Introduction
Ryan: Welcome everyone to the Ai Founders podcast show. Our podcast is dedicated to celebrating the remarkable accomplishments of AI innovators, entrepreneurs, and visionary founders and the captivating stories behind the movements that they’ve built. I’m your host, Ryan Davies, and I have the honor of hosting today’s episode on Future Proofing Tech: Is your organization data ready? With our special guest, Ratnadeep Bhattacharjee.Thanks for being here today.
Ratnadeep: Hey Ryan, thanks for having me. I’m really excited. This is one of the few podcasts I’m actually appearing in this week, so fingers crossed, we can make it work both for us.
The Importance of Being Data-Ready
Ryan: This is going to be great for our listeners. This is just a wonderful topic, and I think you’ve got a wealth of knowledge here that we can dive into for our listeners. Ratnadeep is the co-founder of TechVariable, and he’s played a pivotal role in developing premium digital solutions and consultancy services globally. His leadership has been instrumental in executing major projects in data management and AI-driven solutions with a mission that revolves around driving digital transformation for organizations through data-centric AI solutions. And he has a vision execution to set new standards for tech innovation.
Additionally, you’re the host of Leaders Perspective, a digital transformation podcast series that offers insight and conversations for industry leaders and entrepreneurs. It sounds like we cover a lot of the same groups and information and things like that, which is wonderful. For starters, I think to kick things off, you could cover up that introduction a little bit more for us and share a bit about your journey in the tech industry.
Ratnadeep: Yeah, thanks for that, Ryan. You mostly covered everything, but let me try to add something to the palette. So, as you rightly put it, right. I’m one of the co-founders of TechVariable. We are an eight-year-old company. So when we started, we started as a boutique technology company offering software services solutions. Since we have been mostly catering to North American clients for different digital transformation projects, starting from MDP development to actual product development to covering the entire gamut of product development, From designing, developing, and deploying everything over the past 4.5 years, we diverted our attention to mostly projects in the digital Transformation Domain. By digital transformation, I mean data-driven digital transformation. What we realized over the course of that particular journey of these four or five years and what has really been remarkable, I would say, is the number of organizations that are not data-ready. Why did this question even arise? What kept on happening with most of the big organizations and enterprises was that they wanted to implement some AI use cases; they wanted to get into the bandwagon or the buzz called AI. Now, it has evolved into a generative AI Netherlands. What we made a practice within our organization is that before embarking on any data-driven solutions, we should really do a data infrastructure audit. This becomes the crux of how you can go about thinking about a solution, So that is how we would say we evolved from being a services company to a product company. Now, with all the knowledge that we have gathered over the past 34 years, we have put them into some sort of solutions that we have internally, which caters to being A I ready consequently and to be data ready. So that’s a little bit about me. I don’t know whether I was able to capture your attention.
Ryan: No, I think that was perfect. And you talked about it again, Sort of that baseline audit almost, right? And again, for our listeners, we have everybody, and I know with TechVariable, you also have people at all stages of that journey, right? You’re at everything from precede and like I ideation stage right through to mature organizations that are looking for help and to become data ready. But what does that mean? As you said, it starts with an audit to kind of understand where you’re at, but what does it mean for a company to be data-ready? And why is it just so essential for tech founders to make sure that they are regardless of what stage they’re in?
