for them to offer this to their customers to personalize their small business cards or consumer cards. And so, and when I look at that process change that happens, like for us, it's a long process of card design and approvals and then getting a card into market. You could combine compliance systems, brand systems, creativity at scale. Um, enabling our partners to get their cards done faster, enabling consumers and small businesses to express their individual personalities and styles. And 60% of uh consumers, particularly Gen Z, are looking for how they can individualize, you know, express their individuality. And so it'll create an emotional connection to the card. So we fix process. Um we enhance creativity around the entire system, thousands of issuers, and we're getting a better product into the hands of consumers and small businesses. Well, obviously efficiency is sort of the name of the game when it comes to AI, specifically on internal capacity. Nick, you'd mentioned a little bit when y'all are thinking at New Brands, which is sort of a very tactile brand in many ways. There obviously is internal uses of AI. I think you'd said something like we compress a month of coding into a day or whatever the case was. How are you applying that to the marketing function in inside the company? >> Yeah, I mean, we're really looking at it in as part as a key factor within our content supply chain. So the demands for content continue to increase. I'm sure everybody here is seeing that. You know, we're looking at five, six, seven times the amount of content needed this year versus last year. So significant increases on that. We're looking at AI to help us in very discreet areas to really speed up the development and delivery of that content. So for example, you know, we have a direct to consumer email business on Yankee Candle. you know, we are now able to go from kind of physical layout to coded email in just a click where before that took a lot of coding to be able to get that done. On the product side of things, like all of our products are tangible products. Um, when we wanted to create content, we would have to go to first production, get those first models in, put them in studio, shoot them, then develop the content. now able to take the same files that we ship out to the you know supply chain to produce we put them and create digital twins um leveraging AI and then are able to generate content in a much faster quicker way. So that basically compresses you know 8 weeks of time that we otherwise would go through with these studio shoots into a week or two and allows us to be out in front with our content needs. really looking for very discreet spaces where we can leverage AI to really help us be more agile and move quicker and and really meet the demands that are based on us. I'm curious when you have obviously sort of a portfolio of brands, what sort of safeguards do you put into place to make sure that on a case- by case basis to sort of evoking that brand voice, staying on character, etc. Like when it comes to actually translating an idea into copy and that into marketing, what do the sort of brand safeguards look like for that? All of the work that we do is human initiated. So we we we don't kind of give free range to kind of AI to create whatever they want. It's all human initiated. We call it human initiated AI amplified or augmented. And then we have certain touch points along the along the chain to double check and verify and make sure that it's delivering what we need it to deliver. So it compresses the timeline. It does a lot of the annoying kind of tedious work that would have to be done within a design process, a content development process, but it still has key safeguards both from a brand standard side of things, but also from a legal side of things um to ensure that we aren't kind of veering off into spaces we don't want to play. And then Edwin, I'm really curious you're not just with Uber, but Uber advertising specifically, which is not a new function, but certainly newer than the core sort of Uber product. So theoretically obviously a lot of innovation in a sort of nent wing of of a company. So a lot of opportunity for that. You've mentioned that you all have been working on a study with Adelaide and Canar >> and that sort of I think evokes some of the ways that Uber is starting to think about its advertising product and sort of make that pitch to marketers. Can you talk a little bit about that and explain like attention metrics? I think that's an interesting component. Yeah, I think viewability has always been something that we're uh looking at as marketers and we wanted to have a higher level of accountability and start to work with our partners to say look when we think about the media uh that we're actually delivering for our advertisers, what does attention look like and how does it actually couple with performance? And so we work with Adelaide and they really think about the implicit perspective of attention where they're looking at duration, pixels, formats that they are actually scanning across the internet as well as eyetracking and then the explicit signal of research which is with Canar. And rather than actually coupling that together and using norms, we actually leveraged what people said about the ads that they were running on Uber advertising. And so we're quite proud of what we've actually accomplished with them. It's a custom