The Art of Marketing Architecture with Biswajit Chakraborty

Speaker A: Hello everyone and welcome to the Marketing Roundtable, a digital marketing podcast. My name is Brandon Jones and I am the producer along with the paid media strategist. And I'm going to hand it over to Brian Cosgrove, the host for today.

Speaker B: Hi, I'm Brian Cosgrove, principal at Braindiv and an analytics strategist. And I'm here to speak with Biswajit Chakraborty.

Speaker C: Hello everyone. Thank you for having me, uh, especially Brian and Brandon. Absolutely. Brian just mentioned my name is Bishrajit Chakraborty, but you can call, uh, me Bis. I spent over 14 years, uh, the intersection of digital marketing and technology focusing on transforming digital landscape. I've been with Adobe over a decade. Uh, my current designation is uh, Managing Principal Enterprise Architect, which is a mouthful. So in a nutshell, uh, my role is uh, a marketing technology solution, uh, architect. That means I help to build architecture for a marketing technology solution. Uh, also I am a manager, uh, and there are multiple other architects who reports to me, does the same thing.

Speaker B: All right, so Biz, great to see you. I was really happy to see you at the Web Analytics Wednesday event that we had in Philadelphia. You presented some pretty amazing things there. And I know while we were there we had talked a little bit about your journey to kind of get to this enterprise architect level. How did you become a principal Enterprise Architect at Adobe? Where did you start off?

Speaker C: That's an interesting question. So, uh, let's start my journey. So I um, started um, with my uh, journey from India. So I uh, did my Masters of Computer Application from India, one of the renowned college called Vellore Institute of Technology, and then moved to a company, I'm not sure if you know, that's called Cognizant Technology Solutions. I worked with CTS for like two years, then moved to Adobe. So uh, one of the unique thing of that particular move is um, Adobe historically used to have Adobe consulting practice based out of us. Uh, but that is the first time Adobe started its consulting practice based out of India. And I was one of the first, uh, Adobe India consulting or consultants. Uh, I would say especially uh, that time I was focusing on Adobe Analytics.

Speaker B: Uh, right, right.

Speaker C: Was this the Omniture back then Omnichare or Site Catalyst? Different, uh, changes the name. So it was for Site Catalyst, then Omnicher, I think it is Omnichare, Site Catalyst and then Adobe Analytics. So when I joined in 2013 it was like a 10, 15 people team in Adobe India. Now it is more than thousand. Then in 2015 I moved, uh, to this Philadelphia, uh, especially for one of the big telecom company based out of Philadelphia. Yes, that was uh, Adobe's customers. I was one of the first uh, consultant from Adobe who was sitting on customer location five days a week. Okay. Uh, that was a unique uh, experience. I would say most of the time Adobe's consultants are like traveling or in remote basis. But that was one of the uh, place where me and few other folks were sitting on customer locations five days away. Uh, so during that time, uh, I think we discussed uh, during our um. We discussed during our uh, that webinar that when I came first time, one of my colleagues said okay, your name is Bishrajit Chakraborty. It is too tough to pronounce from today you are based on. So from that day I became base. Ah, from there I started learning uh, other Adobe solutions, uh, like Adobe Target for website personalizations, Adobe Audience Manager for creating building audiences. And from there moved to uh, this enterprise architecture role where Adobe started with this Adobe Experience Platform as the new customer data platform, uh, system. And I now became the, or I think two years ago I became the principal enterprise architect where I'm working with Adobe uh, Experience Platform, uh, all the other application services that Adobe Experience Platform has. That's all. Yeah.

Speaker B: So it's interesting. So you started off in India, you had gotten your master's degree. Somehow you landed one of these early positions with Adobe when there was. They barely had anybody there. And then you were able to move within two years from being one of the few people in India that's doing this to being on site in Philadelphia, halfway across the world. And you are now on site five days a week doing this work kind of, you know, uh, even though you're technically the company, right, the software company, you're so ingrained at that point that you're almost operating like you're part of the, you know, their customer, right. You're part of this telecom company because it forces you, if you're on site like that to get into their politics and how do things get done here and which department does what and what tickets do we need to use. So tell me a little bit about that to kind of go from, all right, you just got, you got a couple years of experience consulting as part of Adobe and now here you are and you are on site five days a week in this other company, in this telecom company, learning their business and trying to figure out how to keep all of your software program moving.

Speaker C: Yeah, that's uh, superbly like uh, I would say, uh, it's an eye opener. When I started with Adobe India it was all the customers. I had like remote customers or I would say iOS remotely and all the customers would be out of us. Uh, and uh, it was over the phone and then giving solutions to their problem. Now uh, suddenly I came five days a week and now my responsibility is to make our Adobe solutions implemented across all their departments. They are like consumer their business. They opened a new uh, system or new business units called mobile or uh, wireless system. And my responsibility was to make sure go and talk to all the teams and implement this Adobe analytics everywhere. So I felt that that was an unique experience where I was able to handle and also learn how to handle different teams. So I first started with the marketing team talking to them what is their marketing goal and then took those marketing uh goals to the technology team, said that okay, this is their marketing goal, come and do this implementation. Marketing team says okay, I don't have those requirements. Uh, I said okay, let's bring uh, both of them in the same room and say okay, tell the requirement to each other. So I am the, I am sitting in the middle gelling to both the teams and trying to convince them why this is important. And that's how uh, like uh, I started this like role.

