[MUSIC] [Jason Hickey] Welcome everybody. My name is Jason Hickey. I am a Group Manager for Product Marketing for Journey Optimizer, and I am joined on stage by Roger Woods, our Senior Director of Product Management for Journey Optimizer and Adobe Target. This session is 801, Journey Optimizer Roadmap and Innovations. If you're in the wrong room, it's totally fine. No judgment. We're going to be talking today about a lot of really cool stuff.

One caveat. Everything that we talk about, that's forward-looking is not a commitment and is subject to change but we're really going to nail all this stuff.

So, here's the agenda for today. We want to give you just the tiniest bit of history lesson on what Journey Optimizer is all about. Why did we build it? Why did we look to disrupt ourselves five years ago and go into the labs and build this product? We want to talk about what led us to this moment, what Journey Optimizer is all about, where we're at right now, and then we're going to shift into where we're going. And the majority of this presentation is going to be a roadmap. You're going to learn about new product innovations that we either just released, right now this month in March, or that are on the near-term horizon. So I'm really excited to walk through all of this with you guys, and Roger going to kick it off with the vision. "Why we're here?" [Roger Woods] Okay. Thanks, Jason. There's a little bit of storytelling, little bit of demo, little bit of discussion, okay? So, I'm going to start with a little bit of storytelling on vision of why we're here. Number of years ago, 5 or 6 years ago, as we were looking at the experience cloud product set and looking at what they do, there's a pretty prime and ripe opportunity to disrupt ourselves. Part of that was because there was this notion of brand-initiated moments that you've heard Adobe and us talk about. but also this notion of customer-initiated moments and melding those and getting harmony between the two is where we saw a lot of friction in the engagement space, customer journey space. So, that was one where brands are saying, "Hey, I'm sending messages. My customers are also doing something else or my guest or my user, whatever you may call them, is doing something else and I don't know how to arbitrate or communicate or deliver experiences. I don't know how to stop something. I don't know how to nurture them through to an end goal or an outcome." And we saw this as attention in a number of areas. Some of those areas came from inbound channels and outbound channels not being connected well in the brand. And we saw that, quite frankly, in our own Adobe applications. There was some things that did really well on low-latency experiences for inbound. Others that were more audience-based experiences for outbound. Fire-and-Forget style. And when you try and weave all that together between brand-initiated and customer-initiated experiences, it cause some friction points. So, what we did is we looked at six markets. This is about five years ago and this is from an analyst's view, everybody is a little bit different but there were markets that were very much channel-based like mobile engagement automation or email service provider, ESP. There were other ones that dealt with workflows and orchestration around cross-campaign management or journey visualization, at the time it was a little bit newer, and then others like real-time interaction management, it is very much about decisioning and logic and intelligence. And same with experience optimization. So with this lens, this is what we set out to do. We set out to bring together one-to-one and batch orchestration of people. Real-time moments, in the moment, as well as the audience-based. So that's what we set out to do. The second thing is bring inbound and outbound together which is a very challenging, complex thing on the Journey canvas Or if you think about the life cycle because you're firing messages at somebody for outbound and you want to check, do they still qualify for that message when they're coming to inbound channels. Is it still relevant? And those have to be connected together. Most organizations aren't even organized to think that way because their orgs are siloed regardless of the product, organization, there's some challenges that we'll talk about at the very end of this session today. And the third one was really we want to bring together. Intelligence through the whole piece. So that's really the foundational... I would call those strategic product pillars for Adobe Journey Optimizer a few years ago that we set out to really solve and push forward. And I'll show you how that then translates into the next 3 years, in just a moment.

Awesome. So, we talked about this session... a lot about product road map and vision. And we're not going to do a 101 'what is Journey Optimizer' today.

Happy to talk to you guys about that. Lots of great content out there. But I do want to talk about where we are. We look at today. What have we accomplished? This product's been in market for just under three years and I think we've done a lot in the past three years. So, how are our customers using Journey Optimizer today? We think about how we've bridged already outbound and inbound and batch to one-to-one and layering the intelligence. And we think about all of these moments. The moments that a brand has with the consumer whether or not they were initiated by the brand or initiated by some real-time event. And we think about how Journey Optimizer is always listening. We have this principle that we call active listening which means that we are ready in milliseconds to respond to any signal and then layer that into the current and full historical context of a user, a profile based on what they maybe have done in a store or the call that they just had with the call center the other day, or a recent order they had or a trip that they took. So, we think of this as this spectrum of tactics and the spectrum of capabilities that go all the way down to what we say as unitary which is one-to-one. Talking to Jason, talking to Roger about their booking all the way to broad-based. Address your entire population, your known universe, or some kind of audience-based thing in between that's based on just signaling to your platinum members or your gold members or that pivotal moment where they graduate from being a gold member into becoming a platinum member.