attention unit and we are driving about 11% uh differences in terms of our Journey TV ad unit versus online video and about 40% differences when it comes to our postcheckout and Journey formats. It's not about viewability or attention. It's about accountability for us. And what we're trying to do is start to tell our partners that these media surfaces matter. And when you can actually start to innovate with the measurement firms to get to that place, it it becomes a lot more meaningful to us. >> Well, and you talked a little bit about that. I think the the Uber advertising sort of pitch as we're not just a utility, we're a sort of means of relief. Yeah. I think is an interesting element. And I mean in terms of explaining why this brand lift is occurring. Can you talk a little bit about that that theory of that not that theory the concept really that Uber is sort of changing the mindset of the consumer as they're experiencing that ad? >> Yeah. Recently we released a study in Australia all around this concept of quick commerce. And I think most people would think that oh quick commerce is about the functional benefits of speed. I'm getting something quickly. And we're starting to see that it's really about relief and release. Uh one of the interesting things is that when we get to that place of Mazo's hierarchy of relief and release, it's not a new sort of thing. Uh the psychological perspective of cognitive load has been talked about since the 80s. John Sweller, I think, was the first to speak to that. And what we want to actually do for the consumer is move them from a place where they're not just looking at Uber as a fulfillment channel, but really as a place that allows them to reallocate that cognitive effort so that we're delivering to that person when they need it. As a result of that, when you're in that moment, you literally collapse the funnel to some degree and you help a brand not uh just meet a function, but an actual need. And the reason why I love that is we have to start to think about how consumers aren't thinking about brands specifically. They're actually thinking about a need and a platform that can deliver that. And when you couple that together, the natural nature is I'm going to pay more attention. I'm going to actually love your brand because you're giving me what I need in the moment. And so that's that's something that we've actually loved, you know, doing research on. >> I'm curious, Cheryl, just I mean at Mastercard presumably people are coming to your products to fulfill a need. Is that a similar line of thinking that sort of shapes y'all's marketing? >> Yeah, I mean absolutely. It's functional and emotional needs always is, you know, what we think about. Um, and interestingly enough, we just launched our own um media network. It's Mastercard Commerce Media and same thing like we have billions of transactions that we're able to you know draw insights on and create an offering based upon segments specific needs uh and understanding based upon their purchase behavior. Um and so you know we looked at this we have you know again you you talk about innovation you know we we looked at what are the components we have at Mastercard from our transaction data to our loyalty capabilities our measurement capabilities and you put it together and uh and out comes commerce media 500 million you know permissioned uh opt-in consumers 25,000 advertisers and an ability to attribute right down to the purchase behavior globally, so domestic and travel and all around the world. And this is a tremendous offering for us, which we did advertise just you know by the way around ade uh you know right in um Uber advertising which was fantastic for us by the way in reaching the B2B audience as well. And so anyway, the you know, how do you innovate? Understand people's needs. In this case, it was understanding advertisers needs to get down to that attributable measurement. Um, and we can do that up to we've seen up to 22% return on ad spend, which is amazing. >> I'm curious. I think there's probably a lot of people in this room who are thinking about starting a retail media network or working with one. They're curious about it. Can you tell us a little bit more about the process of sort of setting that up? like how long has this been in the works? Is this a bit of a new skill set for y'all? I mean, organizationally, is this something that requires sort of new learnings or what's your process of standing something like this up? >> I mean, I think for us, this is where it's like you watch the trends happening and you see what's happening in the world and you go, we could have an offering that's really interesting because of the plethora of data. We have transactions all around the world at you know across every category by every segment and you know we have a personalization capability. Several years ago we bought a company called dynamic yield. Um so we have personalization we have a marketing services capability and a loyalty platform. Um we had card linked offers. So we were already had you know thousands of advertisers working with us on their card linked offers program and it was about stringing it together and you know when you take a step back and you go what's the process for which you decide to advance this you know most certainly we looked at what are big bets and new businesses as a network business which is what we are what are the kind of businesses we