Speaker B: I would say yeah, it's so interesting because it kind of forces you to think about the company's politics, the company, like how do they work, how do things get done here? And it's so much bigger than just being like hey, I got uh, the best training or I know this tool really well. It's like you now know the reality of what it's like to actually get the tool working on site where it needs to happen. Instead of just saying hey, if you have a problem you can call me. You're now proactively trying to get the projects done. So to uh, me that's just, it's really interesting and it feels like it's also a big shift beyond just being sort of a product expert.

Speaker C: Like absolutely correct. So it's like a product expert is always with me but like handling people uh, at uh, different nuances of the businesses, what is their business goals and understanding that prioritizing which one should be going first was really eye opener. And also I uh, would say handling people was the most learning things during this period.

Speaker B: Yeah, it seems like that's the part they don't necessarily teach you in your, you know, in your master's degree or they don't teach you, you know, sometimes when you're, you're on the support side, you don't get that exposure at that level.

Speaker C: Yep.

Speaker B: Okay, so then from there you went to um, so, so okay, so now you've been working on site at this telecom company for a while but you've taken on more clients and more companies that you've been working with. Beyond that, tell me about sort of how that progression, that chapter of your career.

Speaker C: Great. Yeah, so before telling that I need to tell some of the uh, things that happened during my career. Uh, like when you come from another country uh to the US you are in a visa. So uh, so after five years my visa got expired so I had to go back to uh, another country. So uh, in this case my company uh, said okay, uh, you are a good person, you're working nicely with Adobe, so let's send you back to Canada for a year. So that's how I came out of this uh, project. That is the primary reason my visa was over and I can't stay here in the us I can't stay here five days a week with this customer. So I have to go out. So I started taking other customers and also moved out of the country. I moved out of the country, uh, and immediately I moved to the country. The COVID hit 2020 January, I went to uh, Canada, 2020 March. Everything locked down. Um, so that time I was actually traveling to another telecom company based out of New J. And I was uh, in and out in flight from Canada to there and coming back. But suddenly uh, this uh, Covid hit and then I didn't to travel. Everything is over the phone. So I started understanding uh, more about remote work because I was always uh, at first I was remote uh, from India. But later when I five years I was with this customer in Philadelphia, it was a different experience. Now with COVID that experience changed. Now we need to again work on remote. But this time as I have grown up in my career my goal was totally providing business values uh, with different solutions. Uh, so that's. Then I started with this another New Jersey based telecom customer. Then I have taken multiple other customers like some uh, retail customers, some hotel chains. That's how I started diversifying my knowledge and skills uh, through that.

Speaker B: So that's great. So now you've got different industries, you've got telecom, you've got like you said, uh, different retail or hotel hospitality. So you've kind of gone through there. They all have different kinds of websites, different kinds of apps, different kinds of challenges in marketing and with regard to customer experience. Um, you know, in this case you're still working with Adobe so you have obviously a pretty big idea, sort of a lot of playbooks on how to put together different sorts of programs. Did you find as you were moving between them that you were using sort of different Adobe products? Were you using Adobe products differently in one industry versus another? Like how did you know, how did things translate or where priorities changed.

Speaker C: So uh, as I was telling uh, when I uh, moved to Canada I already became the architect by then and I already started working on Adobe uh, experience platform as a cdp. So by then CDP came to the CDP customer data platform and ah, that is the time like all the different customer data platforms started coming. Not only Adobe, Adobe's competitors started bringing a lot of these tools. Uh, and that time when I became an architect I started thinking about, not tool wise, I changed my thought process instead of tool thinking about the business values, uh, why a customer is getting the CDP and what is their business goal with the cdp. And uh, that's how I started changing so using the tool, thinking of the business values and drilling down to a different level called use case. So from the business value moving to the use case then uh, going uh, to the technological capabilities. So like I was uh, that's how I changed my approach. Uh, so instead of earlier when I was here in the Philly, um, then I was thinking about more on a technological side. As I became an enterprise architecture I was thinking about more providing business value. So as you would ask the questions about how my approach changed with different uh, customers I totally changed my technological landscape thought process at the first I changed it to the business objective as the first and then derive the technological capabilities. Hope that answers your question.