Our customers are using this every day for geo-aware location-sensitive contextual personalization campaigns. As I walk through an invisible geofence powered by the app and the device in my pocket, I can get a push notification or I can get a text message. I can get some kind of a signal that's saying, hey, you're really near the store that you like a lot and there's a coupon you should walk in and check it out. Our customers are using us for way more than just marketing use cases. You think about operational and transactional and run-the-business kind of work. Password resets and one-time PIN, two-factor authentication. Check your flight status. Check your reservation. Check your experience or have your experience when you are in a physical location, in an amusement park even, for safety and support and service. So, the use cases that customers are using Journey Optimizer go well beyond just batch and blast marketing campaigns or even personalized promotions and offers.

Here's another way to look at... a little bit of what I was saying about how we're bringing these things together. And we want you to be able to do it on one pane of glass, on one surface, on one workflow, in one canvas. And to be able to say that I have journeys that are initiated from a place of 'read the entire audience'. Or you have journeys that are initiating from that trigger or that signal. And all of these use cases we now have bringing together in Journey Optimizer. So, Roger's going to take a couple of minutes and set up the strategic foundational pillars of how we're moving this product into the future and then I'm going to start talking about individual features, capabilities, and product releases that we're going to do to achieve those strategic aims.

So, this is where it's been really fun. I've got to give a quick shout-out to Jason and his team and the product management team. The last 6 months, we've really been on this journey of saying, what does the next three years look like? And where where we going? We know where we came from, where do we want to go? A lot of that feedback and or that plan and strategy has been driven by people in this room, people here at Summit, customers that we visited globally by listening to what's working and what's not and where they want to go, and where you're trying to drive the business. I'm going to break up this slide, as we go through it, into two main parts. There's the top row which is really some objects that you have in the system. Audiences, journeys, channels, and content. Now, if you'll bear with me, I know not for everybody in here, English is their first language, but I'm just going to call those nouns or objects. Then there's a set of things underneath in the system. Orchestration, decisioning, experimentation, measuring, and governing. Those we could call the verbs or those are really the life system of what you're trying to do for experience optimization. When you take these and you apply them together or you interweave them together, that's what we're going to do today. So, you're going to hear Jason talk about features and functions or new capabilities to solve needs that you have, and what that will do is start to bring orchestration not just against content or against journeys but orchestration against channels, or decisioning, not just against content like 'next best offer' but decisioning against a journey or an action on the canvas for 'next best action'. Ultimately, all driving towards next best conversation, next best experience. So, what we fundamentally believe over the next few years, we've started on a lot of this, you've seen we already have things across these verbs, but it's really bringing them together on a purpose-built system to achieve experience optimization and that's really the goal of what we want to do. So, we'll be weaving that through the entire system. So, just think about that chart as we start talking about it and then the challenge is for you to go back and say, "What are we doing in the space of experimentation and how are we experimenting against channels? How are we experimenting against journeys? How are we experimenting against decisions that we make within the business?" And think about this a little differently than everybody operating just at the noun level or the object level of channels, journeys, audiences. How do we mature the business to start thinking about the optimization objects of the verbs? So, let's get into it.

I want to take all of those nouns and verbs and start showing you some actual product functionality that we're going to release. We'll start with audiences. If you guys were on the keynote, the main stage, you saw Amit Ahuja talk about Federated Audience composition. And it was framed as real-time CDP. I'm super excited to share with you that Federated Audience Composition as a platform native service also applies to Journey Optimizer. So, this is bringing the past 20 years of campaign DNA. All of what we've learned across our decades of experience in multi-channel marketing, cross-channel campaign management in Adobe campaign, modernizing that onto Adobe Experience platform to minimize the data redundancy of the data attributes that you don't necessarily all want to bring into Adobe. To help you govern and control the use of that data and you all have existing technology stacks. You're working in all kinds of different data warehouses. All of that data might not live in Adobe but Federated Audience composition is going to let you plug into those data sources at the source, in both real-time CDP and Journey Optimizer to use them in your campaigns and your journeys. So, we're super excited about this one.

Talked a little bit about this. I want to go through one use case, on the next slide, of how you might use this, because it's not just about bringing data in to build an audience. You're able to enrich that audience with profile attributes or personalization attributes that then can be delivered for that one-to-one personalization in any channel or on any screen or surface. To be able to pull in, for instance, attributes about my local store that I go to. Who is the manager on duty? What are their hours? Do they have any local sales? What's the address? All of that great stuff. It's not necessarily part of my profile. It's part of the information about any given store and Roger's favorite store is different than my favorite store. So, here's an example. You are working through. You need to get some high-value loyal customer data.

What you see here is a workflow and this workflow is super visual. It's no code. It's meant for a non-technical practitioner to step-by-step on a visual canvas. Pull in and merge the data sets that they need from disparate sources. Enrich those audiences with the data points that they need to then activate in any channel across paid media, across organic channels, across email, the website, the mobile app, the call center, or anywhere that they need it. So, super excited about this. I will plug an audience session for you guys at Summit. It's called All Things Audiences. They're going to dig into this a lot more. We only had about three minutes to talk about this one but if you want to learn more about Federated Audience composition and audiences in general and how they apply across Adobe Experience Platform applications, go check out All Things Audiences.

