Conf42 Platform Engineering 2024 - Online

- premiere 5PM GMT

Optimizing Offer Management Through Enterprise Application Integration (EAI): A Data-Driven Approach

Video size:

Abstract

Unlock the power of Enterprise Application Integration (EAI) to boost customer engagement and retention! Discover how strategic EAI implementation enhances personalized offer management, increases conversion rates, and reduces costs.

Summary

Transcript

This transcript was autogenerated. To make changes, submit a PR.
Hello, everyone. before I begin, I want to thank Mark and the con 42 team for providing this opportunity. More importantly, thank you everyone watching the conference live. Or the recording time is valuable, and you have so many options what to do with your time. But you choose this opportunity to learn with me. So thank you. And let's start the presentation. Okay, the topic of my presentation is optimizing offer management through enterprise application integration, also known as E. A. I. But before we dive into the topic, I just want to introduce myself. Want you guys to know my, what my background is. My name is Vamsi and I have about 18 years of experience in solution design, business analysis, and product management. I do have a postgraduate degree in business management, specializing in IT and marketing. Interestingly, I also have a bachelor's degree in technology and electrical engineering, and it's from one of the top colleges in India. The last 10 years of my experience though has been at DirecTV. It's a satellite and streaming services provider. At DirecTV, I drive solutioning in the offers and decisioning space. I should say I'm fortunate that at DirecTV, I had the opportunity to live through the change and, maybe we can even call it paradigm shift off how customers viewed content on their televisions, right? So earlier, where it was beamed in via satellite through the dish to now, customers view content on streaming over the top. So that's another area of great interest for me, but perhaps a topic for another day. Now to wrap up, I've also previously worked as a software developer building order management software. And I've also done sales for cloud solutions, including infrastructure as a service and software as a service. So let's jump in now. So here's how we're going to go today. The agenda. First, we will look at when do businesses need to focus on offer management and what do they have to do to get there? And then what's the need or what's the business imperative for integration? Then I'll introduce what enterprise application integration is, also known as EAI. We'll talk about some types of application integration. And then I'm going to cover some best practices and methodologies that can be used to implement E. A. I and my favorite part of the presentation. The case studies. I wanted to share some case studies that I came across of companies that have successfully implemented. So we'll talk about those. And then, if businesses do want to go down the road of implementing, I want to highlight what the implementation road map will look like. And, we'll talk about what the benefits that businesses can reap out of doing such an implementation. Finally, we'll tie everything together with, offer management and then we'll conclude. So let's jump in. Okay, so when do businesses need to optimize offer management? so imagine this scenario, imagine a new business that just started. in the initial stages, it needs only a few internal software, say an accounting program, a CRM and an email automation tool. Now, let's say the marketing team comes to the sales team and says, Hey, we want to launch this email marketing campaign. Can we get a list of prospects, right? So the smaller business will have these smaller systems and maybe they're from different vendors. And so there is no integration between each other. So someone from the marketing team will have to go in person and, talk to someone from the sales team and ask them to pull this list of prospects. That person will then have to go into the CRM and pull this list and send it to the request and so on. So It takes a lot of transactions or iterations for the marketing team to be able to get the data. And then generate the email campaign. So that's how they run their businesses. And perhaps it's okay at a smaller scale, but what happens when these businesses grow, and customers increasingly have high expectations of personalized offers. And there is a need for customers to relate to the brand and they want to have this consistent, seamless experience across different channels through which they interact with, with the business, whether they're going on to the website or to the app or, they're walking into the storage. that business has a retail store, so they want to drive this interconnected business experience across various channels. the businesses, they want to make sure these customers come back. They want them to be loyal. They want them to make repeat purchases. Okay. And they want to increase the overall customer value. And this has been proven by studies. There are research papers that have proved that, 91 percent of consumers prefer brands that recognize and provide relevant offers. this, ties into the desire of customers wanting, personalized offers. Now, how do companies scale, right? So in order for companies to scale, they must start implementing more systems. they add HR software, a content management system. Maybe some billing and payment solutions and, some marketing automation tool, data analytics, supply chain inventory, instant messaging, and the list can, it can go on now. All of these systems grow within a business as it scales and any business that experience this level of scaling needs a way to connect and rearrange applications so that the data is shared and businesses can provide value out of connecting all of these applications. And on, customer loyalty, customer loyalty isn't just granted. It must be earned. And that is where enterprise application integration comes in. It helps provide that integration between these apps. Now, why does it matter? Why is integration needed? What is the business imperative for integration again? Imagine the situation where the marketing team wants to launch a campaign based on it. previous purchase history for customers. So imagine marketing team having access to that kind of purchase history. Or let's say they want to launch an email giveaway of a product and they have real time access to the inventory of the product. all of this helps having that kind of