[music] [Abhijit Ghosh] Good afternoon, everyone. A very warm welcome to each and every one of you. I know it's been a long week at Summit, so thank you for spending the time with us. I have the pleasure of kind of sharing the stage today with Tadeu Banzato from T-Mobile, and I'm Abhijit Ghosh from Adobe. And in this session, we are going to talk to you about how T-Mobile Leverages Contextual Experiences at Scale to drive Customer Journeys. So let's keep the session interactive. We'll kind of go through the flow and then we will open it up for Q&A towards the end just to make sure that we kind of answer any outstanding questions if we were not able to cover that during this session.

Cool, so let's get started. So as we go through the session, just to level set on the agenda, we are going to cover three key areas as we go through this exercise. We are going to talk about how T-Mobile brought together the diverse set of systems that they have to drive better decisioning, essentially laying down the foundation, that is needed to drive your customer-centric approach. Second, we are going to deep-dive into how once you have the foundation layer there, how you use that intelligence to drive against your use cases and drive brand authenticity. And third, we are going to touch in terms of, what does it mean from a organization level, and carry out something at this scale, right? I'm going to let Tadeu do all the talking after I set the stage in here so that you can hear it directly from Tadeu. Alright, so let's kickstart, right.

When it comes to telecommunication, we are seeing a couple of trends that are raising the importance of strong digital experiences. One, we see an evolution of competition as it relates to growth of MVNOs wireless + wireline bundling, new entrants that are coming in the market, which is needing for us to be more faster and granular, as we engage with our consumers out there. When it comes to, the second is rising consumer expectation. From a consumer standpoint, they want more of you. And what we are seeing in here is a centralization of experience leadership, localization of marketing activity, and a very strong emphasis on network quality as our consumers are using the brands. Which is driving a heightened importance as it relates to effective digital experiences and how we handle customer value management at the core of the business. Third, which is pressure for efficiencies, which I think is, it's not just specific from a telco standpoint, but we are seeing across all, and I know we have business owners in here, we see across all verticals in general, is cost reduction. How can we better optimize our spends on the tech stack? How can we rationalize that? How can we embed GenAI into the notion of marketing and care to drive more automation and reduce the friction that kind of exists in there in the customer engagement like cycle? Adobe carried out a quick survey with asking telecommunication users for their preferences. And not surprisingly, what we saw was, customer views providers as interchangeable and kind of look at changing that every one to two years. What was also interesting was that it's beyond price and reliability that our consumers are asking for, as they spend time on the respective telecommunication brands. They see personalized communication, a key piece as it relates to disruption management, and how they look at trust when it comes to handling, let's say, network outages, as you operate and work onto the platform, per se.

Following an unprecedented change in the telco, customer experience will be key for differentiator for thriving brands. As we look at expansion for next generation networks, which makes it easier for consumers to switch from one network to other, as we know all network portability, which emphasizes an important aspect, of how we need to focus as a business when it comes to optimized retention.

We are also looking at expanded product offerings that are getting created out there every day, which kind of then kind of translates into how you look at personalized acquisition and upsell opportunities by building on that centralized repository. Changes in consumer behavior, which is driven by price sensitivity and inflation are also key aspects that come into picture as we look at improved experiences that drive customer satisfaction. How you can transform a traditional purchase channel by embedding more efficient means of marketing and care as our customers onboard on multi journey. And while we do all of this, we have to keep a first-party data strategy management in mind.

Which kind of highlights the heightened opportunity for personalization first, keeping a first-party privacy approach in mind as we look into these things. With that, I'd like to hand over the mic to Tadeu, who's going to walk us through this in a real-life example of how T-Mobile is doing it on their side and driving some of these contextual experiences at scale as they're transforming customer journeys. Tadeu? [Tadeu Banzato] Thank you, Abhijit.

First of all, I have to open my heart to you guys. This presentation almost didn't happen because Abhijit is not a T-Mobile customer. Shame on him.

But he eventually convinced me to come in here and talk a little bit about the experience that we're driving in T-Mobile.