So, let's get into channels. This is where we have some really cool stuff coming up and we got some really cool demos to show you.

So, channels. Talking about channels and content. How you deliver content in the channels that you have and how many channels that you actually have in our product.

So, I'm going to talk about four channels that I have already released or are soon to be releasing. So, we're going to talk about in-browser messages. This is a new channel that you have on the web surface to very easily deliver pop-ups and overlays and modals. We're going to talk about code-based experiences. This is expanding the persona or the audience of Journey Optimizer to bring in the developer audience and the product manager audience to deliver code-based payloads to any inbound surface. Connected televisions, digital voice assistance, call centers, you name it. Mobile apps, websites, anything. Inbound cards.

You've maybe heard of them as a message feed or personalized cards. You've probably experienced them in your daily life. If you think about when you open the app and you have a set of offers and you can dismiss them and they get penned, and then they're in different orders and they're ranked. Giving you the ability to actually prioritize the order of those offers and which ones are dismissible and which ones aren't and build an interactive personalized message feed for your customers. And then guided setup. How do you actually onboard all these channels? You're looking at Journey Optimizer, maybe you haven't fully implemented it yet or you're just getting started. How to automate lots of that setup to bring a new channel online to drastically reduce the time to value. So, let's get started.

In-browser messages. Super excited about this one. Here's some screenshots from the product. This is a way for you to quickly deliver a triggered pop-up on your website or an overlay or a modal that fully matches the look and feel, all the branding, all the configuration without touching code. So, this is something that if you know JavaScript, you could do this today. That's great but if you don't and you need to do this, this is really exciting channel for you. So, it's a brand-new channel in Journey Optimizer. It's live today, so, you could go do this as soon as you get out of this session, and you could go launch these overlays in minutes. A cool thing about these overlays. Flexible Triggers. So, you can have complete granularity of control across mobile apps and desktop web and the options will be different depending on the surface that you're in because of the context of that surface but with a shared componentry, you'll be able to decide exactly when and how one of these things gets triggered. No code-authoring with out-of-the-box templates. So, you'll be able to very quickly pick from a preset library of options Maybe you need a full-screen takeover or you just need a banner at the top or you need a light boxy pop-up or you can go custom and bring in your own HTML or code anything that you need.

It looks like this. It's amazing.

And then, Rich customizations. So, the ability to take over different parts of the UI, change the background colors, change the height and the width and the styling and all these things just with the sliders and the options that you'll be able to completely customize the the look and feel.

So with that, Roger's going to give you a demo of this right now. Okay, so, I'm at Fréscopa or coffee barista shop and I'm coming in here and this is actually the website that if you go to the mobile lab, we're showing this. So this is the actual site they're doing with the lab. So I'm just going to sign in and you're going to see a set of different browser based in that messages that take me through this experience. So this first one is, "hey, let's take a quick tour." So this is a carousel-style in that message, in-browser messages. There's a coffee quiz. Here's what happens. You can spice it up. You can earn badges and achievements. You can also subscribe to have your coffee delivered frequently. And I think, "hey, that's great. Let me go and subscribe." So I want to add to my barista here and I want this... Call me before you deliver to make sure I'm home. I don't necessarily want it sitting on the doorstep and I want it twice a week, so I'm going to subscribe. Another example of a message. "Thank you for subscribing." "Your journey has started. You're going to have lots of joy." So, I close that out, and then, I want to start to see, I need some K cup. So, let me go look and select some coffee. I see this K cup and I select that one to dig into it more. I get a little, "hey, COFFEE7" at checkout. Use that. I'm a new member. Present that type of thing and I can put that in checkout. I say, "hey, wow!" That's fantastic. I add it to my cart. Product added to my cart. So, you can see all of the different styles of modals, views, pictures, the different triggers, the style of the carousels, all of that extremely simple inside of Journey Optimizer to bring along for a single-channel and a little bit later, we'll show you what this means on the Journey Canvas to start to bring in that messaging or other channels that Jason's going to talk about together in concert. So, that's a little bit about the types of things that you can do with in-browser Messaging. Awesome. So, I'm super excited about that one. Again, live today. You can go try it right now. Code-based experiences. So, this is another channel that's live right now. You can go do it. This is in our product today. So, I call this programmatic inbound. What do I mean by that? I mean that it's an API. It's an API that any surface that you have, visual or not, can call and get a response from Adobe back with a payload, HTML or JSON, to power and experience in an on-demand way. So, instead of us calling out to some engine or your service, it's you calling to us. Those payloads can be static. So, it may be a feature flag to change the search results screen. You're testing, you're doing an experiment, you're doing some kind of product release in your mobile app and you need to validate that it works. You could do a static payload like that to change that experience for everybody or some percentage of your customers or they can be completely personalized down to the individual. For like, when you call the call center or when you engage with Alexa or some kind of a digital voice assistant or when you're looking at your connected television and the app supports this. So super excited. The idea of this, again, is to open Journey Optimizer to every single inbound channel that you might have in your ecosystem.

