Introduction: You are listening to Samsara. Be inspired by NOCODE Superheroes.
Moderator: Hello, and welcome Samsara again. I'm your host, Joyeeta, and we have today with us Suraj Rajan. Suraj is a solid data practitioner who has intense experience of building and scaling data related systems for many businesses. Suraj, welcome to Samsara.
Suraj Rajan: Thanks, Joyeeta. Hello everyone. Before I start, I really hope everyone listening to this podcast is safe and staying indoors.
Moderator: Yes. People, wherever you are, you should be listening to this either on a run or at home. So, you shouldn't be anywhere else. Great. Suraj, tell our listeners a little bit about who you are and what you do.
Suraj Rajan: Yes. I call myself a data technologist. I specialise in data and data analytics. I'm a strong believer in data, and I believe in data driven insights which is accessible to everyone. The key reason for that, is to convert the insights into actions, thereby not just driving the enterprise value but making a difference to one's life. So, data analysis has always played an important role for decision making, and that's the key in terms of understanding what the insights are or to translate those insights to actions. I think the start point is always data analysis. I've had about fifteen years of experience working in data, and I've been through the data journey building data warehousing, business intelligence (ph 01.27), data analytics and data science.
Moderator: Wow. A quick obvious question to me is, how did you get this entire spectrum of skills, because obviously you can single-handedly manage and run a team, recruit, get the solution implemented, understand the business? What were the various steps in your career that led you to becoming a person who has this 360 degree perspective? I'm asking for the benefit of many of our listeners who are thinking, how do they become like you?
Suraj Rajan: I think once you start working with data, you really understand how fascinating it is to work with data. We talk about data, and when we talk about data, we talk about analytics, data trends (ph 02.06) and so on, the whole spectrum of what data offers. I think data is all about finding solutions to real life problems. It expects you to be curious, creative, and challenge all the theories around you, and it's actually a journey. When you talk about data, you talk about the journey towards discovery insights and innovation, and that's what I find really fascinating. When I started working in data (ph 02.26), as a data analyst or a developer, I was really keen to understand what the bigger picture is. What is it we're trying to solve? Yes, we talk about big data, we talk about data democratisation now, but that's the, sort of, problem which we always try to solve, even, probably, say ten, fifteen years ago. But maybe not as, at the enterprise level, but at least, probably, at individual function level. That's what I found very fascinating about data, and that's how I, sort of, expanded into understanding more about data. The key is to understand any corporate problem or an organisational problem, you need to understand your data, and have the ability to translate data into information.
Moderator: So, you bought up a very nice point, translate, which also came up in a couple of our other podcasts where people said that when you have a business problem, or a business challenge, or potential top line opportunity, the ability to translate that into a data driven solution is a big, big part of what will happen next. I think there's a bit of a grey area here, Suraj. Who shall do this translation first? Do you think that, normally, what we expect or what happens, is it the business user who does the translation? Are these the data scientists who sit in between? Do you get consultants? How does one do this translation? Take this statement and turn it into a data driven problem statement?
Suraj Rajan: I think in the last say, ten, fifteen years, the whole function of data has changed drastically. There's been a complete shift in the culture, when it comes to working with data. For example, say ten years ago, we didn't necessarily have a chief data officer, right? It was (toeo 04.08) but I think organisations lately have realised the importance of data to the company and that's where you had CDO, and the whole reason of having a CDO is that you have a strong stakeholder sponsorship and support right from the go. So, in doing so, what a lot of companies have done, is they've changed the whole culture of data. Before, we had business analysts or the business owners of the SME's, the subject matter expert, would probably understand data in a bit more depth back then, but now, companies talk about democratising data, turning your data into information, say you have a data platform which makes data available to all the users, and when I say all the users, it's not just the business users, but anyone. It could be a developer, it could be a CXO (ph 04.54) or it could be an analyst. So, now we'll talk about, say, first class data citizens. Now, I think the whole transformation has gone through in various companies, where any individual who understands data, the key to this is, someone who understands data, because data in isolation is meaningless.
It's only when you understand the context of data, you really understand what the data has to offer and that's (inaudible 05.20) a bit about data journalism, just like, you know, if you want to be a good journalist, you need to understand the five W's and H, in terms of where, who, what, why and how. I think it's really important to understand the relevance of data when you're working with data. These days, anyone in the company can be a first class data citizen. If you have that, you should have the underlying infrastructure that the company has, the underlying culture that the company provides, you should have a good guardianship in terms of data management. So, that includes the data quality-,
Moderator: So, you mean not just lip service, it's, like, really, the company top-down being incentivised to be data driven, not just check boxes, some innovation initiative and, you know, using that as a PR propaganda?