Speaker B: It does, it definitely does but it's still, it deserves to be leaned into. So I'd like you to kind of explain a little bit about um, you know, what are the kinds of major projects that these martech tools. And so I'm just going to mention a few different pieces that we've talked about. So Adobe Analytics, I'm thinking more like clickstream. How many people come to your website, where did they come from, what did they do when they were on your website? And that could also be your apps as well. We've got the CDP now before we talked about the customer data platforms, people had email systems and people had other databases with some of their customer details in them. But the CDP kind of brought a way to centralize it for marketing use, for offers for personalization and other types of things. And it sort of m organized it for that for that Use case. We've got other tools around testing. I'm going to show this version or that version, which version seems to be getting better results. I know Adobe has solutions there. You've mentioned audiences earlier. So now uh, here's a group of people that have done a certain action or are eligible for a certain piece and now we can send them emails or offers, you know, through any, any very, you know through various channels. Can you kind of tell me um, you know with all of these different tools and martech platforms, when you're talking about putting together, you know, maximizing business value, how do these come together like and as an enterprise architect, what do you prioritize, you know, when you're talking to the customer and they're saying we want our existing customers to subscribe to this service or we want our existing customers to use our self service platforms or something like that. How would you kind of take these pieces and turn them into a roadmap or a solution? How would you architect?

Speaker C: Yeah, that's a really, really great question. Before when uh, we talk about Adobe Experience Cloud I always give this example. The example is of a city. So think like Adobe Experience Cloud is a city. So in a city you have roads, uh, which is same as data flow. You bring that data of different buildings which is like different content system. Like uh, you're building a content. Okay, so that's that different services like personalization engines. Okay, uh, where you, you just talked about targeting a B testing and all. So that is the different services. So we as an architect is chilling between all the systems and connecting them to give a business value. So that's the core goal of an architect. So giving an example, let's give an example of or a real time example of a uh, retail customer. So if a retail customer uh, would like to do uh, or target a person if they have added something to cart which is sometimes we say abandoned cart use case. Okay. An easy use case. So what generally it works is you need to think about uh, there are multiple components. As I said, the first component is uh, the data identifying who abandoned that card. So gathering that information and storing somehow for Adobe's purpose, it is Adobe Experience platform that defies that set of people who has abandoned a card. Okay, so that is the data, uh next is the content. So if the person haven't done a card, what should I send? Should I send in banner next time come to the website or should I send an email, which channel? Right.

Speaker B: So here's a question. Who are all the people? Let's talk about the village of people involved. So you're the architect. Are you sometimes kind of coming up with what the. Whether it should be a banner or you know, are you coaching them through the process saying this is how this experience could work or who do you interface with?

Speaker C: Yeah, there are multiple set of people we interface with. One is the data engineers. They build data, um, engineers.

Speaker B: But they need requirements.

Speaker C: They need requirements. Uh, so first that's why I always say that thinking about the business objectives, the use case. So if the use case is abandoned, uh, cart use case, uh, so then the requirement is fine. Like all the people who have Abundant Card. So find out those information from somewhere. It could be from website, it could be from other sources, maybe just giving an example from website. So if it is from website, we'll get the data from website with a data engineer. Then I will guide the data engineer. Hey, this is how you need to bring the data. And this should be uniquely identifying this set of people. Uh, and that's how you should store it inside Adobe Experience platform. So that is the data pillar that is done. We need to go and talk to the content team. So contentive could be uh, the uh, customers who has their own content business or they might be using Adobe's content management system which is called AEM M. Uh, so we can talk to them and say that, okay, based on this particular use case, I will send a banner or I create a banner. That banner will be usable for both the places for email, sending an email and if they come to the website again we'll use the same banner. Okay, Content, uh, uh, we have called asset management. We will have a unifying asset and that content of image can be used in multiple purpose.

Speaker B: Right. So but you know, before all of that there's that part where someone just says hey, a banner. Putting a banner in front of somebody is the right way to kind of affect this behavior to, to offer someone or nudge someone in the right direction. And so how do you, I guess like is it a marketing person? Is it a just who do you start with? That's the business stakeholder that says I just need this done.

Speaker C: It generally starts from the top, uh, down approach. A lot of time, a lot of customers, the marketing lead or marketing executive said that okay, I want to grow this business by certain amount of number. And based on that we derive different use cases. One of these use cases, the abandoned card use case, and they said okay, let's test and find out if it is usable or it is useful to send an email or versus doing the website. Okay, so we can create a journey of that customer, uh, saying that hey, first we'll send an email. If they don't react on the email, if they come to the website, I'll show the personalized image, something like that. So uh, that's the power of Adobe experience platform, having the unique set of data with multiple journeys. Like you can send an email, post the email, you can say show it everything to the websites. But generally it starts with the customer marketing team.

Speaker B: That makes sense. And then obviously they don't always know all the capabilities that you're able to pull off. So there's somewhere in there where it's like you're asking them what do you want to do, what do you want to achieve? And then uh, you know, and then they're telling you kind of these vague requirements and you're like well you know, we could put a banner here, we could do something over here. There's all these different ways we could express you know, that idea that we could drive that, that outcome. And then obviously you have to get some approval with them to say hey, let's go, let's move forward on these different ideas items. And then that's when it sounds like you go to data content and then um, is there another pillar? I guess the journey is the journey itself and the journey is just the rules. Right? Is that the rules of when to show the content based on the data?

