Hi everyone. Good morning, good evening, and good afternoon. My name is Raach Ma. I have around 20 years of IT experience providing solutions to various business domains like healthcare, manufacturing, retail. All these years like I have contributed to legacy application modernization, digital transformation. I would say that's my core area of expertise. So if. Already working on legacy platforms and you are in the process of modernizing it. Then you are in the right place. So in the next couple of minutes we will talk about how a powered solutions would help us in transforming these legacy systems. That's about my quick introduction. If you have any questions after the presentation, you can please feel free to reach out to Arch in the neuro@gmail.com. Yep. Let's write a bit jump into the presentation. All right. So like I mentioned many of us in our career would've worked with many legacy applications and systems, and we all know so much business value would've been tied so tightly. With these legacy applications. But what happens when the business expects these legacy applications to be meeting today's business needs? One of us definitely would start thinking about how I can modernize these legacy applications, right? And definitely we tend to follow the proven and the useful conventional approaches, but sometimes these conventional or the traditional approaches might not. Might not be good enough to meet the business expectations or might not give you the desired in income desired result. So then you will start looking for the alternatives. So this is where I talk about, a powered solution, which is which I call it as a cognitive retail modernization framework which will help me like how exactly I can transform these legacy systems at the same time I can enhance this customer engagement mechanisms. So what would be the the approach that I would want to come with through this presentation is like it would be an embedding of cognitive capabilities with this, legacy systems where where you will produce a intelligent hybrid environment that is going to combine with proven technology with the advanced AI capabilities. Okay, so before we, we go down further into the details, I want to call out what are the present challenges the industry may face especially in the retail sector? So I, based on the research, what I could read from various articles or my own survey. And it's 70% of these legacy systems like they are like more than 20 years ago that they have the age as 20 years. Some of these businesses are still running on these legacy applications. And I would also see that 70 to 80% of the IT budget is being contributed towards its maintenance. Having not having much preference to the innovation, which is not at all a good good thing, good sign, I would say. And 22, I would say like just 20 to 25%. Only 20 to 25% of the data is being utilized for decision making. I would say today data plays a so vital role that you have to use as a main key driver for any of your decision making. So I would say these are the, some of the challenges definitely we should be addressing keeping. Keeping in mind that conventional approaches may or may not address your current business need. So I have to always call out for options and alternatives. Yeah. We, for this challenge, what other options I have to address and then meet today's business expectations. Okay. So like I mentioned, we would definitely try to follow some of the traditional approaches the conventional way of modernizing these legacy systems. But before we go there, I want to call out like what limitations that you may face when you try to deploy these approaches. The first one would be high risk and uncertains definitely. When you use these traditional approaches, it is possible that, you may carry heavy risk. Because these applications are mission critical applications. Some of them might be, and you'll be having so much business logic inside that. And when you use this traditional approach, you may not really take the business advantage of that, and that would definitely be a high risk. And you might not be able to see the desired outcome of it. And the second thing I would say it'll be a migration without transformation, what I meant to say. You are able to migrate it to the new platform, but you'll still be treating that as a. A modern legacy system, meaning you are not able to really achieve what exactly you, you plan to achieve because though you follow a traditional modernization approach using a lift and shift or rip and replace you might be able to. Migrate it, but you wouldn't be able to completely transform the existing system, and you wouldn't be in a position to achieve the desired outcome. And the third one is a gradual improvement without vision. So though the traditional approach still suggests that you can apply the phased manner, but still you wouldn't really be able to achieve the the vision that you have planned for it. What I meant to say, we are trying to do it in incremental fashion, but still somewhere in at some point of time, you'll feel that it still lacks some amount of integration to give you the holistic picture of it. So these are the limitations of the traditional modernization approaches. Now, having said that, these are my limitations. What options do I have? Let me see. All right. So like I was talking about though, one option is to use this conventional approach of modernizing these legacy systems. What if I try to use this in the, in this era of AI and ml AI powered, cognitive retail modernization framework and how exactly is going to help me out. So what is this framework all about? So what are the various phases or the layers it has? Let's see here. So it is primarily would be comprised of for four different layers. Cognitive layer integration. Predictive analytics implementation, hyper-personalization engine in diligent process automation. Let us see each of these layers in a detailed manner. So what I meant by cognitive layer integration. Though you have so much data residing on these systems, you should be effective position to utilize this data. For a better decision making. So you should definitely deploy some intelligent methods or techniques for processing these data. Like you can use APIs or even streams or various approaches so you can use them and you would, you should be in a position to effectively utilize unprocessed these data streams. And predictive analytics implementation. What I'm trying to say here so while you started the processing of data, you should also think about applying various machine learning approaches through which you can process and produce a results that can drive you in a better way for, predictive decision making which is definitely, it is going to be an optimized way for forecasting these patterns. And then decision making. So that would be the second layer. And the third one is hyper-personalization engine. So you have processed the data and you have applied some machine learning models or the techniques on this process data. What would be the outcome of these two approaches? Like you tend to arrive at a stage where you can tailor, you can use this process data and you can produce certain tailored