[Radar Recap] Unleashing the Power of Data Teams in 2023

Hi, I'm Maggie, VP of Curriculum at datacamp, and I know firsthand how overwhelming it can be to keep up with the rapid changes in data and AI. But staying ahead doesn't have to be hard. That's why our curriculum team has curated the best learning content to help you and your team stay ahead, covering everything from data and AI literacy for busy executives to cutting edge generative AI for developers, and everything in between. Whether you're upskilling yourself or your team, we make learning practical, engaging and flexible. It just works. Power your data and AI transformation journey on datacamp.com.

Hello everyone. Welcome to Data Framed. I'm Adele, data evangelist and educator at DataCamp and data framed is a weekly podcast in which we explore how individuals and organizations can succeed with data. In this episode, I wanted to share with you our fourth and final Radar 2023 session that we're releasing on Data Framed. And this one is with Vijay Yadav, Director of Quantit of Sciences at Digital Data analytics at Merck, and Vanessa Gonzalez, Senior Director of Data Science and Innovation at Business Solver. This one was hosted by Richie and had tons of amazing insights on how to build high performing data teams, especially in the environment we find ourselves in 2023. Now, for newcomers who didn't listen to yesterday's episode, just as a warning, the format will be slightly different for these episodes as oftentimes we take audience questions as well and and moreover, the audio quality is that of a zoom call, but we hope that this brings additional perspectives from Radar to a wider audience. Now let's get started.

Hopefully if you've joined this session, you're aware that making better use of data can have a pretty radical impact on your organization. However, just hiring some data professionals doesn't magically make that happen. So today we're going to talk about how to hire and build data teams so that they can have that high impact that you all want. So I'd like to introduce Vijay Yadav, who is the Director of Quantitative Sciences for Digital Data and Analytics at Merck, and Vanessa Gonzalez, the Senior Director of Data Science and Innovation at Business Solver. So both Vijay and Vanessa have pretty extensive experience in hiring data professionals and also building out data teams. So I'm very excited excited to hear their expert opinions. Now we're going to start the session with my questions, so we're going to chit chat for about half an hour, but we'll also leave plenty of time for your questions in the audience. Please do make use of the Me Q and A feature. You don't need to wait until the end in order to ask questions and we'll try and get through as many of your questions as we can at the end.

So.

Hi Vanessa and Vijay, great to have you here. So I guess let's just start off finding out a little bit about who you are and what you do. So Vanessa, just to begin with, what does Business Solver do?

Hi Richie. Well, thank you. First of all, thank you for having me. Super excited to be here. And Business Solver, it's in a very interesting space. It's a software company, so it's a tech company but it works on the HR industry. So we are benefits administrators. So we do software and administration of benefits benefits. So think about these. When you, when you get hired by a company, let's say that you work for Coca Cola, you go in and they, the first thing they tell you is you have to pick your benefits.

And.

How are you going to use them. And they tell you about your HSA and all these different options that you don't really know exactly what to do. Well, we help with that. We create the software, we create the chat box, we create that content for you to have a better experience and help these companies do that. So that's what businessolver does. I lead the data science, machine learning, engineering, BI and linguistics team in businessolver and I have to say I have the best job in the world. So lots of fun.

Fantastic. And Vijay, so I think Merck is a little bit higher profile. But just for anyone who's not heard of it, can you explain what Merck does and what you do at Merck?

Sure. Thank you Richie and thank you audience for joining. I'm so excited to be here. Merck has the name recognition. If you don't know, Merck is some of the products that are very life saving. Merck is one of the top pharmaceutical company and I guarantee you one of your family members must have used the Merck product because Merck is big into vaccination space, vaccines and biologics. Some of the products for vaccine for MMR for example, very popular one HPV Keytruda is immunotherapy oncology product. So quite a big presence, is an international company and I am heading a data science for Merck within manufacturing division and happy to be here. And my experience is primarily applying artificial intelligence, machine learning and data science to improve the operational efficiency and bring some new cutting edge technology for the operations of any manufacturing company and in general other business functions.