Speaker C: Based on the data? Yes, that is correct. So uh, again it starts with the business requirement. M as I just mentioned. But when we talk about the business requirement, uh, uh, we also uh, uh clearly tell that uh, you may not be able to achieve everything in the first go. So let's stagging approach like crawl, walk around approach like create a roadmap for a little bit, six month period of roadmap and say for the first one or two months we'll try to do a quick turn it out and see how system works. Okay. How we can personalize and then build upon it a uh, bigger personalized journey. So that's how I always take an approach like having crawl and walk around,

Speaker B: uh, approach that makes sense. So you dip your toe into the water here, you get something going that's personalized and then you're building out a more robust always on system.

Speaker C: Yeah. One of the primary reason is uh, to be very honest, uh, though I am from Adobe, I'll tell that uh, sometime what happens, some of our customers, they think from the technology first and they try to build the capabilities, they try to integrate our Adobe Experience platform with their CRM systems and then try to do it with some other systems and getting their billing data inside. And by then six months are over and they're not getting the value uh, of that tool. Then their business team comes and says hey, I'm spending so much money uh, and not getting enough return on investment. So that's what I always think of having a crawl work run approach. Start with at least small use case, showcase the value. Okay, this tool is working to be matured. So then get another first use a very simple use case like uh, next time the person comes I'll change the website banner. Then uh, I increase that use case. Okay, next time the person comes uh, I'll use a uh, banner as well as I'll send an email.

Speaker A: Right.

Speaker C: So I grow upon that previous experience

Speaker B: and then when it comes to the data side of that, do you build that out over time? Are you just saying like look, I'm going to do the first case here and I'm going to get just the data I need to do this to get started and then I'm going to need more data later and we can build that out in parallel or in the background while at least the customer is getting some, you know, in this case the company is getting some benefit.

Speaker C: Agree. It's an agile method instead of like waterfall. So as uh, I always like to uh, give value to the customer with an agile method. So first do something we have and then build up on it uh, because

Speaker B: we have some roi at least before six months is over. We just paid for this for six months and we still don't have a lot of value yet.

Speaker C: Correct. Um, that's what we faced with a lot of customers. That's how we started changing our approach, saying that okay, don't boil the ocean, uh, instead start with a small use case, activate that and show value and then get more funding and bring more data. Because what happens uh, a lot of time, uh a set of customers or a uh, customer, a set of groove in a customer, they pay for this particular tool. A lot of other team members, they're thinking okay, why should I work for this particular use case? I, that is not my uh, business goal right now. So uh, to do that like this, breaking the silos, right? So first focus some values and then it will go to their higher level, say that okay, now we are getting value out of it and then uh, that particular executive will talk to the technology team, okay, save this data and then we'll get more data and build up on it. That's how it.

Speaker B: That makes sense. That makes sense. So, yeah, I just, I always think about the people I feel like, you know, in certain cases, if the client or the company doesn't have the right people to interface with you, or you don't have the right development support to get the data, or you don't have the right creative support to develop the content, you end up in these. There's lots of possibilities for bottlenecks. And if you have people that just want to, like you said, they want to bring all their CRM data and they want to do everything data wise before they do anything and execute anything, then you end up in these timeline challenges where you're like you're paying money but you're not getting the return that would help you pay for this.

Speaker C: For.

Speaker B: Right, for all of this. And it is a long term investment you would say, right? I mean, uh, to set this up, you get more benefit over time. The longer that you have it, the more completely it starts to become integrated and even the more trained, I guess your team is on it, right?

Speaker C: Absolutely. Yeah. There are always this people, uh, process and technology aspect, right? So people, when we are thinking about a marketing tool, people is very important. Uh, there are process technologies, but having a mindset of personalization first is very important and quick small goals and achieving those goals will build upon uh, of all your personalization goal. So like if an executive is saying, okay, my goal is personalization first, but still they need to show some value to their executives, say that okay, how we are growing up on this personalization goal. So um, as uh, we executed these use cases, we are getting more data and the data is becoming mature. Ah, we are making more unified profile and then we can target more people and more different channels, different use cases.

Speaker A: This podcast is powered by BrandyW, a digital marketing agency that not only provides a diverse mix of digital services, but offers guidance, experience and true partnership to bring your business to the next level. We offer expertise in analytics, paid media, web development, SEO and design, just to name a few. We've worked with businesses on every level. From local operations to Fortune 500 companies across a variety of industries, we have expertise to help strategize and execute your digital marketing needs. No matter what step in the process you're at, we're there for you. From wireframes to paid campaigns, brand new experts in everything digital. Please reach out today to learn more about how we can help your business grow.

Speaker B: So you, I mean obviously you're an enterprise architect. You kind of know these tools backwards and forwards. I imagine that sometimes you deal with some executives that are just like, I want personalization. And you're like, that could be a million things. How do you help them put together a program where they can prioritize where to begin and how to get things going? Um, how do you demonstrate that? When they said, hey, I vaguely want personalization, and you're like, you turn it into tactical specific things that provide business value. Do you find yourself negotiating with different product managers or who are the people that you kind of need to work with to make that happen?