approaches, how you can reach out to the customers and you can engage them, you can convert them you can make revenue out of them, right? That is hyper personalization engine. And the third one is intelligent process automation. So now we consider data, we process it, and we have, we started working on this process data. We use this process data for making better decisions tailored approaches and process automation. So you should see wherever there's a refinement, there is a fine tune fine tuning required for this process. And based on this processed data you will have a better control how intelligently you can automate some of these processes. You can apply these cognitive capabilities, but not, re-engineering the whole system. Yeah. So why I'm, in one thing I would say why I am multiple times insisting on this. Sometimes we go with this rip and replace as a whole cell approach. That might not be really helping you sometimes. So that's when you have to think about what if I use these AI powered solutions? On these retail systems and when you try to impose this framework what it demonstrates actually what kind of potential it'll give you in in, in the cases of increasing the customer engagement rates. And I would say based on this survey, I would say like 40%. I would say there is a, there is an amount of increase in this customer engagement and also in the overall operation efficiency. When you use this cognitive retail approach. Okay, now we have seen what are the layers various layers involved in this methodology in this framework. Now we will see what are the various phases, how exactly we can implement this cognitive retail framework. Phase one is like cognitive foundation. Phase two, intelligent augmentation and phase three is autonomous optimization. So what exactly meant by cognitive foundation. Now like I mentioned in the previous slide, you have so much data, critical data, siding on these platforms. You have to build efficient data pipelines and even streams so that you can definitely build a core infrastructure that, that, I would say that's the basic. The foundation phase where you will you will establish this co cognitive retail framework. And phase three is very intelligent augmentation. So now you put the foundation, now you have to integrate these prediction capabilities into these operation systems so that you can augment the capacity or the the efficiency of these intelligent systems. And the third phase is autonomous optimization. As the name series you, by implementing phase one and phase two, you are always welcoming. These systems like to be self capable, self-managed or self optimize these algorithms and these closed loop systems. These three phases would really help. Anyone to implement this cognitive retail modernization framework. That way we will see like a tremendous difference between the traditional approach and the the a powered approach. And I would see definitely I have to mention this, by using this approach, there is a success rate near 2.5 times higher for this approach where you'll still be falling that incremental and the phased manner. Approach. Okay. Yeah. I, to make it in a better, under better understanding, I would wanna take a case study. It is about a regional grocery chain, which which like, which, just like any other grocery chain would be handling things like out of stock issues over inventory, overstock. Like in terms of how effective we do the promotion strategies, and once I implement this, what are the responses how it is really affecting the customer's shopping trends and how I'm really, retaining those customers. So I want to take this case today. I want to talk about what if I implement this AI powered approach, how exactly is going to bring the benefits, sir? So I want to call out certain numbers. This particular grocery change, which I'm talking about, it is like having an old application, a legacy application, which is of 15 years old merchandising management system. Now rather than doing this conventional approach of rip and replace of the whole wholesale approach what if I try to deploy this cognitive retail modernization approach, how exactly it is going to bring the differences or the results. So I would say based on the case study, it has tremendously, it has showed the substantial business outcomes with especially the positive return on investment within the eight months. I would say the total implementation cost was like approximately $3.2 million, representing around 0.4% of retail's annual revenue. And it is compared to the estimated 12, $15 million, had it been like you have used this conventional approach. Now I have said that I have used this AI powered approach instead of this conventional approach, how exactly it is going to impact the financial factors of this grocery chain, first year financial be benefits. So it has proved that it is going to, or it has already given approximately $5.8 million financial benefits with 181% written on investment. That is the first obvious financial impact. And the second thing, inventory management. So approximately $3.5 million you can save or you can see the savings through reducing the carrying costs decreas shrinkage and improved sales for enhanced product availability and marketing effectiveness. So under this category, it is approximately like $1.2 million. By enhancing this marketing strategies promotional efficiency, all, all because of introducing these cognitive capabilities. Instead of using this traditional approach of modernizing this legacy systems and with respect to operational efficiency, this approximately $1.1 million savings. Especially improving the delivered financial benefits through reduced labor requirements for administrative tasks and improved workforce ization. So not only these financial results if I want to compare with this traditional approach, it is, I would say it's a true continuous learning and adaptation. This cognitive framework has introduced an accumulated data, analytical experiences, along with its predictive accuracy and business impact increased over time. Okay. So that, that, I would say the general not general, the specific financial category. What are the general benefits if I use this cognitive retail moderation approach? One is reduced implementation risk. If you rightly remember I have called out this as one of the risks when you use this traditional approach saying so much these are mission critical systems, and so much business value or business rules would've been implemented in the system. It is so possible that you might not carry it over when you use this traditional approach, but considering this AI powered cognitive retail modernization approach, you would, it, you would like reduce this implementation risk and foster time to value. Of course. So it is so possible when you use this conventional approach. It'll take some time. It'll take large amount of time, cons compared to these augmented approach. So that way when you use these AI powered frameworks, it is so possible that you can start seeing the results. You can bring in the change in a much faster way. And you can start responding to the market demands and needs and lower investment requirements. When you