Talk about where you can use it. I've listed a lot of these already. Point of Sale is another one that I think about a lot. ATMs, if anybody's feeling spicy, let me know. Kiosks for sure. If you're thinking about brick-and-mortar, you have actually talked to somebody today who's talking about in the QSR space looking at bringing Journey Optimizer into an order kiosk in some brick-and-mortar location. So, super excited about this one. Roger. I don't know who in here is a Creative Cloud customer but there's a little Creative Cloud install desktop app as well that we do reminders and push things too in terms of your subscriptions renewal. Here's a new app update. Those are also areas on an installed desktop application where people are using code-based experiences. Yeah, exactly.

One more channel I'm super excited about. Cards. So, cards is the non-technical, no-code, little brother to the code-based experiences. So, cards use the same infrastructure. It's a real-time inbound call on any surface that you need but it gives you a visual or a form-based authoring mode for anybody who doesn't know HTML and doesn't know JSON to go build a set of messages that can populate a message center or an inbox or some kind of a feed of these cards and these cards can each be individualized campaigns and so every one of your customers can get a different set of cards in different orders. There might be one global card that everybody gets and that's pinned to the top but it's the ability to very quickly log in to our system and build these personalized feeds for your customers and have them in a very visually appealing way in your mobile app or on your website or really anywhere that you need them.

So, that's three new channels that we're working on. There's a lot more. How do you get these channels going? Guided channel setup is something I'm super excited about because when we worked through this, we reduced over 80% of the clicks that it requires to set up our SDK, validate a channel setup, and send a message. Send a push, get an in-app message, get an in-browser message or something like that. I actually have a demo that I'm going to walk through in PowerPoint here, of how we actually automate this SDK configuration, how we streamline the validation using our assurance tools and how we actually get to that in message. So, let's walk through it.

You're in Journey Optimizer. You just got a new app. You've published that app to the App Store. You actually want to integrate Journey Optimizer into your mobile app to start sending mobile push messages or in-app messages. So, you log in to Journey Optimizer, you say, "Okay, I'm going to get started. Let me launch this Quick Start." You're immediately taken into this kind of a wizard where you select the name of your app in the platform or on iOS or Android using Swift and all this stuff. So, you put some quick information in and then, you start this auto-configuration process. This is using all of our APIs behind the scenes to automate hundreds of steps, dozens of pages, dozens of screens into something that it just runs through and gives you a code payload. So, it says, okay, we set up all your configurations. You need to add this code at the top of your app. You need to add this other code, little snippets into your mobile app or your website but it's all completely customized and built for you and all you need to do is copy and paste it. Once you've done that, you can tap into our assurance tools. This is a validation tool that we've built on Adobe Experience platform that live pairs with the app in hand on your device, and tells you that everything is working okay, connects to your app, and then runs through all those validation steps, and then says, yup, it's great. You've validated the mobile app, the SDK itself is installed properly. Perfect. What channel do you want to use? So, you've got the SDK. What do you want to do with it? You want to send in-the-app message? You want to send a push message? You probably want to do both. Let's say you want to send a push message. So, now you've got the push message, you need a couple of configurations for that channel. We need to make sure that we have everything with the push token and that authentication. So, again, you go through this setup. You validate it very quickly. It's going to give you that immediate feedback. You've never left this screen by the way. You're just in this one screen in Journey Optimizer and now, you can see that you've got one channel ready for activation in your new SDK, in your new mobile app, that's push. You could go ahead and configure an app next, if you wanted to. And then, it's going to say, it's available for engagement and it's going to be in your drop-down in Journey Optimizer and then you could go send a push message to all your customers immediately.

So, what I just showed you is 80% fewer steps, 80% fewer clicks, automating lots of stuff and saving you guys a lot of time and really helping your mobile app developers or your web developers get up and running with a new channel in Journey Optimizer as we roll them out. So, it's one example in terms of one channel just for mobile. We have a number of partners or people in here that their roles on implementation and structure and of going to different launch or download the SDK, go into the admin portion to set up the channel, or go here. Five different places to do things. This streamlines all of that. Example for mobile but same thing for web and other channels, just to really drive that value. Yeah, thanks. So, let's talk measurement.

Journey Optimizer. Optimize is in the name. You have to be able to optimize. How do you optimize if you don't know what's going on? A lot of really exciting stuff for measurement. I want to start with the biggest one. We are completely reimagining measurement in Journey Optimizer. Every single report that you see in this product is going to change and it's going to be enhanced by the power of platform and drive consistency with customer journey analytics. That means, when you log in to Journey Optimizer, you are seeing Analysis Workspace. You are seeing the same kinds of visualizations that power customer journey analytics as a platform-level service to drive those consistent visualizations and those behaviors with single-click access to go into customer journey analytics, if you own it, but if you don't own customer journey analytics, this is still the visualization that you get and it's still the reporting engine that you get. It's a massive step forward in terms of our ability to facilitate the data literacy and the storytelling that you want out of Journey Optimizer, it also helps us normalize metrics.