Ratnadeep: I think that’s the fundamental question, Ryan. What happens is most organizations grow so rapidly that they keep on piling up those, how do I put it? Technical lag? I call it a technical lag. You keep on growing so fast that you can kind of forget that technology is probably reaching a stage where it should be. I would call it modernized, if I may say so. Let’s say this organization has a humongous set of data lying within different departments. Let’s probably take an example of a retail organization. They want to do demand forecasting, and they want to build a demand forecasting model. But the problem is if your technology team works on a single data source wherein, you know that as an organization, you have an umpteenth number of data sources lying because each of your team uses different Software to get their work done. So, these are rich sources of data. Now, many organizations honestly don’t use all of their data. They don’t have a very good data infrastructure in place. They don’t have a very good data engineering practice now. What do they do? So, typically, what does an enterprise do? Let’s say they decide that, hey, I think we should establish a data practice within our organization. They would hire a data scientist. Hey, I need this an IML model, or I need analytics, or I need a reporting dashboard to work for me. I need my decision-makers to know exactly what we’re doing and whatnot so that I can report to them the best kind of information we can give them, right? What does the data scientist do? They expect that the data that they get is structured and integrated properly. But when they start looking into the data, they realize that, hey man, this is not something that we expect. I cannot derive any outcome from this particular data set or whatever is available to me. Then, when they look back into what actually exists within the paradigm of this organization, in terms of data, they will realize that there is a huge number of data sources, probably about four or five data sources we can actually use to get the outcome that is not being used. That’s because your data engineering practice has not been set. This necessitates hiring data engineers, data architects, and all those sorts of things. Then you start your data infrastructure journey, which essentially typically takes it’s 10 to 12 months of journey. So that is what I mean to be data ready. If the organizations, I would say 80 to 85% of organizations are not yet data-ready.
Ryan: And that is a staggeringly high number, I guess. But it makes sense because people don’t understand again. Maybe they think they are right. Like, hey, I know my numbers, I know my info, I know my customers or whatever it is, but clearly, there’s a lot of gaps. As you said, the bigger the organization, the more it spreads out, and the more different systems you’re collecting from some you don’t even realize are an area for collecting this data. It’s there to serve a purpose, and that’s not what it is. Well, it’s still there to uncover, right? So I’d love to kind of understand how with this increasing importance of data, I guess, some of the challenges that make it really tough to be true of data ready. But maybe some specific examples that you’ve seen, like highlighting some of these challenges and overcoming them from that sense.
Ratnadeep: I can give you an example of a case study. That will really make sense
Ryan: Absolutely, yeah, that would be amazing. The real-world stuff is the stuff that, again, I think, applies most and we can relate to the most. That’s great
Ratnadeep: Right. So, we actually worked with a multinational FMCG company. It’s a very big company, actually. I won’t be able to name it, but you get the picture, right? It’s one of the largest in the world. So they were struggling with forecasting their inventory level sales and what kind of sales would happen for them to maintain a very good level of inventory, right? So, essentially, they wanted a demand forecasting model. When they came to us, they said that this was the problem. We make our sales on Amazon, Target, and Walmart. But we don’t have a singular demand forecasting tool. We thought that since it’s a very large organization, it has a very good data practice. The data team was big, big enough to have a very good data practice. But when we really deep dived into their infrastructure, as I said, we do it, even if it’s a very, you know, very good data team that they have data practice is very solid. But even then, we do the thing that actually has served us well till now, we do the data audit. We realize that even though the data team is there, they are not particularly interested in working in that particular use case, if I may put it. So what was happening was the target data was lying in silos. Amazon data was lying in silos, and Walmart data was lying in silos. This actually gave us an insight into the fact that they have not a singular data infrastructure set up or a data warehouse set up correctly. Even before embarking on the journey of getting this demand forecasting model done, we actually covered it through our platform, which we call Data Steroid. Essentially, as I said, it brought down the number of months, which would have taken 6 to 9 months for this to have happened. It essentially brought it down to six weeks, and within nine weeks, we were able to deliver even the demand forecasting model to them. It started with really understanding the data sources and then bringing it to a structure that is ready for analytics, which is ready for AI. This is a very good example that I can give you.
Ryan: I love that. I think that’s just great, like somebody who already has a lot of the data, all of the stuff in place, but just couldn’t connect those dots. And then, as you said, OK, we want this, and it’s like before you can do that, you need to take a step back and kind of get that from there. And again, maybe that’s just not their understanding of the maturity of their data readiness from that standpoint. Is there a framework or any set of indicators that any tech founders or business can use to assess that maturity of their data readiness and what that road ahead looks like, in your opinion?