Speaker C: Yeah, so when we, this kind of discussion comes with you, like giving an ex. Let's give an example. If any hotel, uh, chain is saying, I want to personalize customer, uh, once they come to my hotel. Okay, okay, so let's take that example. Uh, I say, then we say that, which kind of personalization do you want? Okay, we have to personalize where, what is the medium? So we need to drill down different sections and based on that, uh, we see what are the capabilities we have in our product. So giving an example, uh, that I worked with one of the hotel chain and uh, they did one, uh, personalization in the inside hotel tv.

Speaker B: So, okay, the hotel TV monitor.

Speaker C: TV monitor if you are already a loyalty customer. Uh, so next you go and swipe your car to the hotel and uh, switch on the tv. You will see hey, um, uh, this name and welcome to this hotel. Okay, that is uh, sometimes very uh, like I feel very uh, nicely personalized and bringing those people said, oh, my name is here, so they know me. Uh, so it feels personal. So in that case, the how we have done so, the business, that particular hotel businesses say that, okay, we want to do personalization once the person comes to the hotel. Okay, how, uh, so and uh, even before how we first asked why do you want to do that? So they said, okay, these are the reasons because I feel people will feel more personal when they'll come to the hotel. Okay, that's a valid reason. Now how we will do that and what are the channels Based on that, we'll derive all the different small use cases and then try to execute one on one use case. Uh, and if that particular feature is not available in the product, go to the product manager and say that, okay, customer wants this and how can you prioritize this use case in the product? Something like that. That's how we deal with it.

Speaker B: I think it's so interesting. So, uh, when they do something like that, how do they know it worked? And I know that's a uh, you know, it's like, okay, great. Now it says welcome, you know, biz, you know. And you're like, okay, great. Um, we did it. Do they ever ask you questions on like do they do surveys? Like, how did they kind of triangulate? Was that the right way to do it? Or could we use different colors? Like how do you navigate that?

Speaker C: Yeah, so there are multiple ways after that. The first thing about doing any journeys, uh, or any personalization journey is the measurement like which you just asked a question. So Adobe has a tool called Customer Journey analytics, which is an omnichannel measurement system. So from any set of data, uh, we can bring, unify the information and say, okay, this is the report. Okay, the measurement report now coming. Exactly how to show the success of that. There are multiple ways. So like for this particular hotel business, they say that, um, I am going to send a survey after the hotel checkout.

Speaker B: Uh, sure.

Speaker C: We used to bring those survey data directly inside Adobe Experience platform and report using customer, uh, Journey Analytics. And so that. Okay, this is how their NPS score has increased. So NPS is one of the key

Speaker B: the Net promoters motorsport. So it's uh, would you recommend this to someone else? Right, that's the general question.

Speaker C: Okay, yes. Uh, and we actually uh, found out that this particular hotel chain got a really good NPS score which was more than 8. Uh, that was.

Speaker B: People would recommend them and that personalization was part of that, part of the driving factor.

Speaker C: Correct. And you said that, um, like should we use this color versus that color? So that is generally we use one of our tool called Adobe Target. So that actually can easily do uh, AB testing or multivariate testing and it can give a report which one was more successful. And we tie that report with this NPS score, Net Promoter score inside Customer Journey analytics and show that the red

Speaker B: ones and the green ones. And so we take this version and we look at the NPS scores of those people and we take that version, we look at the NPS scores of those people and we say this one's getting better scores. This is the version we're going with.

Speaker C: That is correct. That's how we do it. And that's how we can tell that, okay, going forward, uh, this red one is the winner. So going forward everybody will get this red one. Something like that.

Speaker B: Okay, all right, so that's, that's exciting. Now this all leads back to the, the data. So a hotel chain, a lot of times, you know, they have like rewards and loyalty accounts. So you know, are you a first tier level or you know the gold, platinum, whatever, you know, whatever, whatever their, their system looks like. Um, so that's when they get a lot of this information. But there are still a lot of businesses out there that don't have a ton of information about their customers. You know, if they didn't already have this legacy loyalty program. And it seems like that's becoming more and more important. Can you kind of talk a little bit about that?

Speaker C: So uh, as the main, as this time progressing with cookie less world. So let me give you quickly information about quickly cookie less. So the website, uh, uh, when you go to any website, website sets a cookie in your browser which is like a small code tell that, okay. Uniquely identify you every time you come to that website. So earlier uh, most of the browser used to support third party cookie. So what is the meaning of third party cookies? That if you go from one website to another website you can still identify that person. Now third party cookies are going away. As third party cookies are going away, uh, it is becoming harder and harder to identify a person cross domain from one website to another website. So that's why getting first party data is very important.

Speaker B: Right. So the fact that you were on a site that maybe shows an interest category, you were on a site looking at, you know, about vacation, you know, spas or something like that and then you come to this hotel site we would love to know and promote, why don't you look at our spa services? But we can't really get that information from the other sites. We don't have a way to exchange that with them without, without these third party cookies. So um, that means that we just certain things that might have been possible 10 years ago, 15 years ago just aren't possible now.