Okay, brilliant. Well glad to have you both here. So let's talk data teams now, I think just to make sure that everyone's on the same page. The idea of a data team is a little bit fuzzy sometimes because there are so many different roles involved. So can you tell me what does a data team consist of to you, maybe Vanessa, do you want to go first?

Sure. As you say data, it's very broad, it has very different areas. You have the teams that they're going to be using the data, you have the teams that are going to be governing that data and you're going to have teams that they're going to be cleaning that data and organizing it and architect. How are we going to use the data? In companies, companies like ours and others, they have a ton of different data. So you have to make sure that you have those pieces. When we're talking data teams, I always think about depending on what area you are. When you're thinking about users like in my case that is more the machine learning, AI analytics space. We're users of data, so we're not necessarily organizing the databases like what I'm doing now. In the past I have had teams that they're doing more the governance, the data quality as well. But what you're going to have to think about is that when you're using the data, you have to have a really deep knowledge not just of how to use it, but also of the data itself. You have to understand what's the architecture, what are the limitations, where does it come from, how, where does it go, what's the size, the quality of it. Then you can really work on it. When you're thinking data teams, I think as a huge environment that you can specialize maybe in one area, but you have to be cognizant of everything that happens surrounding the data. So I don't know if that answers the question, Vijay.

Yeah, so lots of different roles from sort of data cleaning to managing, analyzing and several things like that. All right, so Vijay, I guess as being part of a larger organization, do you have a different take on this or no?

I think yeah, and Vanessa touched some up on that. So if we think about the data, right, so you start from a source, somewhere data is basically generated, right? And it never dies somewhere it is just moving, right? So think about from the source, it starts and keep moving till it goes and is stored somewhere. And then after that somebody using the data, enabling the data for decision making process. But the data teams are starting. How do we ingest right from the beginning from the source, you are basically transforming the data and storing somewhere Then getting the insights out of that using some type of model. So data creations, ingestion, cleaning and storing and then using it. I think all the roles that you see in overall process from source till getting insights from that, I think anybody who is enabling all those functionality are enabling the insights. I think that's what the data teams are quite a number of roles. We can talk a little bit more on that during the chat. But I think overall, I think anything between source and making use of data, those are all the data teams.

All right, fantastic. I think a lot of our audience data camp tends to be data analysts, data scientists. It's important to remember that that's just one section of a sort of broader data use case or workflow throughout the company. All right, so what a data team is. Oh, sorry, go on now.

I was going to say that just so you touch the point. So the people who are making the plumbing work for data is a data engineering portion. Right. Then data is. There's a data scientist. And then once they store, you need to manage, you need to govern, you need to create the visualization, all those data governance and visualization folks and machine learning people. So everybody's kind of playing the role from beginning to end to do that. So those are some of the roles that you just talked about.

Fantastic. And so we've figured out what a data team is, but we were going to talk about high impact data teams and high impact sounds very cool, but it's a little bit abstract. So I was just hoping you could both provide some examples of what it means to have an impact as a data team. Maybe Vijay, do you want to go first this time?

Sure, absolutely. So the high impact data team. So at the end of the day we have really understand is that any work that is being done by data team, it is all about the business outcome. If whatever the work being done, if the work outcome is not giving a value in terms of either growth or improving the efficiency or cost savings, whatever the outcome is defined in high impact is how quickly you can give back value using the data. That is the criteria. Now there could be cases whereby high impact, but it might take a lot of time. Right. So high impact, you know, depends on the corporation. What do you mean by high impact? Are we talking about millions of dollars? You have billions of dollars what that number is, but you can again vary. But the idea is that you want to and data is still new. You know, it's a complex landscape, right. And you want to drive the value quicker the better. Now you want to go low Hanging fruit, deliver the value quickly, whatever the data is available. Based on that majority of the cases, data is not available to make the high impact. Right. So there are certain piece of work while you're delivering in the short term, leveraging the data, existing data, but then you're building your future with the new data assets for the high impact. So I would say low hanging fruit. Go and deliver the result, show the value of data in a bigger corporation, not everybody sees the value in the same way, but if a data team, a data leader, can basically go and deliver the results quicker, I think that's when you have your allies on your side. So depending upon how, how quickly we can deliver, I will go for that. But there's some foundational work that needs to be done as well, so we cannot ignore that piece as well. So do foundational work while you're putting the future, but deliver the results quickly and make the business impact possible.