consistent view of whether it's customer data or company product data, supply chain data, having that kind of view for marketeers, it really helps in making their decisions. making their campaigns alive and then, the ideas flow through, instead if there is no such integration, all of these systems are siloed. So there are specific individuals or teams that have access to a specific system, but all of that is leading to a fragmented experience because the customer data, where the orders where the customer orders are placed, they're in the order management system. The payments are in the payment and invoices and the leads are in this year and the customer interactions are elsewhere. And so if there are these are siloed, then there is no value driven out of it. And then the marketing team or the sales team or the customer service team, they struggle to make sense with outdated or conflicting data. So that's where you know, integration would help. Now again, imagine a campaign has been created, they want to get through approvals. So typical way would be to go up the chain and have executives approve the campaign and typically that would require someone to log into the campaign management system and approve the campaign or, maybe send them an email and have them respond. Do an email with approval, but how would, if the integration is built straight into the instant messaging? So as soon as campaign is created, the, the messages go into, the, to, to instant messaging, and then it goes out to the executives, Through I am at seeking their approval. So for them, it's a one click experience for the executives, and it makes it much faster to go through the bottom line again, being that, process of integration. It helps remove these bottlenecks and in several of these business processes as well. Now we know a marketeers rely heavily on analytics, right? So business gather information from customers at every interactions. Each interaction is a data point and having that kind of view of data in a single place real time makes the job of marketeers or analysts that much more easier and much more efficient. So I'll talk more when we discuss the case study, but each interaction, it doesn't have to be limited. All of these interactions can be connected. Integrated together into a single view of all analytics and, integration can make that happen now from a sales perspective, right? I've told you I've done sales before. So I know how hard it is for me as a salesperson. When we call it the prospect to realize that they've already made a purchase, And that happened probably because the lead didn't flow through to me in time, as a salesperson. So having, again, that kind of interaction from the CRM to the sales team, it helps nurture leads quickly in a timely manner. And if leads, they're not nurtured on time, then we know conversion rates take a hit. Now, I also want to talk more about personalization for marketeers to be able to do personalization. If there is no integration, they have to rely on manual segmentation and often what happens is that manual segmentation, it's not real time. So data gets outdated fast. So by the time the campaign is launched, the segmentation has changed, or the customer has moved and so on, So again, having that kind of real time access to data will help marketeers react to changes real time. And, provide effective personalized offers to the customer. So there are studies again, that say if there is poor integration, there's a loss of about 20 to 30 percent revenue. And, so all of this is where, enterprise application integration, comes into picture and it helps. So now, let's talk a bit about that. what is EAI? what's the definition of enterprise application integration? the definition is that it refers to the seamless connection and coordination of various software applications. But the way I see it is, it addresses lack of communication. So if you look at the diagram here, there are several applications, right? There's a CRM, there's a supply chain management, there is a ERP, the catalog, there is several applications. So how do all of these tie together and how does the data flow through for business to make sense? And that's where EAI comes in. What it does is it enables real time data exchange between these systems. And breaking down the silos. So they're no longer individual systems alone. You know that the data in a meaningful way is transferred and translated across all these different systems and available for business users to make decisions. So again, the data is being replicated, but it's not really. Duplicate it in that it resides everywhere. the data flow is real time and more importantly, though, it facilitates that kind of agility in responding to market changes. So the moment business needs to react to market change, the data is already there and they're able to make reactive decision instead of having to fetch data from a different system and then go through an analysis and then make that change. okay, now there are a couple of different types of integration that can happen. I just want to briefly touch on that. I listed them, here, so there is a presentation integration that primarily deals with, having a UI level integration. So it merges, one or more applications, but it's more at the UI level, right? So it, so it's also referred to as screen scraping in some context. while this method is still employed in some places, it's considered outdated. The advantage obviously is that there is a unified user interface. So all of the information is available in one place. So perhaps, it is appropriate for smaller applications, but then it definitely is outdated and there is not a lot of sophistication. So it cannot be a comprehensive integration and a lot of customization might have to be, built in, the second type of application integration It's called the business process integration. Now, let's take an example. The CRM system, right? It has the information about invoices. And if it is integrated directly with the order management system, as soon as the order is placed in the order management, the order data flows through and the invoice is generated, right? or think of another scenario. Think of, let's say, as soon as an order is placed, it's connected to the scheduling system. So if, something requires a technician to go and install something, Then that data that flows through scheduling system and the scheduling also happens seamlessly at the same time, an order is placed. So the advantage clearly with business process integration is that it