Speaker C: Yeah, there are ways of doing that. Uh one, uh, there are some data providers, uh, or it's called a probability probabilistic matching provider. Like if you heard about Axiom or Merkel, uh, they generally provide uh, probability probabilistic data. Say that okay, maybe Bishrajit have such spa, uh, they don't know if this is Bishrajit, maybe Vishwita search spa. But based on that you can show an ad. So that's called uh, like a data provider. But generally because this third party cookies are not there, um, are not going to be there in the future. Um, it is very important to have a first party data collection. Having some reward based system. We always say having a reward like provide your email address and interest, I'll give you 10% discount, something like that.

Speaker B: Yeah, I mean we're moving in a world where privacy is very important to people and they don't want site, this site knowing information from that site that they're, they're, they're less interested in that. And it sounds, you know, a lot of the browsers, Apple, Google and other other organizations, they've, they've restricted some of that communications.

Speaker C: I did Google so far Google still uh, have these third party cookies, some third party cookies. Uh, Apple is very strict with that. So uh, that's why especially Apple launched few years ago called ITP Internet Tracking Prevention. But they stopped all this cross domain tracking. So no more cross domain tracking. And if you think like Apple all the mobile browsers. So in the US more than 50% people use iPhone.

Speaker B: Yes.

Speaker C: All their iPhone browsers are part of that. So people think that mobile is very less, but mobile is not less. Uh we have done uh, analysis and we have seen that a lot of customers, uh, Data has around 50% of Safari users, they are coming from the mobile devices. So uh, we need to think about those set of people and how to uh, personalize them. Uh, and I think as we are talking earlier, uh, about some rewarding system like hey, I will give you some kind of reward and, and you share some of your data with us so that we can personalize something like that.

Speaker B: Yeah. I mean again if you know more about their, the whole idea of customer journeys means you kind of need to know who the customer is. And so you know to that point giving them some reason to have a want or uh, direct relationship with you means that they have your data. It's not, you know, they can have your data and just use it only for the purpose of providing you better service. But you need to come up with or invent a reason why they would want to log in or authenticate with you.

Speaker C: Absolutely.

Speaker B: Yeah. And that, and then that opens, that unlocks all the different ways that you could send a nice little signal to them that hey, you know, this is a little more custom for you. Um, what are some other really interesting things besides I think that was a really good example that you used for personalization with the hotel piece where people don't even think beyond the web or email. They don't even think about. Well what about the monitor in the hotel room? Um, what are some other, I guess interesting executions? Just where. Okay, you said they're doing it right, they're doing personalization really well.

Speaker C: Uh, before even telling the right, I will tell the wrong. Okay, so I'll tell my experience. A few days ago I was uh, so I went to vacation. Before that I wanted to shop, do my shopping for my shoe. Uh,

Speaker B: the sneaker or what kind of shoe?

Speaker C: Yeah, I was looking for a sneaker. And actually I am already a customer for them. They have my information. So, uh, I went to their website. Then they figured out it's Bishwajedar Bails. Uh, uh, I saw some shoes and then I just left because I didn't have time. And I will come back later and see. I started getting their paid display ads in Google or somewhere, wherever I'm going. Uh, because I was booking some hotels. I went with their, uh, Expedia Summer and then they started throwing all those ads. Uh, then I started getting some emails. And those things are happening in minutes, not hours, not days, like in minutes. I felt like it's like a innovation on my personal, uh, uh, privacy. I felt, oh, that's, that's, that's not good. So I left that website. I uh, basically left that company, uh, and purchased from a competitor. Uh, so like, what I feel about any personalization is, uh, needs to be helpful, it needs to be timely. And um, because sometimes it makes, uh, it gives a feeling like it's too creepy. So that's what I felt.

Speaker B: If they're over aggressive, you know, you showed them a tiny bit of interest and they're aggressively sending, uh, ads and other things everywhere. You're like, I can't escape them. My entire world now is this one sneaker company and then you ended up buying another sneaker. So, uh, so, so that's what not to do. Okay, now let's pretend that they did the right thing. What would be an exciting way for them to make it special for you?

Speaker C: Yeah. So if they did the right thing, they become more timely. Wait for some time, maybe 6 hours or 10 hours, 12 hours, and send a nice message with an offer. Like saying that, uh, you have seen this one and this one has. You can, if you do it in certain time frame, you'll get 10% off on the shoe. People will go and buy. So having an offer integrated with the journey would help. Uh, right, because lot of time when uh, everybody is price sensitive or most of the people are price sensitive. So when you show some offer with that particular journey, that helps. So, um, I'd, if I'm in their place, uh, I would first give uh, some time to think to the customer that, okay, what, let's wait, let them think what they're going to decide, and then slowly give a nice message saying that, hey, you have seen this product

Speaker B: and

Speaker C: if you come in this particular time frame, you can get 10% off or 10% off from the next buying. Something like that, some kind of. So those, that way you can engage the customer and feel valued instead of like, okay, you came to my website, now I'll run behind you. Something like that.