I am, and I'm not going to do extensive introduction here but I want to make sure that I level set that, I am the platform manager for everything related to messaging coming out of T-Mobile. So any outbound communication, could be a service, could be a marketing. I look at the platforms that are enabling T-Mobile to get prepared for the future, the future that we have been seeing here with GenAI, all that creativeness that Adobe is bringing to the table. I am the person that is making the future ready from the messaging standpoint, from T-Mobile.

I like to start my, actually my presentation. First of all, I'm not native English speaker, so don't get to cut up on my bad English here, but I'll do my best to entertain you guys as well. So I like to bring in that quote from our technology president. "Our network was built to enhance the way people live their day-to-day lives." And I'll try to use that quote to also tie up everything that we're doing to put our customer in the center of everything that we do.

It's super important to say first, because that quote came from a technology guy, and is super humanizing, right? That quote essentially says that we need to do our best to connect people, to connect families, to connect friends, wherever they are with the best technology possible. And the technology is just the enabler, right? So this is a quick overview of the T-Mobile 5G penetration,. And I'm building this story with that and starting with that, just to tell you that T-Mobile has more than 325 people using our 5G network right now, which is the fastest one in us, and probably the fastest one in the world.

We have about 2 million square miles of 5G penetration in US today.

And it's good too, because that whole expansion and the way that we've been doing this so quickly, so fast, so nimble, also forced us to review everything that we had at home to be able to get to the same fast, into the same pace that we were delivering the network for our customers, right? Just to give you guys a little bit of a sense of what I'm talking about in terms of volumes, T-Mobile today sends about 4 billion SMS messages yearly. 2 billion emails, I think we're almost doubling that already.

Millions of push notifications, millions of direct mails. We also integrate our platforms for outbound calling and everything else, right? So although we're talking about messaging, about calling, we also consider sort of messaging as well, because we're reaching out to the customers.

Let me give a step back, one thing that I was forgetting. So T-Mobile, we consider contextual experience broader than just dynamic content to our customers, right? What I mean by that is that, we try to obviously reach you the best content, one-on-one experiences, et cetera, et cetera. But we need to get the full context of your household, or the full context of everything that you're doing, so we can be more targeted and more one-on-one communicating to you as well. And it's not there yet, we're following that journey. aAnd that journey began when I joined the company. Not saying that I'm the only responsible for it, but that journey started when I joined the company, looking at all in every single platform that we had in place to orchestrate outbound messaging. And we're talking about 14, 15 systems that sends out communications to our customers, that used to send out communications to our customers about three or four ESPs, email service providers, which made, which might sound that our future was really, really far away for us to do contextual experiences. And I guess a lot of companies here today, also have been going through this path of identifying legacy stuff that are hidden in the closet, and then trying to get those skeletons out and figure it out how to solve the best way as possible for our future.

So we acknowledge all of that, we identified all of that. So my responsibility was to build a three-year roadmap to identify how we're going to migrate from that state, where we were doing a lot of somewhat manual spread and pray messages, things like gives me chills only to talk about, but we acknowledge all of that. And I put together a roadmap, which essentially was going to build our future state for contextual experiences, which is also tying back to all the AI opportunities that we are going to drive in the next few quarters or year. So it was a challenge, right? Because I had to, although I own the platform, I don't own the messaging strategy. So, and I have some of my stakeholders here as well listening so they can come back and gossip about me in the company. But essentially, I had to really partner close together with them, and we are sort of tied to the hip, to the whole conspiracy that we're bringing to the table to build all of this amazing stuff that we're building inside of T-Mobile. And on top of that, everything that we're preparing T-Mobile for the future in a way that we can experiment faster, innovate faster, even fail faster, right? But the thing is that we are now able to recover much faster than using 14 message systems that we were manually sort of managing.

So I'll enter to the chapter two and talk a little bit specifically about a very interesting and sort of first of its kind for T-Mobile specifically, but I guess some carriers also have the same issue, which is network experience optimization, or we call internally NEO.