use this traditional moderation approach, it is possible that there is some amount of skill gap. There is some amount of efficiency problems. There is a possibility that it'll we tend to see some unexpected surprises. If I use this AI powered modernization framework it is definitely you have a great chances of lowering your investment or the capital investment on these requirements and then preserved institutional knowledge. Like I kept saying in all these legacy systems you have so much business logic rules have been encapsulated. You. At any cost. I don't think business is at the liberty of losing them. So by using this augmentation methodology, I'm not going to move away from it completely, but still, I'm going to modernize these systems res, be able to respond to the business current business needs. So what I'm trying to say, you are, I'm still able to retain the business knowledge what has been there in those old systems. At the same time I'm able to respond to the new business needs. Having said the bene the benefits and the financial impact, what sort of challenges we may face when you start implementing this retail modernization approach? So data quality issues, like I said, so data plays a very critical role. If the data is not being properly maintained, not refined enough, so whatever the data you are going to supply for this moderation approach, it is possible that it may not give you the expected outcome because the data is not in a very refined format. It is possible that it'll give you undesired outcomes and integration complexity. You have to understand I'm trying to modernize these legacy systems. During this process. It is possible that I may have to face certain integration challenges and complexities when I'm trying to bridge this gap with these a platforms. So you have to be careful enough and you have to consider ahead what sort of those challenges. That way it, it'll be easy for us in the roadmap. To make the process much smoother and skillset evaluation. So this AI powered approach, though, we start seeing the results of it, I would still say it is it is like a quite a rapidly evolving technology. So you have to be in a position to choose the right set of skills and talent to be able to apply their strengths and weaknesses while I work on this moderation approach on the change management. It's so you are doing, you're doing a big leap from this legacy modernization legacy systems, modernizing them to, a cognitive frameworks. It is going to definitely be operational challenges. You have to bring in certain change management procedures for effectively handling all these operational procedures. Okay, technical debt. So not only these these four factors. The fifth factor is what it is going to talk about having, we have already mentioned about this increased integration complexities. The, these approaches are going to accumulate this technical debt. When it is implemented without appropriate architecture, vision, and discipline. So these are the five challenges that we may face when you, when we try to implement this framework modernizing or enhancing these legacy systems. So having said, those are my challenges that we may face. What exactly the considerations that you have to keep in mind during its implementation. You have to establish a clear architecture standards. Absolutely. I agree with that. So you should always have a a standard way to design a coherent architectural framework that ensures shortterm enhancements and supporting the long-term system health and implement data governance. Who will not agree with me. Like when you are trying to deal with such critical data, which is going to be playing a key role for your business expansion, you have to have some governance in place. You have to develop some master data, go master management capabilities, and you have to have certain frameworks in place for maintaining the data quality and develop talent strategy. Like I said, it's a quite evolving and rapidly changing technology platform. So you have to have a strategy like how exactly you can make use of the present trends and techniques, and you can assess the talent and the skills of the people. And you can, put them in the right way. So you should also see do you have an internet expertise or you have to invest so heavily so that you have to hire some new people while you want to achieve this. No the fourth one is prioritize change management. So in the previous thing, we have considered this also one of the, one of the challenges we may face. So you have to keep this as one of the implementation consideration that you have to have a comprehensive change management process for effective utilization of these new capabilities. So I would want to conclude, it is all about like how we effectively you balance like both legacy systems, knowledge that is still playing a pivotal role at the same time. The new features, the new enhancement that you can leverage from this AI powered approach. I would want to call out the three various areas where you can balance it in a very you can do it in a very balanced way, enhance the customer experience. So these cognitive capabilities. I'm pretty sure it will deliver personalized interactions, that is gonna increase engagement rates by up to 40% while preserving the human touch that defines exceptional retail experiences. So that is one, one factor. The second one is preserved technology investment. I would've already spent so many so much money on retaining this making so, so stable. My platform so much operational experience and this refined algorithms that I would've been having on my systems. You should you should go in a very amicable way to protect these valuable institutional knowledge at the same time you leveraging these advanced, the cognitive capabilities. And compelling financial deals. This would be the third area where I would say you, you would be able to still deliver substantial operational improvements and compelling financial returns with significantly compressed timeframes. Yep. What I would want to finally say is you have a legacy system. With so much institutional allergy knowledge residing on this platform, you have to assess in a very thorough manner, should I be placing what sort of approach or strategy that I can use. Should I be using a whole wholesale approach of lift and shift or rip and leap replace, these approaches are proven ones, but still are in the a in the age of AI and ml, you should really take one step back and then see what if I try to deploy these AI powered approaches like what sorts of capabilities that it is going to introduce? Is it going to just modernize or is it going to enhance further? So you should always have such sort of decision making. And then see when I introduce this kind of new frameworks. Am I going to minimize the risk? Am I going to maximize the value? If that is the way, yes you, you have all the possibilities of effectively apply these trends in this digital age. At the same time not losing this institutional knowledge, which has been proven for so many years and helpful, being helpful for your businesses. Yeah I'm done with my presentation. Thank you so much. If you have any questions, I would be.