All right, so really the impact depends on, like, what your goals are and like, are you actually achieving the results you set out to achieve and how quickly are you doing that?

That's right.

All right. Okay, so maybe let's go back to the start, like how you build out a data team. And I know Vanessa, since you sort of fairly recently moved to business solver, this is something you're kind of involved with. So what's the first step in building out a data team?

Yeah, so it really depends on what area of the business you're going to be doing. Like, are we, if we're talking about a data team that is going to be bringing in the data, cleaning it up, preparing it for use, creating those databases, are they going to be on cloud? Are they going to be on prem? Whatever you're doing, that's going to be one piece. But let's focus on the data science and the analytics team to give us an example of how you build it. I think something very important, as Vijay was saying, is that you're going to have to add value. So where are the opportunities for the company to add value on using the data? In some cases it's getting insights from the data and that would be like the business intelligence teams are great at it and visualizing that data. But in other cases, like machine learning AI, you're going to try to automate processes, predict behavior, and change behavior with it as well. In those cases, it's going to be a little bit different. What I think that is very, very important is that you have the skills you need to cover the whole process or the whole cycle. Like if we're talking data science, a data science cycle that you're covering it all, you need to be able just to, if we go on the machine learning side, you need not just to create models because if nobody uses those models, then you're going to fail. What you need is to create models with good data. If the data is not good, then you're not going to be able to get a good output from those models. You need to have good data and have data engineering teams that they're going to help you with that, cleaning it up, getting it ready, moving it from point A to point B. Then you're going to need data scientists, computational linguists, depending what you're doing, machine learning engineers, that they're going to be creating models to use that data, then you're going to need to be able to productionalize those models, get that result from your models somewhere for you to be used. You're going to be having to. Machine learning engineers are going to help you on that side. Then once you have the output in a good place, then you're going to have to work with your IT teams, for example, to integrate it, let's say in Salesforce, or be able to send an email with that information or do that integration piece. Your data teams are going to depend a lot on your business, business and how you do business. But you're going to need to have all those different skills you cannot have when you're creating your data teams. It's not good to have eight people that have the same skill and then one that has a different one. You need to find those skills that you need in different places. Now the tricky part is that data teams and machine learning and bi, it's not very compartmentalized. You have people that do a little bit of this and a little bit of that. So what I always think when I'm creating my teams is keeping it pretty open and see what are the strengths of the people that I have. Then when I hire, I hire for those other skills that I'm missing. But I make sure that I take advantage of the skills that I have in different teams. So a lot of times, one example, my BI analytics and my BI engineering teams, they're separate but they collaborate a ton. So it's a, it's, it's a team sport. So in a soccer team, you don't have all goalies or all defense or all offense. You, you have to have a little bit of everything. That's how I see data teams. You need a little bit of everything for it to work. To work.

Well, that's brilliant. I love the, the soccer team analogy. Yeah. A team of 11 goalies would not be. Wouldn't score. Wouldn't score very much, I'm sure. I mean they got a good defense anyway. So I'm wondering, is there a natural order to hiring? So you suggested that you need to have some sort of data infrastructure in place. So maybe you hire the data engineers before the data scientists. I don't know. Vijay, do you want to comment on this? Like what order do you go about building out?

The.