So essentially, while we were building the roadmap and enabling the platforms, we had the networking team or the engineering team from T-Mobile coming to us and saying, we need to find a way to make our network to communicate to our customers. And when I say network, essentially no human interactions there. And whatever the network was doing in terms of upgrades or in terms of new spectrum of 5G being enabled, or new business being enabled, like home internet that we didn't have in the past, we were thinking about, I mean, they were thinking about a way to communicate to those users or customers, but they came the old mindset of, let's spread and pray the message, and see what we get out of it, or just communicate to people that might not be relevant to that situation.

So it felt like changing the tires while we were riding, right? And essentially, felt like that we were riding a Formula One car, and we had to do that at speed that was not imagined before. So we had to enable, in delivery...

we had to acknowledge that we wanted to enable delivery of contextual experiences on day one, right? And I'll talk about the platforms, how we were rationalize all of that. But essentially that NEO use case was a way for T-Mobile to challenge our own status quo on the way that we were working. So we partnered up with a bunch of people inside of the company, multidisciplinary folks, engineers and et cetera, and I felt super stupid in some of those meetings. But it was super laser-focused that this challenge was going to be disrupting everything that we do in terms of outbound communication for the company, and in terms of how T-Mobile also bring the whole mindset and experience at place to put the customer in front of everything we're doing. So we knew that we had to activate multi-channels. We knew had that we had to operationalize everything at scale for not just the customer, and I'll get to the details of the architecture that we thought of, but also how do we make sure that we're building something that will be possibly reusable for multiple use cases in the future, and that we wouldn't have to rethink and rework, and we were building the pipes, as I mentioned. I mean, I'm a platform person, so building the pipes that would eventually enable any network communication use cases to be live. So we came up with the idea of... I mean, in some of the iterations that we were doing, we had to figure out a way to identify which of which of the towers across the globe and across the map that I showed you before, the users were connecting from. So we had to come up with a calculation. So essentially in a time span, if user connected to a specific tower for more than 50% of the time, we were calling that their home tower. It could be an office, could be whatever, but for the sake of the iteration, we started calling it Home Tower. So we could be very focused and targeted to that specific customer when the network was getting to any updates or anything.

And I'll just dive into the use cases. I think that's more interesting anyway. But first use case that we wanted to implement is actually just notify the users, but instead of sending to a specific area code or a specific ZIP code, we were tying that up to the network and the specific tower that the user was connected. So we know where user is connecting, how long they're connecting to a tower, if they shift from one location to the other, we all know that.

The second one was actually, if there was a planned outage, we wanted to let them know, the user know that was connecting to that specific tower upfront so they could prepare themselves for any service deprecation that was happening while we're doing that maintenance. So we would send an SMS or an email or a push notification to the user and say, "Your network tower will be out next week because we're doing a upgrade, or we're doing a transformation there." Then signal improvement. So when the network was available and upgraded, we wanted to communicate to that user that they had the best 5G in town. So they had to take advantage of that. And then we had the idea of the fourth use case, which is essentially enabling T-Mobile to do more contextual messages, content and experience, and also do cross sell and upsell and et cetera, right? Because we know that the user connected to that tower or the device connected to that tower. It's not a 5G-enabled device, or we knew that the user didn't actually turn on the 5G on their iPhones, for example. So we knew all of that, and we wanted the network to communicate to that user directly without to having a human sort of nitpicking everything out and pinpointing whatever attributes we wanted to have.

So, I know this is a little bit bearing dump. I'm not an architect by nature, so if any architects are here, please don't throw any rocks at me. But it is a horrible design, but I'll try to make sense of it.