Assessing Data Maturity and Readiness
Ratnadeep: Yeah, that’s a very good question, Ryan. We actually have a tool that we use internally when we get a very good leader. We understand that there is something here. We give them a list of questionnaires, which is kind of a survey but not a survey. It’s an AI readiness test. We call it an AI readiness test because it sounds much more buzzier, if I may call it. It is essentially a data readiness test. We test what stage they’re at, what kind of culture they have within the organization in terms of really adopting data tools, really getting ready for the next stage of digital transformation, which can either be an IML or can also be some analytics or reporting structure those sort of things, we also understand their ETL or ELT landscape, whatever suits the technology infrastructure they have. We also understand what kind of data pipeline they’re working on; those sorts of things basically describe where they are at this particular stage. When we do this data audit, this is one of the first preliminary activities we do, and then we dive deep for a two-day workshop or three-day workshop.
Measuring Impact and ROI in Data Initiatives
Ryan: I guess that’s the key, right? It is to really understand that maturity where you’re at and things like that. Is there a way that organizations then can also measure the impact or ROY of the data initiatives that they’re doing? Because I know that’s a big thing, right? People say I can always measure revenue. I can always measure this or that, like end goals, but how are you able to communicate that impact in terms of ROY and that investment that they’re putting in?
Ratnadeep: What we offer to our leads is that this data infrastructure audit that we do is free of charge, honestly, because we really want to tell them where they are at. Whether they adopt our product or not, we recommend other solutions as well. These are our competing solutions. Why don’t we go with them? This would make more sense. This would not make more sense. This product has this. Our product is this, we give them a very good comparison. In terms of ROY, it’s very difficult to get an ROY calculator for a data life cycle management solution, if I may call it. Essentially, what you’re telling them is, hey, if you invest this much, you will get this much out of this, but how can you measure human productivity? I can give you an ROI in terms of if you have 8 to 9 data engineers sitting for 8 to 9 months. You know what kind of cost I’m talking about, right? This gets reduced to six weeks, and only one person is getting involved. He or she is covering the entire gamut of data life second management without needing to hire that many people for that many number of months, then you know what I’m talking about, what kind of saving I’m talking about. I think that’s the best way anyone can have for any product.
Ryan: I think so. I mean, that kind of speaks to itself, and I think for leaders hearing that, it resonates in terms of, yeah, that is really what I want to do is drive this data-driven culture in the organization because it’s going to lead to so more connect, connect both internally and externally better and make better strategic decisions. How are you finding that tech leaders, can foster that data-driven culture with their organization, and why is that so critical to have that top-down mentality in that sense?
Fostering a Data-Driven Culture
Ratnadeep: I think Ryan, it’s not just the leaders who are responsible. Honestly, there are also people who feel that data-driven culture is only built by the technology people. It’s not tech, even non-tech people. Honestly, these non-tech people are the ones who are mostly involved with generating the data within the organization because these people work on different Software within the organization on a daily basis. So if these guys really understand the importance of putting in as much data as is possible within the different Software without really ignoring some of the fields that exist within the software. Once they start realizing the importance of that, they would actually contribute to this data layer, as I call it, for an organization. Also, they should understand that they should have enough knowledge shared within the organization that their competitors are doing this. Why can’t we also adopt the same things? These people are already using these technologies. These people already have a data infrastructure in place. These people can actually predict many different parameters within their business, important parameters, or kps within their business just because they have adopted AI. Why could they adopt AI? There comes your answer because they have a solid data infrastructure in place.
Ryan: I think it brings it all together from that point, right? We need to make sure that we’ve talked about this numerous times now, but just the incredible importance of having everybody buy into it, understanding the key connections as to why you’re doing it, and what the expectations should be kind of out of it from that. When we’re talking about looking ahead, there are a lot of tech founders. They’re going to face choices in terms of their data infrastructure and technologies, and there are different tools and platforms that are all out there, including yours, of course. Do you have advice on how to select the right ones for you, and most importantly, are there emerging technologies that I think are going to play a really significant role in the future of this space to help organizations move forward? And I’m sure you don’t want to spill the secret sauce hereof of your road map going forward, right? But I’m sure there are a lot of things out there in terms of how people should be looking at making a decision for how to bring somebody into this space.