For goal-based journeys...

When you start a journey, what is your goal? What are you trying to improve? Are you actually trying to improve one thing and prevent something else? You're trying to drive conversions up but call center calls down. Having a unified set of metrics across all of the experience platform applications, that you can select those metrics in your journey to set a goal for. To have outcome-based experiences. So, every single journey that you'll run in Journey Optimizer will run against a goal that you've set for it and not only will it just have a goal but we're going to look at the journeys you've run in the past and help you benchmark yourself against your goals and against how well are you performing in this journey. How close are you? The visualization I have in my mind is like the bake sale and you've got the thermometer and it's like how close are we to our goal? Like that's the kind of visual that I have there on how performing are your journeys and how do we get you better at running better journeys.

Inside of this, you'll see some live reporting on the Journey canvas itself. you'll never have to leave the Journey canvas to see some of that initial metrics and your progress to that goal.

So, let's see it.

As Jason was saying, what this also does is sets us up not just for the bake-sale perspective but what I mentioned earlier, moving from, "I want to deliver an experience" to "what outcome do I want to have?" Either for the users or for the business. And then everything that you saw on main stage and you'll see tomorrow in some of the keynotes with Generative AI. You can picture where we're going with this. Once you know what the outcome is, say, what's the best journey, what's the best experience, what's the best channel? So, once you can measure, we can then apply the nouns or objects and we can generate suggested things and be able to test on that. So, here's the view. I'm going to walk you through just in terms of timing. Where we're at is June for public customer access by request. So, if you're interested in some of this, or you want to participate, reach out to Jason or me and say, "hey, we'd love to participate on these new AJO reports." So, here I'm in Adobe Journey Optimizer. You can see this Reports tab. These are now the combination of all of my journeys. So total entered for this time range, right now they're individual reports, but you can see here, here's how many people are entering, exits, conversions. I can also see a list of various journeys that I have. Same thing for campaigns or even aggregate channel reports. You'd be able to do here. But let me drill in to the summit 2024 report, you can see this is an individual journey then. Some new visualizations in terms of that flow diagram with some of the numbers. The other thing you'll be able to do is collapse some of these together and get numbers or stretch them out, move them around so you can start to play with this in an interactive way in terms of that journey. You're seeing I've got different actions or journey steps. I can now also then drill in and say, "hey, I want to look deeper and I want to filter." So, you can start to then filter and again, this is the same data as a journey analytics from the data set in AEP, and then we're just presenting this in Journey Optimizer. So, maybe I want to filter this by a specific channel. Say, I want to filter this by email. So, you can start to then look at this journey report just by individual channels, if it's an omni channel report. You also might want to then, if you have an experimentation running, drill into some of that experiment. Which variant is driving the highest lift? What's my confidence? And start to dig into there. The other thing that I can... Oh sorry, I skipped a part there. The other thing I can then do, I've been in Journey Optimizer this whole time, and we have access to Journey Analytics as a brand and I want to go in. I want to do even more deeper analysis on breakdowns or dimensions. There's this Open in CJA button. Open that in CJA. We take the exact same report, open that up in CJA. With the information, you can see I'm still in the Summit 2024 report structure, and I can come down and start to look at this and then maybe I want to even break down this report, the action report, even more. I can drag in some of the dimensions, start to look at that report, do a breakdown by that through the power of Customer Journey Analytics. The other key thing here then is if I really like the way I've built this report, I can also then save it and keep it as the report or edit what I want to do. Maybe I want to add another key metric. So I come in here. I want to add purchases to my overall visualization of my headlight reports. And now I want to save it. So I can save this report and it updates that. And I have this little toggle. "Do I want to update the report back in AJO?" If I say yes, then the next time somebody comes and looks at this specific journey report, they'll see the breakdown of total purchases on that report as well. So, the analyst then and CJA can share things back to the journey manager or channel manager inside of AJO. So, save that and I've got my report. Last thing, a lot of you have asked, "hey, can I schedule reports?" So, yes, bringing the power of some of that to be able to schedule report, deliver me a PDF of some custom report I've done on a specific time frame. There's a number of customer journey analytics sessions throughout Summit. So, they'll be talking about this as well. Jump into those if that's of interest to you. I am pumped not only about the unification of the personas that use Journey Optimizer and Journey Analytics and having that storytelling through data together but I'm more excited about what the foundational metrics layer allows us to do. So, this notion of being able to pick a metric, set a goal, drive to that, and then use that for intelligence and predictive, it is really going to power your business strongly going forward and help drive things. So, super exciting work between AJO and and CJA. Love it. So, let's talk orchestration. We tease this out with goal-based journeys. So, what else do we have going on? So, Right now, we're talking about orchestrating journeys. Earlier, we talked a little bit about orchestrating audiences. So, with journey orchestration, I'll talk just super quickly about what I mean. I talked, at the beginning of the session, about the principles of active listening. Listen for a signal. So, something happened.