Our journey with Adobe, started way back when we acquire... I mean, even before we acquiring Sprint. But at the end of the day, Adobe has been working with us and trying to pitch a lot of products to us. But at the end of the day, what we have started is that by, I'm not mistaken, 2018, I was not part of the company, so I don't have that specific date. We started ideating in how to bring AEP as a platform for the company. Essentially, we started working and enabling AEP or RealTime CDP at scale by 2019, 2020, if I'm not mistaken, or a little bit later than that, because we had to get our legacy stuff together, right? How we had to consolidate all the data, we had to consolidate the way that the data was going to be transformed or was going to be injected inside of the system. So it makes sense now, makes sense tomorrow, and also in the near future, or longer future, let's call it that way. So for NEO specifically, we also had to figure out, what were the systems that we needed to tap into to get all the attributes and all the data that I was mentioning about. So device data, device activation data, user profile data that is connecting to a specific tower. There are like a bunch of systems, I would say, six or seven of them, that we had to tap in and understand how the data was going to be distributed, how the data was going to be actually transformed for us to put into streams and events inside of the RTCDP. And also we had to figure out if we had all the attributes already available in the RTCDP. So in the user profile, or else we wouldn't be able to communicate to that customer. So that was a challenge by itself. But then for Neo also, we were introducing a new way to communicate. So, and I was mentioning about Adobe Campaign in the past. So T-Mobile today, we live with those two solutions in place. We have Adobe Campaign v8 or Classic v8, and we have AEP, and how we make sense of that, is that ACV8, well, I mean, we call internally, ACV8 or maybe ACCV8, or Adobe Campaign V8. We use that for all the batch marketing activations, essentially, because it does a much better job in terms of volumes for blasted communication, lookup tables. And there is a flexible way for us to isolate side load data instead of having to figure out, every single time, how we would put that data inside of AEP or inside of the RTCDP or whatever, right into the user profile. So it gives a flexibility for our marketers, our CRM folks to operationalize that and start orchestrating things that almost impossible to be done in the past with the 14 systems that I thought about, and the three SPs.

So we found that to be a very low-hanging fruit. So we delivered a CV8, early last year. So essentially that low-hanging fruit to enable and to start changing the mindset of the marketers or the operations team that now they can do more contextual experiences at a batch level, but it is helping us to prepare for the future of AJO and everything else that we're bringing to the table. So with AJO, we listen everything, we have all the orchestration, I mean, obviously, those are not real messages, but gives sort of a context and sort of a hidden microphone, gives sort of the context of what we're driving with Neo, First message, just outbound communication, super targeted, super tailor-made for that. From the context of that user that is connecting to a specific tower, we're sending the message. So now we're not relying on the ZIP codes or the area code from the phone to send out a super-targeted message. And that can really open a lot of opportunities for us.

Then after sending the first message, the user continuing to the journey with AJO, and we can now give more some real-time context for that user. So obviously, dynamic naming whatever, or first name into each of the message. But if you look at the second one, we have an opportunity, we send an a link to the user now saying, "Activate your 5G device," but we know which device the user has, right? And now we can do that context of, "Click here to figure out how to activate it. If you are an iPhone user, you'll get a specific message, or you'll get linked to a specific iPhone-tailored or customized or contextualized landing page, and you could send that same for an Android user as well.

And then the fourth example, obviously, I put a little highlight there, because we're driving some contextualization to that. But then when we start talking about the opportunities of, now we know that a user connected to that tower that just got updated, we can send messages to them that are contextualized with cross-sell and upsell upsell opportunities. We know that the user will never get a benefit of a 5G ultra capacity if they still have a Nokia device. Sorry, I didn't want to make a joke on that. But essentially, that's what we're building.

So some of the outcomes, right? The whole experience we were building was actually to not just to make the network communicate by itself to the customer, but also we were trying to drive cost reduction from cost to care becase that's essentially one of the problem that we wanted to address, and I had that on a slide, but I forgot to mention before.

So whenever we were doing upgrades or, introducing 5G to a specific region, we saw an incremental number of volumes to care, calls to care, which was a concern for us, We were going super fast and we were getting more and more calls. So we wanted to validate that assumption that, if we communicate more transparently, more honestly with our customers, will that help us to reduce those calls to care? And then some of the results out of it.