Suraj Rajan: Absolutely. Definitely not. Say, for example, you've been working with a lot of customers, right, so when you talk to any customer and ask them, saying, when they're making decisions, do they always consult data? Do they have access to the data they need? And how trustworthy is the data that they have to actually make the decisions? The last key question is, how do they trust their team to accurately interpret the data? I think those numbers are probably very surprising because a lot of companies still don't have that maturity when it comes to understanding data, interpreting data or using data to make data related decisions.
Moderator: Yes, a very, very good thing. I think you bought up some really good points. One is, you spoke about a first class data citizen and we'll explore that in the five whys, and also, you mentioned about the infrastructure being right. I'd like to know a bit more about this citizen, first class data citizen. Who is this? What are you talking about, for the benefit of our listeners? Who's the first class data citizen?
Suraj Rajan: When we talk about first class data citizen, I think, let's take a step back and understand how we can actually facilitate that. In the last five years, a lot of companies have, what's called, data democratisation, as one of the key objectives. So, for the benefit of users listening to it, data democratisation is actually making data available, have a robust platform, where anyone with, say, even with less technical experience, have the ability to understand data. Basically, there is no gatekeeper or any bottleneck when it comes to accessing data, and that's what most of the companies are trying to achieve. So, what that means is, once you have that in place, then you can actually use data for day-to-day decision making or understanding at a much faster pace than your competitors. So, (inaudible 07.51) competitive advantage. Now, how do you go about doing that? The problem over the years is, people have been working with data, but they've been working with data in isolation, say, from a silo system. What that's happened is, it's, sort of, created an SME function especially with those silo applications, but once you democratise data and once you have a more data driven culture within organisations where any problem statement, I take a step back and just look at the data, and then make informed decisions using the data. It, sort of, converts me into being a data citizen, or a first class citizen, where I'm actually using data for day-to-day decision making.
That's always the first step in terms of having an appropriate infrastructure or a culture. Who can be a data scientist? Anyone can be a data scientist, right? That's the aspiration of most companies. When it comes to first class data citizens or data scientists, the prerequisite to that is having a robust platform, a data platform, which enables an individual to access the data and make decisions, have the ability to interpret data.
Moderator: So, you're basically saying a first class data citizen is someone, who regardless of their business outcome, is empowered by data?
Suraj Rajan: Yes.
Moderator: And is actually, actively using data, data related workflows, it's embedded within their work and they're using it to manifold accelerate the real job, right?
Suraj Rajan: Absolutely.
Moderator: Another point you mentioned about the five W's. Can you elaborate a bit on that? What are those?
Suraj Rajan: Yes. So, there is a standard thing about journalism. If you want to be a really good journalist, the key questions that you really ask is who, what, where, why, when and how. I think that's really applicable even when it comes to data journalism, as I would call it, because data on its own, like I said, is meaningless, but to actually be able to work with data, you need to understand the relevance of data, you need to understand, if something happened, who was responsible for it, why did that thing happen, where did that happen, (TC 10.00) and what's the date and time that something happened, and how did that happen? So, once you answer this question, it gives you a bit more context to the data, and that helps you translate that data into information.
Moderator: So, you're saying the context and the, you know, setting of the data-,
Suraj Rajan: The relevance.
Moderator: The relevance, the organic production of it and the use of it, is equally important. Data by itself is not information unless all of these things are known, right?
Suraj Rajan: Absolutely, yes.
Moderator: So, here is then my questions. You asked me, as we move towards a world where, you know, everyone wants to be more and more data driven, at least those who are not providing lip sync, they're making an effort. So, when they do that, there is either over or under reliance on data, either they completely ignore the importance of it and they're losing out, as we see in several sectors, you know, what Amazon is doing to several retail-, we can see that happening a lot in a couple of traditional sectors, but then, at the same time, there's an over reliance. I've seen this data, so now this is going to make sense. Without this metadata and the context that you just mentioned, we're falling short of making really good decisions. So, what, holistically, from a system wide perspective, as a person who's actually been inside and outside the system of creating data related changes, going into organisations, building their system, solving it for them, what do you think could be an all around approach to solving this? How do we make sure we're not going crazy about data, but we're also not ignoring it? What is the balanced approach and what are the various things we can do inside an organisation so we have that?