I think one of the element that we have to consider before we building a data team in a holistic way. So think about there are four. Any data leader and data organization, there are four piece of work or four pillars that can be basically working on. Number one is development work, right? So whether you're developing a data product or developing a data pipeline, right. So there's one area that is to be done is a development work. The second one is research and development, R and D. This is the remit of the people who are basically working on the new ideas. Necessarily they are not developing actual data product, they're thinking of something new. Right. So that's the second element of that. The third element is operational support. Right. So once the model goes into production, you need a group of people to basically support that as well. And the fifth is more like consulting role. So you as a data organization, you're providing consulting services within your organization to other teams or to other business. So if you take these four elements of the work, depending upon what is the limit of that data organization is you want to be getting people in full, you know, in four buckets. Right. Once we decide that, you know, let's say 50% on development, 25% R&D, 25% this and maybe one area, we are not doing anything. I'm just taking an example, for example, consulting, not everybody's doing. So once that allocation basically takes place. Now let's understand the remit of the work that this data leader, data organization has in order to enable. What are the different skill set in covering those three areas that you're focusing on? Development work, R and D and support. What do you need? What is the body of support basically needed? How much body you basically need to support. And you go back and look where you are in terms of what resources, skillset you have. What do you need for your remit? You take the delta and basically make a plan to either hire a recruit or develop the people to meet that need. I mean, that's the most organic way to really think about building the data team overall.

Okay, so it does seem like again we're going back to the idea of trying to fill out different blocks of skills and make sure you've got a balance. I'm wondering what happens when you do find you have gaps. Have either of you used sort of upskilling or tried to fill in gaps by training existing employees?

Yeah, no, definitely. And that's where like for example, data camp is super useful. You're not going to find people that, with the same experiences with the same skills that you need. And also the other piece with data teams is that it's a super fast changing environment. Like we're like four years ago, we, the needs of the skills we had were, they're very different than what we have today and that we're going to have in five years. So we need to have like data teams. Used to be they are very, they're, they're learning all the time. So it's sometimes when you have gaps, it's a lot easier just to get your own team to learn these new skills that you're missing than trying to find the skill. Like you're not gonna, you, you cannot have 300 people with skills like each of them with one skill because then can you imagine coordinating that? Even that would be crazy. So sometimes you're just, if you're needing, if you have a team that they are like heavier on the data science side and you don't have a DevOps team, then maybe you start like your, your data scientists, they learn some DevOps skills to fill that gap. If you have a small team, you see it even more that they have. You're going to have a machine learning team that has data science skills, data engineering skills, because they need to know more to be able to get done what needs to get done. But yeah, that's something that I always focus in my teams a lot of the time. I always want them to have time in the week for development and for learning because there's always going to be new things if they don't know them or things that we did not need before that we need now that we have to learn about. So that is something that, it's very, very important for anybody that wants to work in data. I think that you have to be ready to be a long life learner. That is going to make you like a very good data person. If you're not there, then it's going to be a little harder. You're not going to be doing the same thing, the same way for very long, I don't think.

Absolutely. I have to say I quite often get questions from people saying, well, how long does it take to become a data scientist? I'm like, I've been doing this for 20 years. I still know like 1% of data science is always something new to learn. And so that was a really interesting opinion, like from the sort of startup point of view where people do have to kind of shift roles and take on responsibilities for like maybe outside their area of expertise. I'm curious as to how it works in a larger organization. So Vijay, perhaps Merck, I'm presuming you have like sort of maybe a bigger like Data Academy style setup or do you have any sort of official upskilling program?