9% of the users recall receiving the message from the first one, 79% did not see any disruption. So we started seeing that makes sense for us to continue doing this, scale these use cases forward and so forth. And then I'll jump into the second one, which is more relevant as well, is that 88% that recall receiving the message, receiving the first message. Also noticed, sorry. 88% of the users that received the second message also remember that they have received the first message, which is good, meaning that we tried and learned by sending the first message. Does that gets really relevant for them when a tower is being upgraded or not? Most of the users obviously didn't see any disruption on their service, but I think that the biggest number that we want to call out here is 95% of network confidence level. What that means that those users not only like T-Mobile, but we'll also recommend T-Mobile for our customer, for their friends customer, and I did that for Abhijit. So I had to recommend, and he's probably shifting right after this call. So, we're not stopping there, right? There's a lot of opportunities that we're bringing to the table. We have about 3000 service message use cases that we're bringing to this whole architecture that I showed before. And that's helping us to accelerate some of our context, contextual experiences that we are planning for our future.

One of the biggest call out in this slide here is that our home internet users and that NEO use case will definitely tie together super well, because home internet users essentially use our routers and device to connect to our ultra-capacity 5G. They use that for work, for everything. And whenever a tower outage is going to happen, or an upgrade of service is going to happen on their region, they can get prepared much sooner, so they can continue working, they can figure out how to get the whole experience in place and continue using our network So I'll pause for now and I'll introduce you back.

Thanks, Tadeu. So that was amazing. I mean, 14 marketing systems, three ESPs clearly, it was not an easy task. Like, could you help us like talk to us a little bit about the transformation that really happened in there in terms of processes, training, communication, how you started measuring it. I know we spoke briefly about the network sentiment analysis, which kind of revalidated the notion of how NEO was performing, right? And how you saw an increase in terms of the trust that customers have on the quality of service. But what else was going on, like from a business standpoint, from a change management perspective that we were seeing in here? And not to belittle, but consolidating 14 systems, I should say 17 systems and building on that rich customer profile, which is not just setting you up for today, but for future use cases. How is that helping you, as we go through this process? First of all, consolidating the platforms not only bring a ton of savings for the company in terms of money. And, I can tell a little bit individually, so I'm not state for everyone here right now, but there's a lot of money involved on consolidating on those platforms. So that's the first thing that we have acknowledged and to get buying from all our leadership across the board. That was the first step, and also making the rationale that we had to do that or else we're going to be super late in the game of contextual experience and data driven decision making and everything else, right? And AI, as we're hearing a lot here the last three days.

It's hard to say because it was super hard.

I'm at T-Mobile now for, what? Four years. I guess my hair fall. Like I was there for 10 years and my beard get more wider and wider every day. But I think that the key thing is that, it is an ever-evolving sort of framework and experience that we're bringing to the table. So one of my key stakeholders is here.

He might call some BS from my end right now, but we've been trying to partner up with everybody in the company to put together this framework and make sure that everybody is sort of driving to the same vision and to the same future state. So what we started doing is educating everybody, especially the people from operational side, is that we brought in and also when we were doing enablement for AEP at the time for the data stuff, we already opened up all the Adobe training for everybody inside of the company. So whoever was curious to have a training in whatever platform they would be able to do that training. But then after that, we actually did our T-Mobile tailored training with operations, with everybody that was also interested. We opened up for the whole company. So people from different areas, let's say, people that were actually operating AEM, et cetera, could also join those sessions.

And then desk site coaching. So we enabled a lot of Adobe folks helping us to desk site coach, all of that. We also brought in a few Adobe consultants to help us out drive the enablement of the features and all that transformation from the technology side.

But again, I mean, it's always evolving. I mean, every day we have a new demand and we actually had to put together a strong process to intake of new data, to intake of new features into the platform. And I'm the person that actually interface that directly with Adobe for us to enable new features and requirements for the products that we're implementing and the vision or the future vision that we also have.

Sorry to bring up a question which we haven't discussed, but can you put a context in terms of, what was the TTL before when you had 14 systems or three messaging platforms to deliver something of this capability? I know you said you didn't have the system in place. But if you think about it, you had to deliver something of this nature with that kind of a setup versus how you kind of rolled it out. Now, now would you differentiate in terms of from planning to execution? It was a little bit insane, thanks to my leadership, but essentially, we're talking about two years of enablement of the data platform with the RTCDP, one more year for implementing a CV8. So we implemented a CV8 last year. We're migrating all of the lineup businesses there, and we couldn't migrate faster because we didn't have all the data from all the lineup business because of the upstream systems.