Ratnadeep: OK. So I would like to answer it this way. I believe you cannot just miss the bus. Generative AI is here to stay, and LLMs are here to stay. Now, we need to truly be an organization that is using LLM in the most effective way. You need to understand your organization’s data right. Now Organization data, and enterprise data, as I said, may lie in silos, or you may have a solid data infrastructure in place. These are the two possibilities. As far as adopting tools is concerned, If you are a very large organization, having a solid data infrastructure is non-negotiable. Without that, you cannot proceed to your next steps. Everything you do will have a repercussion in terms of not having the right or correct KPIS, which you calculate, not having the correct reporting that goes out to your decision makers. Now, when you select a tool, you have to understand whether it is actually covering the entire data life cycle or not because you cannot again make the same mistake of taking one tool for data ingestion, one tool for data engineering, one tool for analytics, one tool for AI, this will again mess up your entire data or technology infrastructure. Let’s put it that way: You are essentially telling people that, hey, you are already using these many software programs, and you can bring them into a common platform and derive insights from them. Now, if you expect people to buy four Software to do the same thing, it becomes a problem. So, I believe it should be a solution, and there are a number of these solutions in the market. It’s not just our solution. There are many other solutions that actually cover the entire gamut. Again, one of the things that I really believe, as I said while I was explaining the same point, is that LLMs and generative AI are here to stay. They are only going to get more and more mature as we keep on exploring more possibilities out of it. I really believe enterprises can leverage it with so many use cases and possibilities. I believe, you know, we are at a stage in the whole technology landscape. Where we really are in the golden period. People say that, yeah, I will take a job generative AI. Will just steal all your jobs and stuff. No, that is not going to happen. I truly believe it is only going to make you more and more effective as an organization.
Selecting Data Infrastructure and Technologies
Ryan: Yeah, I think so too. I mean, there’s a lot of fear around it, fear-mongering almost with a lot of stuff. And I’m sure again there’s lots of ethical considerations and all of that like that we didn’t even touch on necessarily. I think you’re absolutely right that the idea of this is to strengthen and embolden people, empower people more to do more in organizations, to do more from that perspective. I know, obviously a big piece, you know. I guess looking ahead, anticipating maybe these opportunities in the data landscape for tech organizations, preparation is key. Are there elements in place now that you think people should have that are the new table stakes, I guess, in terms of what needs to be there from that sense, where you see, you mentioned if you’re not here, you’re a lagging, you’re behind, you’ve got to adapt to this now. So those things where you’re like, hey, this is the minimum bar at this point. Are there any key indicators that you think are in place? It’s probably a complicated question. It ranges from organization to organization, right?
Ratnadeep: So yes, I believe Ryan in two things. First, do you know your data sources? That’s the first question. Do you really know your data sources? You should not be in a place as an organization where you are not sure where your data lies, or even if you know where your data lies, you don’t know how to use it. I know so many big organizations that still have legacy applications running. How do you make sure that the gold mine of data that is there in your legacy application, which your people are anyway using, we’ll keep on using? How do we make sure that without really needing to modernize it? How do we make sure that we keep on using those legacy applications?
But even then, you know, modernize the other things in the business. So there are ways of doing it. There are ways technology can help you achieve that. But at least people should be at this stage to really identify the data sources. Secondly, and one of the most important things is to know what kind of use cases they have as an organization, let’s say, two years down the line, I want to be at a stage where I can forecast my demand where I can have as a bank. I want to be able to do hyper-personalized marketing. But you should have a vision for that. You should know what kind of use case you would like to work in, and only then can you go backward and backward and then achieve it, one step at a time.
Ryan: I love that, and I think you’ve kind of talked about it a little bit more here, and as we close off, you said AI and Gen AI are kind of buzzwords these days. Even as you said, you’ve named your tool around that to try and create that second layer of interest and kind of capitalize on that. I guess how AI and Gen AI specifically, I mean, maybe we could talk about for you and TechVariable, are really creating that a revolution around digital transformation, I guess, and being able to empower it even further.