Suraj Rajan: When people talk about collecting data, I think a lot of people fail to understand the data provenance because it's really important to understand who actually collected the data and why. Like you said, we have a plethora of data when it comes to any (mw 11.55) because, I think every day work (ph 11.57), the world actually produces about 2.5 quintillion bytes of data, so when you have access to so much data, how do you make sure that the data you capture is meaningful? So, I mentioned the word data provenance, so what that really means is, having the ability to differentiate what's primary data against inherited data, what's the difference between captured data against exhaust (ph 12.22) data, and what's the difference between raw data and process data? Because if you have processed data, which means there's a lot of inference and assumptions being made already as part of the transformation, so how would that enable you for any, sort of, further insights or analysis? I think when companies are building any sort of data platform, they really need to understand these elements of data, and have a clear differentiation in terms of what that means for the data management. Once you've done that, once you have a robust data strategy, then I think it, sort of, simplifies the whole data analysis aspect of it.
Moderator: These are all super valuable inputs, so I'm going to go a step further, should I say a step backwards? You obviously know what you're talking about and you've done this a really long time, you know the pitfalls and you know it inside out, it's very clear, Suraj, you have high clarity of the whole process. But someone who's just looking at these processes for the first time, or taking a first step in this journey, an organisation that has just had it's first CDO, or has just begun to look at being data driven, what would you advise them to be the first few steps to be done, in order to embark on this journey successfully?
Suraj Rajan: I think first and foremost, you should have a clear vision in terms of what you're planning to do, or plan to achieve, because without a strong vision, it really makes it difficult to understand what you're after. When I talk about vision, it's also about understanding, not just depending (ph 13.48) on the results or outcome, but you should be really crystal clear in terms of what is the problem that you're trying to solve? Without a clear problem definition, any, sort of, solution or product that you try to build may not really be applicable, or make it irrelevant. So, it's more about understanding the problem, have a clear strategy, in terms of what the vision is. So, say when companies talk about, 'Okay, we're going to democratise data', what does it mean? You should really have a robust statement saying, 'This is what we want to do, because by doing so, these are the benefits that the company can benefit from', and then you drill down into other various details.
Moderator: Yes. So, you're saying something that I've heard from quite a few people who have come on this podcast so far, especially those from the data background such as yourself, data scientist, data analyst, data producers, data journalists, CDO's, they're saying that the question that you ask is, basically, your solution, your answer, your data system is only going to be as good as the question that you ask. But it's a chicken and egg problem, sometimes when you're starting out with this whole deluge of data, we don't know what are the questions we should ask, you know, we're in an exploratory stage, we're trying to discover, are we sitting on a gold mine, or this is a wild goose chase? What advice would you give to narrow down those questions? How do we begin to even ask those questions? I'd say, you know, like you have thermodynamics, the rule zero before you even go to one, two, three, whatever. What is step zero to ask the right questions?
Suraj Rajan: I think, like I said when I started the podcast, I said any experience with data is actually a journey towards discovery insights and innovation. So, once you had a strong statement in terms of what you want to do, so, let's take a statement saying, 'I want to make all the data in the organisation available to all the users', so build a data platform which is robust, you've been through the data qualities, you understood the whole data management, the guardianship of data, and you had a culture where you've established an organisation model, you have ways of working, but then, the beauty of working with data is, as and when you work with data, once you've got data in one platform, it helps you unravel mysteries. So, playing with data actually helps you come up with unexpected answers to a lot of problems. (inaudible 16.06) has always been craving for optimisation techniques algorithm, right, because you need to use your brain at every step on a given, sort of, data, and you come up with ideas that can then deal with data in a better and efficient way. So, you don't have to have all the answers on day one, but the key is to understand the whole business proposition. How do I make sure that what I'm doing benefits the business, or, say, it solves peoples problems, or solves whatever problems? Once I have that statement and once you start working on it, then you discover a lot of things through the journey, and that's where having access to the data then makes it relevant to build on those insights and then automate (ph 16.45) those decisions.
Moderator: So, a close interface with the business users, and empowering them to some extent to enable an access to data is also valuable, right?
Suraj Rajan: It is going to be valuable. Exactly.
Moderator: I'm trying to think of value in that respect. We're trying to enable the business user who's probably not going to be data scientists full fledged or a data technologist like yourself, but to some extent, if they can take a few steps and cross a little bit of these bridges beforehand, then you can, kind of, meet in the middle.