So a CV8, one year, two years from AP, ACV8 and AJO. I mean, this year is our AJO year. We're migrating almost 3000 use cases. Our goal is to migrate 60% of our use cases into AJO and have all the batch campaigns running out of ACV8 by the end of this year for all the line of businesses that we have. Just to kind of familiarize it with the other terms that we are hearing. So basically saying, you already have a well-oiled system, which is there to run your business-oriented or batch-oriented use cases. And now we are looking into the next-gen set of use cases, which are more customer-driven, leveraging AJO, and you're really harmonizing the overall ecosystem and mix and play here, where you're learning through the different interactions by consumers having on your platform. So yeah, essentially, the RTCDP is sort of our brains for everything in terms of messaging. Like all the data rely there, all the data is like consolidated there.

There are attributes coming in from different sources, and it's a ever-evolving way of getting the data in.

But yeah, I mean, it goes AEP, like was in the horrible design that I did, but it goes AEP, AJO for real-time and event-driven messages. And the same profile, all segmentations happen here in AEP, that same segmentation go to AJO. And we use that to send the orchestrate the batch messages. It all tied together. And when we get to a future of consolidating all that, that's our goal. Okay. Interesting. We want to kind of take it to the closure and just talk about the key learnings and the takeaways here. Yeah. Some of the lessons learned, right? For me, we had to challenge and disrupt the status quo of the way that we were working, not only for messaging, but we had to influence a lot of people from the data side as well. A lot of engineers, a lot of people that didn't buy the vision at first, but ended up going with whatever we were bringing into place.

You always have to have a forward thinking, not just strategy, but also attitude.

So there's always someone that will come and raise their hand, but that's not how we are doing business here. And you have to help them to understand what the future is. You have to bring them the vision of maybe five years from now, we might not even have a job, right? So we need to start jumping into these new platforms that we're bringing into place. And I think one of my passion is actually learned by doing.

Don't be afraid of trying to push and try to make things crash. I think that's the whole spirit, this whole spirit that we have been trying to bring to the table.

Things will not be perfect at first. And we don't want to boil the ocean to get to something that will not deliver.

Also be tying that up to, be open to test new things Be brave enough. I mean, you fail. I mean, everybody fails. And the way and how fast you can recover, and that resiliency needs to be part of the whole business, or else you'll just be a fail, a pile of fails, right? Yeah, and always be customer-obsessed. I think that's a key across the board. So we came up with the idea for Neo, because of one of the engineering concerns, and we're finding that much more relevant for future communications to our customers. Imagine the beauty of having, for example, a network tower in a region that is having a flood.

We can communicate that to that person that is at that moment, at that place in time and say, "Hey, there is a hurricane coming, there is a flood coming. Be prepared, because you're connected X amount of miles closer to that tower versus the other. So we can be very targeted to that, and we could actually open that at scale, not only for T-Mobile, we could actually think about, if I would do a foresight here, we could actually be able to generate revenue for the company using that whole technology and infrastructure that we have put in place.

So key takeaways, right? Acknowledge and embrace the problem.

Sometimes, we are trying to push away some problems. We need to acknowledge that, we need to fall in love with that problem and try to solve for that as soon and as fast as possible.

Always focus on scalability. I think that's tying back up to the 14 platforms that we had in the past. I don't think anyone was thinking in scalability at that time. And maybe the technology was not there for us to think about it. But everything that we're driving is with a purpose of, we might get one use case today, but by the end of the year, we might have 3000 use cases there. So we need to make sure that the scale is actually by designing everything that we're doing.

Experimentation, innovation. And I think my motto is actually "Done is better than perfect." So we just drive stuff. We make it happen. It doesn't work today, let's figure it out, let's fix, let's optimize, let's make it happen in the future as well. That's it, Thank you, Tadeu. It was great.