A person walked in a store. A person called the call center. A person graduated into a segment or exited a segment. A person just completed a purchase. Apply conditions and rules, and then trigger those actions. Those actions could be to send them a message or it could just be to update some database or it could be to enrich an audience, and then send a message sometime later.

Innovations in orchestration. Redesigned journey canvas for all the practitioners out there. This I hope is very exciting. So the actual interface of our journey canvas is being totally redesigned with clearer, I hope, node access, with more information about each action or each condition without double-clicking. With 24-hour journey reports, what I mean by that is, reports without ever leaving the journey canvas. To be able to very quickly see how many folks are going in one branch of your journey or the other branch. Global exit criteria. This one's a little in the weeds for the practitioners but the ability to say, "At any point in my journey, if something happens, kick them out." They're no longer part of this journey. And set that once. Your practitioners are going to love this because it's going to save them hundreds of human hours to be able to do this once at a journey level or at the global level, and say, "This is your exit condition, no matter what." Journey Simulation. On the main stage today, Ahuja talked a little bit about this but using AI and all of the data, if you should opt in, of your previous journeys to simulate the performance and execution of a future journey.

So, let's dig into that one just a little bit. So, here's an example of a journey that you may want to run. So, the model we trained on previous journeys that you've executed and who flowed through those journeys and who did what or who exited or what kind of errors there were on your data only and you can get estimates of the journey that you're attempting to run to see if you're building dead ends. Have you made a journey where nobody actually makes it to this branch or the balance is way more out of whack than you thought it should be or that you need to be to fulfill the business case that you've actually been tasked? And so the ability to simulate that journey before you run it will help you run better journeys over time.

Experimentation. Really excited about this one. That's my background. I've spent 15+ years in experimentation. That's where I got my start in digital marketing. It's probably the closest thing to my heart personally. So, what are we doing with experimentation Journey Optimizer? There's a lot of really cool stuff that we're doing. I'll start with acknowledging experimentation in Journey Optimizer was not built for an experimentation professionals like that's what they do every day. We don't need you, when you run an experiment in Journey Optimizer, to necessarily understand statistical concepts, or error rates, or power, or what model is being used. We want you to be able to very quickly validate that you have some ideas and which options are working in any channel or against any kind of object that you want to run experimentation against. But we do want to give you tools to run better experiments. So, we're building a sample size calculator in Journey Optimizer, not ever popping out of the UI, that's going to give you what we call effect detection guidance and that's going to help you understand how many variations can your traffic support and that's going to be different in an email versus your website versus your mobile app. So you'll be able to very quickly say, "Hey, I've got an idea for an experiment. How many versions of this content can I actually support statistically in the model?" And quickly get guidance on if you've got too many or you can afford a few more. We're going to help you scale winning variants. So, that's pushing the winners. If you run an experiment, especially, in an asynchronous channel like email, where not everybody, maybe, is seeing it at the same time or a journey that's taking a month and people are entering that journey on different days of the week or different days of the month. If you get signal very quickly that B is better than A, be able to route the rest of those entrance in that journey or be able to route the rest of those emails to that experience, so that you're not continuing that test beyond when you need it.

Non-destructive traffic allocation. That's a fancy word. What I mean by that is the ability to start an experiment smaller if you're more risk-averse. So, you actually don't want to run the experiment at 50-50 to a 100% of your audience. You want to run it at a subset of that and then, as you know that, okay, you didn't set everything on fire to actually start to incrementally roll that experiment out, to auto-scale that experiment up. A lot of really cool stuff here in experimentation.

Decisioning. What are we doing with the decision engine and decisioning principles? Really cool stuff, again, this Decide layer. You'll notice, and I haven't spoken to this directly, all of these concepts directly map to that vision statement. That North Star that Roger just mentioned about how are we organizing our thoughts but also our teams and our resources, and our capacity to go attack these areas.

So, let's start by just talking about what decisioning actually mean for us. When we think about decisioning, we always do the who, what, where, when, why thing. Decisioning as a problem of arbitration. Decisioning could be personalization. It could be what offer do you get. It could be picking the winner of the fight of who gets the banner on the homepage, for how long, and to what audience. And that can be a very political thing. Or it could be you know the decisioning of how often should you get something or how many times until you stop getting it if we're not getting the signal back from you. So the who, based on the rich profile data on experience platform and all the attributes that you're sending in, all of those signals, the event signals, the transactional signals, the contact signals, all of that goes into a what, which is some representation. It could be a visual representation. It could be instructions to a call center agent to pivot the script and say something different or mention something organically in the conversation that you're having while you're on the phone and you have no idea because you're not interfacing directly with us. The how. The rules and the when, and the where and the how, all come together in terms of who gets it, why do they get it, under what conditions do they get it, when do they stop getting it, and then what happens when they actually complete the action that you want them to take.