Suraj Rajan: Absolutely. If you look at what's happened in, say, IT sector in the last ten, fifteen years, before, I think most of the data was actually owned by technologists, but now it's a business who owns the data. Like you said now, business understands the importance of data. So, data is something that's actually bridged the business and technology, say, in the last five, six years. The whole idea why we, sort of, democratise data, for example, even what (mw 17.38) is trying to do, is to (mw 17.39) empower all the business users because they understand what they want to do, just that they don't have the data to do what they want to do. It's those people, you know, that we are trying to enable and make sure that there is complete autonomy and there is complete empowerment when it comes to building solutions using the right product.
Moderator: Right. So, you're basically saying that as much as we want to have more and more data science and data knowledge in the world, which is one side of the world, there's, like, equal importance for the business users to get empowered, to understand, and there should be empathy on both sides to meet in the middle, because the main goal is to solve the problem, right, that's the main goal?
Suraj Rajan: Absolutely. Yes.
Moderator: Suraj, what do you do when you're not solving this terrifically difficult problem? What else does this person who knows so much about data and has spent an outstanding amount of time in life building value for huge organisations does when he finds time?
Suraj Rajan: Well, to be honest, I don't have too many hobbies, but I love travelling and I like to stay informed on technology, basically, what's happening in the fintech world. I'm really curious about the whole finance business side. I did a lot of work on blockchain, so I'm really curious to know what's happening in the blockchain world. And of course, I've been supporting a few start up companies, especially having been associated to UKBA, which is UK Business Angels Association, and that's where we met, if you remember, a few years ago. And also being affiliated with companies like Syndicate (inaudible 19.11). Basically, what I'm after is, because there's just too many things around us, I really want to compartmentalise and stay focussed on pretty much most of the things where possible, but having said that, we have a gorgeous furry baby, Lincoln. She's a Havanese breed who basically keeps us busy. And, of course, you learn from, not just people around you, but also animals around you, like, Lincoln is very curious and she has a (mw 19.37) FOMO which is fear of missing out. I think that really adds to your personality as well.
Moderator: Yes, definitely. I'm a bit curious about your career and how you've come here. I know that you've reached a point where you can walk in and create a data solution, actually implement it and bring value to traditional (TC 20.00) businesses like the national grocery chain and things like that. You actually understand the business problem, you can help them solve it, you know how to deal with the stakeholders and you can create the solutions. But these require different types of skill sets, right? You're obviously a person who has to manage people, you're a person who also has to understand the tech, and you're a person who also has to deal with the business stakeholders. So, a variety of soft and hard skills come into play. How do you do that? How did you train yourself for it? What are the variety of different roles? What type of education did you undertake? I'm just trying to, you know, help out some of our listeners who are probably thinking how they can get into your role twenty years from now. How would they plan out, starting out from here, like, if they are right at the beginning of their career? How are they going to plan out their next few steps so they have a well-rounded profile to do what you do?
Suraj Rajan: I think, well, before you start anything to do, you really need to be passionate about, what is it you want to do? For example, if I was in a completely different and alternate universe, I probably would have been a pilot or a scientist. I was really keen to understand how time travel works, so I was really, I was in a mood for (ph 21.12) (toeo 21.12).
Moderator: Time travel, where would you go, Suraj? Future or past, where would you go?
Suraj Rajan: It's always the future. If you want to advance through the years a little bit faster than the next person, then you need to explore it space time. So, I was really fascinated by it because, as we all know, gravity doesn't just pull on space it also pulls on time, so I just wanted to understand, in depth, how things work. But then I think I really was fascinated by the whole technology, because I did my engineering and (ph 21.42) computer science and that was the early 2000s. So, I think what really excites me basically, is the amount of impact one can actually bring using data and technology. It's the personal growth, it's the excitement of taking various challenges, and being able to work with all the talented people around you. I think that's what really drives me. When you start a career, when you start working with data, I think it's really important to understand the different characteristics of data. When you talk about big data, we always talk about the seven V's of data, right, which is volume, velocity, variety, variability, veracity, visualisation and value. So, it's really important to understand what this means (ph 22.23), it's important to understand the context of data, understand the data journalism that I had mentioned earlier. You need to have a clarity of the business problem.
So, if you want to learn something, start small, stay focussed, but then, most important is, you should have fun playing with data because (mw 22.40) with data helps you unravel a lot of mysteries, a lot of (mw 22.43). It, sort of, takes you to a deep dive world, which is what you guys experience with (mw 22.49). So, as long as you're passionate, I think, yes, technology is growing at an exponential pace now, so a lot of current problems can be solved using technology.