enhances, and improves efficiency, right? Like I said, the installation example, as soon as the order is placed, the installation is scheduled or the order info is sent to the warehouse for fulfillment. Now, all of these happens due to the, the kind of business process integration. So when that happens, it helps Automate routine task and teams can, concentrate on, more high value tasks, right? So there are some challenges with business process integration. Anytime we are integrating two systems, there are those integration touch points, that is a potential security vulnerability. But of course, it can be taken care by using, appropriate methodologies. Also, it can have a system wide impact in case of one system failure. So there is that. for example, if the order management system fails, all of the other systems that rely on data from this, cannot function, right? So it's like a single point failure. But again, a smart, thought out architecture can address such pitfalls. So now the other type of, integration, it's called the data integration. This is where the data is transformed in a way that systems Then at the data generated in one system can be understood and interpreted by other systems. So the advantage there being that it enhances data quality because the data is refined to be able to be interpreted by other systems. There is also, that improved collaboration between them and. Also, in terms of analytics, it helps, having a kind of data consistency and data quality and overall, it improves a lot of, efficiency in processes. obviously there are challenges with it, but mostly has to do deal with data quality. if the data is not maintained properly, then that lack of quality, rolls from one system into the other. yeah. that's something that can be, avoided. Now, the last kind of, application integration is, it's called the communication, integration. It's called the communication integration. It's a type of integration where a lot of middleware technologies, like APIs, microservices, where the benefits are that some of the solutions are flexible and can be implemented regardless of platform where it is deployed. So it may be on cloud or on premises, right? There is also scalability, right? Not having to replace the whole thing, but just scale those integrations quickly, Also, it helps with asynchronous communication. As in the previous example, if one system fails, it doesn't have to immediately impact other systems, of course, there are challenges with that in that it's more complex. So we need to take care of how technologies are utilized, and it may require some, specialized expertise during implementation and maintenance. So there could be that performance impact, especially handling large volumes. But again, nothing that cannot be handled by making some smart choices in terms of scalability. Now, there are some security risks. Anytime an API is exposed, we need to make sure that has the appropriate security functions built in. So that takes care of any vulnerabilities. Now, okay, now that we learned about what and what is EAI, let's talk about some of the. Best practices. So there are a couple of different, methodologies. I don't want to deep dive into each of those, but just high level, right? There are a couple of different methodologies. So one, it's called the point to point integration. This is when one application can just talk to another application through a one to one integration. This is, of course, not scalable, because over time it builds up to multiple one to one integration. And Then there's a hub and spoke model, right? Where the hub is used to facilitate. data transformation across different applications, so it's much faster, but it's also complex. Then, there is the bus integration methodology where all the participating applications in the bus integrations, they have common standards. So that's the way they communicate. Then there's also middleware, which is some form of application integration. Where there is some logic that can be built within the middleware. So it's abstracted into that middle layer. So businesses choose to replace vendors in the future. they can do so easily. the last one, but I think most widely used right now in the industry is the microservices integration and what microservices are. They use across most enterprises now, and these are lightweight applications that each service single functions, and they are lightweight. And due to the fact it and due to that fact, they can be deployed anywhere, right? So they can be deployed on the cloud or serverless. So this is a very nimble, very agile, way, these microservices are very agile and very nimble that they can be built to achieve, these kind of integrations, between various systems. So these are certain methodologies that can be used, but Regardless of the methodology, the best practice for businesses or organizations implementing E. A. I. Is to have clear objectives in mind. So so they need to make sure they're aligned on what is the desired, business outcome, right? And also go through this kind of comprehensive planning with all stakeholders, right? Whether it's finance, management. Legal, IT, marketing, sales, everyone's right. So now the next thing is, as we are navigating the complexities of new technologies like AI, internet of things, blockchain, now businesses need to arrive at some kind of industry standards so that these are interoperable. So again, Like I was saying, if a business desires to switch out a vendor, they should be able to do so without having to rewrite the whole business logic. So that is something that needs to be taken care of. we already talked about having the data quality and consistency that so that's an everyday job that people need to do. And definitely businesses, they need to implement agile practices. For development or for ongoing optimization, then they're using, E. A. I. Okay, so this is a slide I want to talk about. I came across a couple of different case studies where companies have implemented E. A. I. Sephora. I think most of us would know that it's a retailer of cosmetics and beauty products, but actually they want to be more than that. they want to build that experience. They want to build a global beauty community, delivering those in store experiences, building that connection with the customer. So that's what they decide to do. But, they were in a situation where they had these disparate systems, whether it's managing affiliates in each region, the each country had their own affiliate relationships and each of them were managed