could be in and and diversify from our you know traditional line of business and it was the assets were all sitting right there. So us, we had an ability. We have a great global media team. Uh and so the ability to couple our expertise in marketing and media with our services capability and tech platforms, our data capability, this is how it all it all came together as a big working team. But it was designed as a big bet and let's go figure out how to do this. And it was a fairly quick crop test, you know, under a year to get this out the door for us. >> And what's I mean, I know it's very early days, but what's sort of the initial reception been? What are you hearing from partners and clients? >> The reception's been great. I mean, again, I talk like we think about this as as marketers, you know, what's the holy grail for us? Can we get the right message, a personalized message out to uh the right consumer or business person um you know with the right offering whether that be you know a loyalty offer, a message uh whatever that might be and then can you track it back to the purchase and know that you had the incrementality. It's we can do this and so people are really excited about this prospect um you know as part of their mix. There's multiple things we all do as marketers. You know, we do crazy things and crazy experiences and then we have things that we need immediate return on right away. And so, it's a balancing of our portfolio as a marketer. But it's an exciting ad addition for both, you know, me who's actually managing our our marketing and then uh and then also our partnersh. >> Well, and then Nick, Cheryl's talked a little bit about the sort of personalization element of AI. I know you talked to me a little bit about personas and how that sort of guides a lot of the thinking and the implications and usages of AI at new brands. Can you unpack the sort of idea of personas that y'all are working with and how AI is maybe making that a little bit more sophisticated? >> Yeah, I mean the personas the best way to think about the personas is is basically as a brand you have your target consumer. This is your target consumer brought to life. So basically we we take all the data that we have on a given target consumer create a AI persona around them. So for example um one of our brands is Graco one of their uh one of their target consumer their target consumer is the first time expectant mother. We have reams of data on who this person is what they like and with each study that we do we learn a little bit more. We create the persona based on our own proprietary data because we don't want any kind of hallucinations going on. And then we're able to leverage that persona in a variety of different ways. So if you're the brand person, it's kind of like having your own little like Jarvis sitting there that you can ask questions about your target at any given time. You can say, "Hey, what do you think about this?" Or, you know, I'm seeing this trend. Is it relevant? And they will give you from their perspective what they think about uh it and give you a really consumerbacked answer at that point in time. The other way that we're doing is we we've also taken these personas and we've this is a word identified them. Um but we're using them in the innovation process. So we will either give them a brief and say hey we are looking at developing this new type of stroller that does this that services this need. You know how would you think about it? Would this be attractive to you? How would you make this concept better? And then they will go and they will build it and they will also come up with other other ideas that could potentially solve the problem. bring it back to the brand team for review. Or you could come up with a concept yourself and give it to it and say, "Hey, here's what I think. What do you think?" And it'll then go through and say, "I love this, don't like this. This is confusing. I don't know what you're talking about here." And give us that kind of feedback that we then can use and take and validate in other areas. Now, the personas are overly positive. That's the only kind of danger of them is like they'll never tell you you're wrong. Great idea. They're a little bit passive aggressive sometimes, but they are not they're never going to say, "Yeah, bad idea. Kill that." So, you have to kind of still bring a level of judgment into the equation, but it it significantly advances your thinking, your development at the same time, decreasing the the time from, you know, spark of an idea to validated concept and allows us to just be uh quicker in how we approach things. >> I'm so curious about this. How long have you all been working with these sort of AI enabled personas? >> About a year now. Uh and I mean it's we started off small. We started off with maybe like three or four different ones depending on on the key brands. We've now expanded them to almost all brands and we're also seeing the kind of overlaps between the different personas. So sometimes it's really interesting especially as you're looking at you know whites space areas to kind of partner two personas together and have them work on a particular innovation challenge project brief to see what they might uncover. >> Yeah. In agentic collaboration. >> Yes. Well, um, is has there ever been a moment I would be so curious with this sort of synthetic data concept of like we can produce market