That's all leading me to experience decisioning. So, for those of you who are in Journey Optimizer today and you might be thinking about how you use decisioning in Journey Optimizer, we often have described it as Offer Decisioning, or Decisioning against content offers. Who gets the right offer, win. Next best offer is probably something that you've also used or heard. We want to move beyond just offers as a construct of decisioning into opening into decision and against all of those nouns that Roger talked about, Decisioning against journeys, decisioning against channels.

What's the right channel to actually send me on in a journey, or what is the right journey for me, or what is the right path of a journey for me? And then maybe, what's the right product in a product recommendations use case? So, experience decisioning is in beta now and coming out very soon.

It's a re-imagination of our entire decisioning engine. So, we are abstracting some of the stuff with just about offers and generalizing that across the ability to apply it to all kinds of different concepts. How that's going to materialize to you guys and the product is All New Workflows. You're going to be able to see decisioning on the journey canvas. You're going to be able to engage with decisioning directly in the code-based experience composer. We're splitting apart the content representations of the what. What are you actually decisioning against? Against the how, and the whens, and the whys. I'm going to show you a couple of screenshots about the decision and selection strategies. So, the ability to very granularly describe eligibility. Who's eligible for this or how many times should they see it? And then decisioning beyond content offers. So, we talked about all the different things that you'll be able to apply decisioning against.

So, here's an example of how the workflow will actually work. So, you have the Decision Item Creation. So, what are the attributes that I want to use in decisioning? Be able to actually impact our AI models with different features or different profile attributes to influence that model. And all of the frequency capping and those rules. Different kinds of selection strategies that you'll apply that might be in terms of what's the population of the offers or the experiences that you want to deliver and how are those different for different channels or different contexts. And then you'll create all these. The selection strategy. Selection for decisioning, decisioning could be because I said so. It could be a rule-based approach. It could be a prioritization-based approach of if their propensity is X, show them this. If their propensity is Y, show them that. It could be something a little more complex than just rules. It could be a formula. Something that's dynamic and waits and takes into account multiple kinds of data points and then weighs those against each other for some kind of a value or an outcome or you could just let it all go to the AI and say I want to optimize against a goal or personalize against a goal and let the AI figure it out. But you'll be able to set those decision policies and then execute those in Journey Optimizer. So I think we're going to see a little bit of that.

Okay so what you'll see if you're familiar right now with offer decisioning moving to experience decisioning, this will look a little bit different inside Journey Optimizer. But what I'm also going to do right now is connect a couple of those verbs. So I want to connect first a decisioning and experimentation and then we'll connect decisioning and orchestration. So in here, I'm going to create a new campaign. Some of this, I'm going to breeze through really quick because I want to make sure we have time at the end for some questions.

We'll do the code-based experience so I created that campaign. Now, I'm going in to say, "What is my JSON payload that I want to deliver? So there's really no visuals here but I'm going to come in and say, oh, first I wanted to create experiment, so there was a Create Experiment toggle at the top. This experiment is not on content. I'm going to test against a manual or rules-based model against an AI ranking model and say, I'm going to experiment against decisioning. So I'm going to create two treatments and then within each treatment, I'll go in here to Treatment A and start to grab a decisioning policy, create that decision policy. So, I'm going to do a manual ranking for this one. Let me add the strategy that I've already done, so this would be your list of selection of strategies that you have. You can see I've got priority-based ones, model-based ones, so this one's going to be manual. So Selection A and my fallback is a Winter Jacket for this example. So now that I've created decision policy, I'm going to go in and then also decide how I want the visualization or the treatment, the presentation layer. I want to select the product name, the product price, and also the image URL of the item. So I've done that and now that's finished up in terms of the Treatment A. Treatment B is going to look very similar from a presentation layer but first of all, I'm going to go in the decision policy. It is going to look a little bit different because I want to select the rule or the AI-based policy here. So this one is now using the AI model for this decision and policy, and you can see selection strategies, see that's just the name, you would have your model name, whatever you're doing with my fallback of winter jacket. So that's created. I'm going to use the same presentation layer here and then come in and that's my second one. So, let's then fast forward and say, I've been running this now for a few weeks. I would have a report that starts to show, it's been running for 26 days, 11% lift on the model. So, this is where you can start to say, does my model perform better than manual ranking in my rules and that's a new way of experimenting and again, back to those objects, I'm not experimenting on content. I'm experimenting on my decisions and my decision policies. So, a little bit of different way of thinking about it but it's more expansive and really opens up. A lot of customers say, "I want to bring my own model." Super. You can test models that your data science team might be doing or other types of things as well. So, really powerful functionality when you start to bring experimentation and decisioning together. Let me move briefly then to bringing decisioning and orchestration closer together across omni channel experience. You'll see here... oh, by the way, that previous one like Jason said, experience decisioning here, Q3, next few months, we have it ready. We're just tightening some things up. And then that will be in the next quarter and a half or so. If you're interested in seeing that or participating and getting access to it, let us know. This one I'm going to talk about content decisioning because that's the one we're doing in the short term. Similar time frame. Q3. But you can also see a shout-out here of branch decisioning on the canvas as well. But I drag in this content decisioning node, I can select that, go in, and then I'm editing the content. So, it's a decisioning node in the canvas where orchestration is going to call out to experience decisioning and say, "What do you have for me?" And so now we're going to create what that decisioning policy looks like. So, again, I'm going to run a promotion. I'm going to name that and then I'm going to go add a strategy sequence. What are my selection strategies that I want to have and I can also set the fallback. So in this case, I'm going to pick two different selection strategies with the ranking model, both based on audience. So, selection strategy A is the first one, the fallback from there is B, and then the main fallback is winter jacket. So that's all created. Let me fast forward. Then I'm going to pick the same thing of the name, price and the image URL as my presentation layer. Okay, so that is now embedded and this is usable throughout the whole life of the journey. So, it's contextual sitting there, in life of the journey. What some people are saying is, "I might want to send an email. somebody come back to the website, and I want to make sure the same decision policy was ran even though I sent the the email today and they come back tomorrow to the website. I want that informed through this. So, that's what we're going to do then here. We're going to bring in some channels. This is actually a custom action. So, a lot of customers use Adobe for the delivery. They also have a mix of other destination systems they might be using, or execution systems. Maybe they're using somebody else for email, somebody for push, somebody for internet messaging. This custom action allows you to take that content from that decision and deploy it through any channel that you want, that might not be native through Adobe. So, that's what this first one is in the custom action and I'm going in. I'm adding this expression. I'm picking that content decision, in terms of... right here, in that custom action. Now, that payload is being sent to whatever that destination might be. So, for this example, let's just say that one is email.