Speaker B: It's so interesting because I think a lot of times they're like, if I get a signal, just go all in on this. All the tactics, they just, they just, they just bombard you and what you're saying, you know, these timing pieces, I feel like, you know, so, so I know Adobe has the Customer Journey Analytics, Customer Journey, you know, Optimizer. I guess what I'm. There's, there's got to be a place where you can put in these timers, right? Some of the martech tools, they just are either on or off. You just tell it to go and then it just goes right then and there. And to even slow it down or put that delay in is extra work. Is that something uh, that some of the Adobe tools can help you with? Is that one of the use cases?

Speaker C: Yes. We have this, as I was telling earlier, we have this at the end, this Adobe experience platform that all those data and based on that we have different Journey, uh, pillar, uh, we have Adobe Journey optimizer. So that is under the Journey pillar where you can create a full fledged journey with a timer. So you can say okay, if the person visited my website, wait for six hours, uh and it can also pull the offer from the offer decisioning uh, and say that if they have selected this part set of product, give this 10% discount and then target them to an email or target, if they have already have an email, mobile phone, send them a push notification so you can have multiple channel communications and also uh, include or exclude some kind of offers with them.

Speaker B: So whose job is it to go and kind of like come up with all of these different journey campaigns? Do you find that they're like new job titles or who's the best at kind of putting together all the business logic for this. Who are the types of people that you work with?

Speaker C: So generally uh, with most of my customers this is the responsibility of the marketing team, right? Some are in the marketing who are knowledgeable about how to write a journey. So one of the great thing of Adobe Journey optimize that you don't need to write the code right and you can do an uh, like you can build the canvas, drag and drop and it will work.

Speaker B: But it's very user friendly now to do it to kind of put together that flowchart or, or.

Speaker C: Yeah, it is, it is very good for the business users, uh, to build that journey. So most of the time, with most of our customers, we are seeing that the marketing team owns this journey optimizer tool and they build the journ, uh, based on certain use cases or based on certain, uh, audiences.

Speaker B: Yeah, I'm just curious because someone may not even know that that's a job that they can, you know, a career path that they can go and pursue. It's like constructing all of these journeys and how all of these different rules work.

Speaker C: So, uh, one of. Recently I have seen, I, uh, have seen in LinkedIn. So this journey optimizer people are in high demand. Um, so as you are talking about this, okay,

Speaker B: are looking for career path, building journeys. This is it. This is, this is a place to lean in.

Speaker C: This is a place to lean in.

Speaker B: This is a way to really advance your career. So that's exciting. I mean, I feel like that's the one thing about Martech and how things evolve here, is that it's constantly creating new jobs. And I know one topic that I can't avoid bringing up here is AI. And um, I know Adobe does a bunch of things in the AI space. I mean, people still sometimes associate Adobe more with the creative platforms that they've made their name on. But, um, how is AI showing up in different parts of these Adobe tools?

Speaker C: I personally feel, uh, so we are in a particular critical juncture of this AI systems. Every day you are seeing this new AI systems are coming and at the same time we, we are also unsure where this AI is going. Okay, keeping that in mind, uh, I would personally say that AI, uh, innovation is happening and Adobe is one of the forefront of AI. As you just mentioned, there are multiple types of AI we can talk about, like generative AI or agent AI. Adobe is that creative business, like, uh, Photoshop and all. So we have this creative, uh, generative AI that you can build, write a prompt and it can give you a picture. Anyway, in this marketing system we are creating different agentic flows. Agentic AI system. If you think that Adobe's architecture, as I was telling you, at the bottom you have Adobe experience platform and there are three pillars in between. We are bringing this agentic AI orchestrator, so we will have different agents that can work independently.

Speaker B: So I get it. And I think people have heard the term agent AI, but they don't always know what that really means. So what is something that, for example, without AI happened this way, but now with agentic AI, there's less steps or there's something else getting done that otherwise wouldn't have been done. What are some of the ideas? Or just to help people understand what do some of the agents, how do they help? What are some of the things that they do?

Speaker C: Yeah. So by definition, the agentic AI system works is it can do your work automatically. Okay.

Speaker B: Right. Yeah, yeah. But people are afraid of that because they're like, oh, no, I don't want to lose my job. Yeah, yeah, on that.

Speaker C: I will always tell that AI is there not to replace you instead of help you. Uh, so don't need to be scared or afraid. Yeah, you need to learn AI. You need to learn how to run AI. That is important. Uh, but at the same time, coming back to this, uh, AI General agent. So we are creating different agents like Journey Agent, that can build the journey based on a prompt. Okay.

Speaker B: So you say, hey, I would like to generally not annoy people, but I want to send a message to them about six or eight hours later if they haven't come back to us and purchased sneakers that, uh, hey, here's a special piece. And then it can kind of help you build that journey.