information via artificial intelligence? Do you ever get some bit of feedback from it that's very counterintuitive and then you go, is that real kind of thing and ever have to sort of confirm it with an actual member of the target demo? I mean, like, is there any like suspicion with some of the data that might be surprising? >> I'm sure there is. I don't have a good example off the top of my head, but this is why we have some of those fail safes in. So, while we might do a large amount of work with the synthetic audiences, at the tail end of anything, we we bring in real live audiences to gut check it and real live people just say, "All right, any kind of like gross, you know, errors, you know, in in the way that we're thinking and that that can help ensure that we're not about to kind of head out with something that is just a massive mistake." >> And this ties into something that you were saying earlier a little bit, which is this this broader idea of of at new really trying to lead a little bit more with emotion in some of the branding and move a little bit away I don't want to say move away from the performance elements but expand the funnel a little bit and think lead with emotion in many ways these personas I'm assuming would would sort of help guide those athletes >> yeah absolutely I mean I think we've had phenomenal products >> I think the focus for a number of years has been just talking about it through a product lens which takes you to a very functional rational space that's important but we all know like humans are not always rational and emotions play a big impact on how folks choose and purchase their items. And so what we're talking about is like how do we pair these fantastic products, but make sure we bring a little bit back more of that brand storytelling back to the equation. We want these products to not only perform and deliver up to expectations, but to also elicit some kind of response. You know, it makes somebody feel something because at the end of the day, we're participating in categories where, you know, the difference between the products are real, but but they're minimal. I mean, a blender is going to blend regardless of which brand you get it from. So, how do you make sure that the right answer to the question of which blender do you want is an oster blender. Well, you've got to make sure that you're bringing the best product along with a meaningful brand story that's going to resonate with that consumer and elevate that brand to the top of the the purchase decision. And I know everybody I think in this room would probably like to have a license to do a little bit more highlevel emotionally oriented branding. Last question on this front. What does that conversation look like with with either the CFO or where, you know, trying to sort of justify that budget in that spin when you're saying this isn't as performance tied as maybe some of our other efforts. What's your pitch to the money? >> I mean, I think there's a little bit of both. Uh, a little two two things. One, both of my CEO is a former CFO and my CFO. So, I got two CFOs to worry about. Um, but thankfully, both of them come from PNG where brand is like a way of life. So the bar for me to kind of convince them anything is a little bit less than I think it would be with your stereotypical CFO. But there's also plenty of studies that we're able to kind of reference and show that you know this the idea of building your brand is critical. Building kind of what would have been called top of funnel is absolutely mandatory because you otherwise your lower funnel needs converted to much higher level year after year after year. You have to drive that kind of upper funnel stuff and so showing them with data is is a very compelling convincing way to do that. Now, we've got to be choiceful on the brands that we support and make sure we're putting kind of critical mass from a funding standpoint behind brands to make that impact, to make that difference. So, that's really where the debate is is, you know, we have 50 brands at new like which ones are the ones we're going to lean into and then we're going to be able to drive because we can't afford to be able to kind of lean into all of them. Yeah. >> When talking about obviously performance, I feel like that's the sort of calling card of Uber. y'all are moving. I don't know if it's from performance, but I would say also a little bit more toward a high level brand function, but I think still one of the best things about Uber is that they're very intuitive. You're talking about a sort of contextual product that y'all are working with this sort of if you're going to a Starbucks or near a Starbucks. I mean, it's sort of expanding the idea of it's it's both context based and it's performative based or it's performance based, but it's sort of guided by AI. >> Yeah. with our one of our products, the right offers products. Um, I think context and timing >> uh sometimes are as important and outweigh even the way we traditionally think about targeting. Get two bucks off at a Starbucks even though you're going to the office about 500 ft away. That's a product that we've been sort of working on that delivers that right sort of deal, that right sort of offer so that a consumer is like, "Yeah, I'll take that." And you know, one of the best case studies is last year at around this time, we worked with Kalúa brands and I had my first espresso martini at