Ratnadeep: We have a very good tool called word-wise.Essentially, it’s an LLM-backed chat application that works on your enterprise data, which helps you derive insights from it. So let’s say you are a head of sales, and you just type in, hey, I want to know the top five most-selling products in my organization for the past three months. Then, the chat response will be here. These were the products: product one got this sale, product two got this sale, and then you also get a graph out of it. You can add this particular graph to your PPT and then share it with your decision-maker. It saves you a lot of time. You don’t have to go to Tableau or any such visualization tool to get the graph or generate insights. You can question it in a human understandable format and the response will also be in a human readable. It’s essentially Chatgpt on steroids, if I may say so, but on your organization’s data. Then, secondly, security. The people are concerned about sharing their organization data with Chatgpt because it gets shared across, you know, they can use that data, right? What our platform does is the LLM technology that we use does not even get to know your data and what kind of data it has. It can generate insights without having access to your data. So that’s the inbuilt source that we have internally, which which enables us to do it. Also, all our platforms, even our data engineering automation platform, which we call data steroid ironically is both Data Steroid and WordWise, are on our on-prem deployments. It’s not a SAAS application. We customize it according to what the client needs and how the client needs it. And we deploy it either on-premise or on their private clouds. So that is how these things need to be balanced. You need to be very aware of data. Security, and data governance is ensured through a platform called WordWise, data democratization. These are the three important pillars of data life second management, I believe, And I think we are trying to cover as much as we can.
Ryan: I love that, and it’s, I mean, so important for safeguarding this, this data. There’s sensitive and nonsensitive, but it’s all sensitive. Like you said, it’s all proprietary. It’s all stuff you must be so careful about. Another point to consider all of this data readiness with that, you know, red and deep, what I’d love to do is just kind of turn it over to you here at the end he. And how can people get connected with you and take advantage of you? You mentioned some of the great things that tech variable offers. I know there’s going to be people that are going to be reaching out asking about it. So, we may as well cover it here to get you connected with some of our listeners as well.
AI Adoption Dynamics
Ratnadeep: Sure, The best way to reach out to me is by email and LinkedIn. I’m reachable at ratnadeep@techvariable.com, and on LinkedIn, you can search for Ratnadeep Bhattacharjee. I think you’ll get it. There are very few Ratnadeep Bhattacharjee on LinkedIn, I believe Ryan probably will be gracious enough to add that during the release of this podcast as well.
Ryan: Perfect. Absolutely. We will take advantage of that. Ratnadeep and his team, just an unbelievable wealth of knowledge, have given, you know, free tools and things that we can use here to help with that audit, help with that understanding of answering that question of is your organization data ready? I don’t know. Is it? Am I? How do I know? Where am I going to get that answer from with someone you can trust? I think that’s absolutely perfect and a great place for us to leave off. I want to say thank you, Ratnadeep, for being here today and for sharing your journey and a little bit more about data readiness and how organizations are preparing for this for the future, AI is playing such a crucial role in this as well. I think we could probably have you back for a dozen more topics based on the things we talked about here. So I really appreciate it. Thank you so much for being here.
Ratnadeep: Oh, the pleasure was mine.Ryan. Thanks for having me, and looking forward to hearing from you soon.
Ryan: Excellent, and thank you to everybody for joining us on this enlightening journey through AI innovation. And we hope you’ve been inspired by the incredible stories and everything shared today. And remember, the future is driven by pioneers like our guest, Ratnadeep Bhattacharjee, and the limitless possibilities of AI, so stay curious, stay innovative, and keep exploring the boundless horizons of technology. And before we sign off, I will have a small request for our dedicated listeners. If you enjoy our podcast, please take a moment. Leave a review, subscribe on your favorite platform, and share it with those that those of you know. It’s your feedback and support. That helps us bring more amazing content and incredible guests like Ratnadeep today. Thank you very much again. And until next time, this is Ryan Davies, signing off, take care everybody.