So, I'm going to send an email now, after that, and then, let me bring in web. So, you guys saw earlier, the browser-based messaging. This one's more on the full website and I want continuity there. So let me edit this. I go into the edit. This is the visualization from Journey Optimizer right now for the web channel. I come in. Here's my website. This is the content space or the placement that I want to do something, so, I can come in here and select the decision policy. I want to reuse that same decision policy so I would select Reuse that. It's the same winter deals that I'm doing for the email and reuse that and then it comes in and so now it knows that's the decision policy. Now let me do the presentation layer. So it's a little different, when it's constructed through us, you can pick the presentation layer right here. I select I want to use product name for right underneath the image and then I also want to select the image. I want to do the product image here, so, now it's going to dynamically select based on what the decision policy came back with. It'll pull the name and the image from there. So I've now put intelligence behind this and I've driven consistency between email and web and potentially, even multiple execution systems within or outside of Adobe to drive orchestration and decisioning together. So, that's really... you know what? A lot of power that we have here that we're bringing forward, that marriage of decisioning and orchestration coming together with content and channels. So, it's lighting up for those boxes on that chart that we had. Awesome. So, we're on the home stretch guys. Last little bit here.

What else is coming? More embedded channels. Roger showed you some code-based workflows. This will be visually embedded in every single channel that Journey Optimizer supports. New content catalogs to support product recommendations, content recommendations, media recommendations. And a lot of new AI capabilities. AI model insights. So, last part. Two more slides. This going to be a session tomorrow. I'm going to highlight it here in a second. But you heard a lot about Gen AI on main stage today. You saw a little bit about the AEP AI Assistant. So, I wanted to contextualize that for everybody here and make sure that you are aware of the next session tomorrow. What does that mean for Journey Optimizer? Two key things. Number one.

The AI Assistant for knowledge generation and productivity. That is using the natural language chat experience, the data that's trained on all of our publicly accessible information that Ahuja mentioned, our documentation, our experience league communities, as well as optionally, data about your system to be able to help you troubleshoot or ideate or help with productivity, get insights about audiences and how you've created them. That'll, of course, be available in Journey Optimizer against all of the native Journey Optimizer concepts.

Then, on the creativity side. Content generation. Variant generation. Being able to generate text or assets or both, or even entire layouts of emails based on the prompts and signals of how that might actually work. So if you have a back-to-school campaign or the Black Friday, Cyber Monday campaign, and it's not just that you're going to generate a content that goes in a box on the email, you can actually generate the entire email itself and the layout of how many rows and columns and how big it is and what copy is going in each. So it's actually constructive generation of markup, not just assets.

Controlling the tone and then experimenting on those variations. The whole point of this for us is to not just say that some Gen AI content is better than what you have already. It's to be able to validate it statistically with an experiment. So, we're super excited about this. All of this is still in pre-release and we're working through when we'll make it available but wanted to set that aside for you really quick.

Running out of time. So, I want to respect everybody's time here.

What did we do today? We talked about what is Journey Optimizer. Why did we build it, and how did we get here, and where are we going? And I hope that you were able to get a glimpse of our strategy.