Speaker C: The journey. Inter Journey. Uh, at. As of right now, it will not publish. You need to review everything, then publish.

Speaker B: Yes.

Speaker C: Future. Maybe it will publish by itself. Okay. When, uh, it will become.

Speaker B: And I think that's where people worry about it doing something strange that you never wanted it to do. And how do we, uh. Yeah, how do we give it all the point now?

Speaker C: I would say, uh, that it is a helper. So that means if you, with a prompt, it can write a. Build your entire journey. But you need to review and find out what are the missing links. Is it even going to work? Are you using correct audiences or not? So at first you need to do a little bit more trial and error and find out how it works. And then, uh, build on that.

Speaker B: It can't tell you what to ask it. Right. So that's the one thing is you still need to come up with ideas. You still have to have a vision. You still have to start thinking about journeys, about sequences of different actions and communications that you may want to share with people and what that might look like. Um, and then it can really build whatever it is that might help you do that. But you have to.

Speaker C: But your brain is yours.

Speaker B: It takes a human to coach it.

Speaker C: Yeah, basically, brain is yours. It is your helper to build that idea. Basically.

Speaker B: Right.

Speaker C: So you can. Like we are building.

Speaker B: We went from Having to hard code it to now you can sort of construct it with drag and drop thing to now you can sort of coach it through prompting.

Speaker C: Yeah. So through a natural language you can just write something and it can build something. So like we are building different agents like in agent, like site optimization agents, workflow agents, data engineering agents, uh, journey agents. So and it is a multi agent system. Like each agent can work independently and it can give its input output to each other. So that way if uh, you write like okay, build a journey based on this audience, uh and get this particular content, it will bring everything and give it to you. And now you go and validate and okay, everything follows fine with the publish that what is actually doing it is reducing, using a lot of time and investment from you where you can think more towards the how to use how to build the journey instead of. I would say you can invest more time towards uh, the strategy part instead of just actually building that.

Speaker B: So it's not just assembling. Right. It's like it's power tools, you know, instead of. So it allows you to move faster and get more done. So that's exciting. Um, so if you think about just these different areas, I feel like being the person to kind of do the type in the directions or come up with those prompts is going to be the high value, the high leverage role in the future. Um, what are your thoughts talking about

Speaker C: a prompt engineering kind of role?

Speaker B: Yes, but that doesn't, Just because you know how to engineer a good prompt doesn't mean you know how to engineer a good idea.

Speaker C: Absolutely. So that's what I, sometime ago like when uh, this OpenAI came up with this generative AI and then everybody was saying there will be a new role called prompt engineer and that all vanished in few days. Because I personally feel everyone uh, should know how to run an AI system. That's what I, my personal thing thought is that everyone should know how to write a prompt, how to write a good prompt. Uh, that way you can run an AI system. Okay. So um, uh, instead of having a prompt engineer, you should have a skill called prompt engineer inside you and with your marketing skills. So you need to have a marketing skills and with a prompt engineering skill if you might to have a coding skill, with a prompt engineering skill. So uh, you need to be able to write code or at least fix code, but you should be able to write a good prompt which can give uh, you a good code and then you can fix that code basically. Similarly like for any agentic system you can write A good prompt, it will build your journey, but you should know what to write exactly.

Speaker B: You should be able to review it and say, uh, yeah, want to tweak a few things here or there and fine tune it. So that sounds great. So, uh, we've covered a lot here. I'm kind of curious about sort of lessons from the field. Any kind of like tips and tactics. You've been doing this, you're on the forefront of doing this with big customers and large projects. Um, what are some things that people should start to really think about moving forward?

Speaker C: Uh, I personally feel that, uh, one of the things that always all customers should think when we are talking about marketing and technology is the business value. A lot of time I have faced with my multiple customers, uh, interactions that they, uh, think from a technology capability first, then they go to the business. I always tell them, no, come back, think about the business values, business objectives, then derive the technologies. That is the first thing, first advice I would like to give to anyone. Uh, and second thing is learn AI.

Speaker B: Get comfortable with it.

Speaker C: Yeah, you don't need to become a machine learning engineer. I don't know to write machine learning code, but you should know how to uh, give a good prompt so that it can help you, uh, to do quickly your day to day stops.

Speaker B: All right, well, thank you and I hope the listeners here have learned a lot about personalization and how to think about it from an expert. Uh, they've also hopefully learned a few things about from your career journey that might help them on their career journeys. Um, as this martech world keeps evolving

Speaker A: and changing biz, this has been great. The insights that you brought were extremely helpful. Uh, I'm sure to anybody who's looking to go into a career in this, especially as you mentioned, customer journeys, everybody in the audience look into it. There's a lot of opportunity there. So thank you for spending time with us. Learn more about Brain do in our marketing services on our website at www.brain.de. we can also be found on LinkedIn and other social media platforms with the handle Brain Do. If you learned anything helpful from this episode and if you'd like to see more from us, please consider subscribing or giving us a follow. Otherwise, we'll see you next time at the roundtable.