separately. And then there was also the data, right? Sephora did not know that a customer making a purchase in the store that had previously browsed a certain product on their app. they did not have any of that view and that's when, Sephora what they did is they implemented EAI, where they brought together all the customer information. So now, they are able to recognize the customer, when a customer walks into the store from, Sephora. they know that with the app, they're able to recognize that customer and they're able to provide those proactive recommendations saying, Hey, perhaps you're interested in these products and show them relevant reviews. And so they're able to do that with by implementing. that really helped them with increasing their conversion rates. Also, talking about affiliates, they were able to consolidate all the affiliates across different regions. And, in managing their relationships in a single place. So that also led to a lot of cost savings and rise in average. Order value from these affiliates. so that was a good, implementation Now, I also know about another company shell, that use this kind of implementation Shell they have different loyalty programs like v rewards go plus Fuel rewards. There are different loyalty programs across different regions, but then, for their pricing analysts to be able to launch an initiative, they had trouble curating the data from, across these loyalty programs and make a decision. So Once they implemented EAI, though, they were able to get to this data. So their pricing analysts were able to decide, for example, this particular gas station, are they getting more customers who are just driving along the road? Or, they're getting customers who live in the area, right? So that level of data was available for the pricing analyst so they can make, appropriate pricing decisions. again, these are some really good success stories where companies have implemented EAI. And again, there is studies that back this up, implementing AI businesses have reported 29 percent increase in open rates for emails and also where this helps is, it's also helps in reducing acquisition costs, right? because companies are no longer, they're no longer sending this mass market emails, they're sending more personalized emails based on, customer interactions. and that is made possible. Due to having this integration of data across applications. So, all Okay. Now, I did want to talk about the implementation roadmap for E. I. it has about four phases planning, modeling, implementation, and then the maintenance phase. Planning phase. It's where the applications are identified. You know what kind of applications need to be integrated? What is the current infrastructure? What is the scope? What do we want to achieve? And also select the appropriate methodology, right? Then in the modeling phase, that's where we decide what is the integration architecture? What business processes do we really want to integrate? And how do we want to map the data so that it is consistent across systems? So that kind of data modeling happens in the modeling phase. And, at the end of that, we have a project plan as well as a blueprint. To be able to do that implementation, then comes the implementation phase where businesses are actually implementing those connectors and adapters, implementing that data transformation, integrating those processes. And then, of course, testing it to make sure the desired business outcome is achieved and then last the maintenance phase. Of course, when it's implemented, it's You know, being able to monitor and manage. And if there are errors or anything being able to fix those, that's part of that. Also, for businesses, we all know that you know that needs have all right. So how do those changes get incorporated back into the system into AI? that is the maintenance phase. And, that's also provide support and troubleshooting for users. Okay, now by implementing E I, there are, of course, tangible benefits for businesses. These are quantifiable real benefits. There is increased customer retention loyalty. There is revenue growth. We talked about it. Customers no longer want this generic. Content or generic offering they want to be able to relate to the business. So businesses are investing in, reaping benefits of personalized offers in a way that when a customer is presented an offer and if they accept or decline it, every action they take. Companies get the data to be able to suggest the next best, action. and all of these are being made possible again through EAI having that kind of access to the data. So there is definitely a lot of cost savings because of the operational efficiency. Nowhere we talked about where one person doesn't have to go into multiple systems, but that the data is available across, in order to make those decisions faster. we talked about having reduced acquisition costs. Yeah. Okay. So now just to reiterate the benefits that offer management can get by using E. A. I. there's personalization, where businesses track customer behavior and preferences. They can boost engagement. So there are companies that are, offering interactive features in their apps and gamification and all of this made possible with E. A. I. Convenience. Of course, customer, if they make an online purchase and earn some rewards, they should be able to redeem those immediately when they walk into a store. So it provides that convenience to customers. And again, for marketers, all of these interactions lead to Valuable data that can, drive insights and further improve the loyalty program, right? and earn that customer, loyalty. Okay, I think that's it. I want to leave everyone with this thought, how, in this competitive market, I strongly believe EAI is a very powerful framework of tools and processes. Of course, for optimizing customer management, providing them the right offers, giving them more personalized offers, deciding what the next best action is. all of these resulting and, driving business growth, and it becomes important that by integrating data and system across enterprise, they can deliver this kind of consistent experience and increase the revenue and earn that, loyalty with that. I think I'm at the end of my presentation again. Thank you everyone for listening in. I really appreciate each one of you. Thank you.
...

Vamsi Srivangipuram

Principal Software Engineer @ DIRECTV

Vamsi Srivangipuram's LinkedIn account



Awesome tech events for

Priority access to all content

Video hallway track

Community chat

Exclusive promotions and giveaways