Moderator: You're basically saying that to get to your career point, you need to be not just really good at some vertical (ph 23.08) skills, you need to be on top of trends, you need to know where the world is headed, you need to be a little bit plugged into media? Basically, you have to be quite well-rounded to be able to implement an entire data solution from top to bottom? It's not just about coding up and creating a server and downloading things, it's about more than that?
Suraj Rajan: Absolutely.
Moderator: And that goes the same way with the business users. They're not just working on their business, they're trying to empower themselves with all these NOCODE tools and things, so they can meet you midway, and together, we are trying to transform businesses and entire sectors, hopefully, economies, with the current situation, that will be the need of the hour. We'll walk into a future where our transformation and innovation, and we already have walked into that future already four weeks ago. Things that didn't happen in years are happening in weeks, and we're hoping that we can continue building this world with more, if not the same, velocity than before, and the new normal will probably require us to look even more at data.
Suraj Rajan: Yes. If you look at-, ideally, it should be a joke, but it's not really a joke, which is on social media which said, 'Who led a digital transformation of your company? Was it the CO? Was it the CDO? Or COVID.' So, a lot of companies which really have tried to get into digital transformation has managed to do so in the last one month. But having talked about transformation, I think a brilliant use case that recently came out during COVID-19 was the healthcare digital transformation especially in Taiwan, which is just 130 kilometres from mainland, China. They had a detailed mapping of who actually got in from where, and they were able to stop a lot of transmission early, so you integrate with the national insurance database with the immigration and customs database, and they could track a fourteen day travel of all the histories and symptoms. So, that shows the part of data and technology, so just using the right amount of (inaudible 24.59) integration of data. For example, of course, you need to have a circle of trust because all the hospitals, clinics and pharmacists have been given access to this information for every patient, but once you, sort of, establish that circle of trust, you can actually do wonders.
Moderator: Absolutely. I think it's all down to trust. Everyone talks about the power of data, but I'm always thinking in my head that it's like two sides of the same coin, it's power and trust. Data is powerful when people trust you enough to enable you to use it, and they trust you enough when you actually are going to do something powerful with it, so they're two sides of the same coin.
Suraj Rajan: (Toeo 25.36) For example, the numbers that comes out of China can be trusted because, yes, (toeo 25.43).
Moderator: Yes, we just don't know what's going on.
Suraj Rajan: Exactly, and we can't trust the US numbers because there isn't enough deaths. So, like you said, I think trust plays a crucial role when it comes to working with data.
Moderator: Yes. I think that's a very valid point. We saw with all our customers as well, working across different sectors and using our technology, there's a lot of trust element that goes into data. Data, in the end, looks inorganic. It's not, it's generated by organic methods, but people always think that data is this inorganic thing, and if it's not a human, then how do I trust it? But then, at the same time, they can't trust things without data either, you know, McKenzie (mw 26.20) was saying, 'In God, we trust everyone else must bring data'. So, without data, they also don't want to make completely wrong decisions based on intuition, things like that. So, we're at an interesting cross section in data and technology history, where the entire world has moved online, and there is an outstanding amount of conjecture in what will happen next. One thing is for sure, everybody wants to have more data, and where I worry is that data is not always equal to information. So, what we should be yearning for is information, not necessarily data, which is data plus trust plus metadata, plus people and systems that respect that and honour that. All of that goes into what will be called the information system of the future. I, for one, am really looking to you, me, Jenna, all of us, playing a big role in this new ecosystem.
Thank you so much for your time, Suraj. It was absolutely amazing speaking to you, and finding out your experience of building data solutions, the key skills needed to get there. I'm learning on some terms like how to do the W's, the questions that you ask for data, and what does it mean to be a first class data citizen? You spoke about how different organisations are changing their mindset, how COVID has accelerated digital transformation, and what it really takes to build a solid data powered system, and a lot of the component is trust. We spoke about infrastructures that can generate that trust. Thank you so much for your time, it was really amazing to speak to you, and I hope you get more time with Lincoln now that you're at home.
Suraj Rajan: Thanks, Joyeeta (ph 27.57). Really enjoyed our conversation and really hope all our listeners enjoyed too. Thank you.
Moderator: Everyone out there who's listening to Samsara, you heard Suraj, he made some really valuable points about how to go about thinking about a data powered organisation. If you're just starting out on the journey, you can always reach out to him for any, kind of, help or advice. He's available on all the social media, and if you get in touch with us, we will put you in touch with him. Suraj is a fabulous resource to understand where to get started on your data driven journey. Until next time, this is your host, Joyeeta, signing off. Bye.