the Ritz with my friend Emma just yesterday. Pretty awesome. Um, and uh, it it was across all of our surfaces, including on Uber Eats. And what was amazing about that is as a result, we literally drove about 188% increase of what they're usually used to in terms of what happens on delivery. and quite literally 7% increase in places that were featuring Kalúa and that espresso martini in the bars and the restaurants. Imagine if you could work with a platform through advertising that through that relevance you can actually move people >> to go somewhere else. And I think that's where we're at when Cheryl was talking about it, that when you think about a platform, that serendipity and that relevance is so critical. And I think that context becomes something so real. Uh and obviously the measurement piece of it, the brand scores uh that we work with with Canar were through the roof. But I think the big takeaway is if you're delivering in that moment, you can add a lot of value not just for the platform, but definitely for the brand uh in a very new context. I mean, sort of it's so funny because I feel like contextual advertising is, you know, the original advertising, but the way that y'all are sort of using AI to make it a little bit more sophisticated and obviously optimize those efforts going forward, how do you continue to sort of push that boundary? Like what is the sort of next chapter of these intelligent contextual ads look like at Uber Advertise? >> I think it starts with the utility of what we bring to the consumer even before the ads. M >> what do we mean? And when you think about this concept of AI, machine learning, the learning engine needs to work with logistical precision in order to get a rider and a driver together. That's get an earner to deliver a meal or groceries to somebody. it. We're quite literally using that same format but being very useful to a consumer in delivering optimization formats and scaling up to say look what an advertiser is going to want from Uber and Uber advertising is opportunity to actually have high attention units that make people go oh I'm seeing that I'm computing it I it's impressing something on me and then the as Cheryl was talking about earlier the ability to actually close a transaction and say was there incrementality for my brand and those are all the things that we hope that we're going to be delivering for our partners see in the long run. Well, then sure, I kind of wanted to ask one last question. I think this is super interesting. This idea of multi-ensory branding and Mastercard has leaned so much into. We've been talking a little bit obviously a sort of perpetual conversation about ROI, how do you justify that, etc. When it comes to these sorts of efforts, >> what does that sort of measurement look like for y'all? >> Yeah, I mean, we we really have looked at our branding and how do you innovate on our branding? And you know it happen you you follow the consumer behavior and you go okay we're moving to the small screen and our logo is getting smaller there and our name is sitting under the logo you can't read it and small screen everybody knows our logo and our brands and what it means and stands for trust experience acceptance so we dropped the name from the logo so that was site then we moved into sound recognizing the rise of audio and smart speakers and such uh your brand may not always be seen. So now what do you do to create a branding queue and trust uh we created a Sonic brand. So whether you're uh in a Uber or whether you're in you know whether you're checking out an e-commerce you get our sonic brand reinforcement and then we even have a haptic brand that you could feel a set of vibrations that gives you a sense and peace of mind that your transaction was safe. So uh and then even restaurants where you could actually get all the senses, taste, sound, smell, etc. So you know, but again, you know, tell your CFO you're going to open a restaurant and then they say, "Are you crazy?" Um right, you know, what's the margin on that? And we're not in the restaurant business, we're in the experience business. And we track a couple of things. One is uh anytime somebody has an experience or even gets an invite to an experience, we see a lift in both brand perception and affection and then specifically usage as well. And so one we're able to track back to the reason why we have an experience strategy. And then when you look at things like Sonic and Haptic, the reinforcement that you get of trust uh is two times greater when you have that Sonic brand there. And that's not only a return for us, that's a return to merchants, more likely to go to our merchant partners. Think of small businesses and the ability to reinforce that trust with an e-commerce transaction. And so, uh, we'll look at those measures, um, particularly, but yes, it's an interesting one as you get into the, you know, experiences business, how you actually convince, um, you know, the system to go along with you on these. >> Yeah, absolutely. Well, that's all the time that we have everybody. Thank you so much to our panelists and thank you all for being here. >> That was a conversation recorded live at an AdWeek event. You can find more coverage, analysis, and interviews from across the marketing and media world at adweek.com. Thanks for listening to Adspeak. If you like the show, be sure to follow or subscribe wherever you get your podcasts.