Oh, we have very official upskilling program and upskilling is a big here at Merck. And I think that's, that's really, that's the very fundamental organic way to basically change the culture of the organization when you have the upskilling program. And as part of the skilling that we have two pharma program, one is being more like technical in a group settings, whether it's R or Python. And by the way, Data Camp is a big. At Merck, we're using Data Camp very heavily. We're using other platforms as well and we have curated our own content basically for the training that we're conducting. One of the other element which is in upscaling is mentorship. Right. So there's a group setting that you can have, you know, weekly. And the way we run our workshop, we have weekly workshop, every week, fixed time, people join the cohort and you basically finish from beginning to end. And the new cohort basically begins around 8 weeks to 10 weeks time to get to that a different level, right. There's some fundamental, then there's some advance and they have a different track, right to that. I think the highest impact we found was a mentorship program because that is one on one. And the way we do here is that we have mentors who are expert in certain area of data, whether somebody is coaching on Python and R visualization, anything of that nature. And we would identify those volunteers, people who are willing to give their time to coach and they're the mentees who are interested into getting coached in certain areas. So we pair them up based on skill set block the mentors and mentees and you do that very effective one on one. Of course, scalability is kind of slow. You have a limited number of mentors and mentees But I think having the workshop style, the adult learning is different. People who want their workplace are different than student learning. When you're in class, there's a hand holding element. I think what we found in upscaling is really, really important. Having the workshop, live, workshop and cohort kind of scenario where they can work together and data community within Merck is big. So we bring people together and if you can make them interact with each other, they can find their organic and natural way to interact and mingle and ask the questions what they are learning. That's another way of upscaling. While you're enabling through a data community. Those are some of the avenues and enablers here at Merck that we leverage in a big way.

Back to what Vijay was saying, something that has been very helpful for us on the upscaling side is making sure that the projects are done by teams. We don't have just one person working on something. I always try to have at least two or three people working on something because then you have a more experienced person that it's helping the more junior people to learn. But at the same time the ones that they are more junior, they come with newer skills from their education that they just got like in those cases. So having that, that mix and working on the same things, I find it that it's a very organic way of moving forward and you have to use also some more formal ways of scaling. But I definitely find that getting people together and having that community and that culture of helping out and of, of helping others, that's what makes the difference to get there.

Yeah.

And I can add another element of that. So I, you know, we hold hackathons on a very regular basis. Hackathons are the big way to really make people learn. And one of the element, when I run the hackathon, when we make the teams necessarily, we are not hiring and putting people in those teams who are only data people. We bring people from the business side of it, put them in the team. Their job is necessarily they're not even contributing in any way. They're just being part of that team, just listening and seeing what is happening. That is a learning that you cannot teach other way. When they see things, live without anybody teaching them anything, just listening to that, it has a huge impact that way. So upskilling is not necessarily. You put someone in the room, I think building a community or something, engagement, the way you can engage them. I think that is another avenue for upscaling as well.

I was pleased to hear that both of you are big Data Camp users. But a lot of those ideas about not just learning, but also having that community aspect where you're doing cohort learning, where you're having multiple people on teams together, that kind of social aspect from the hackathons, that just seem really important in motivating people. All right, so related to that, I mean, there's always like tons of people who are like wanting to get their sort of first job in data. So I'm kind of curious when either when you have new hires or when you have someone transitioning from a different part of the business into data, how do you go about saying, okay, this person has good enough data skills to join my team?

Yeah.

Oh, Vijay, Yeah, sure, I think.

And listen, that is. That is always the question on the mind. One of the element for me is that when you talk and we talk about the business value, business impact, right? One of the skill set I'm looking is the business acumen skill. If you cannot see how the company is making money, if you cannot see how the business process in that company is organized to deliver certain values is being there. I'm not necessarily looking pure technical, but I'm looking into that aspect. It's really, really important because it's very difficult to be, you know, not knowing what the business wants and how to even understand the business value, how company makes money, how the business process is connected to do that. I'm always looking that kind of skill set in that aspect. Take the broader view of that piece. Right. And then of course comes to the technical. Right. So technical one, I think, you know, curiosity and someone, how good somebody's willing to learn. I can guarantee you that if you go and try to hire the super expert, either you be ready to give the right job, otherwise the guy is going to quit fairly soon. So if you're looking for someone you want to develop, then the one thing you want to be looking and paying attention is that the curiosity of that person, how are they asking the right questions to the business? So the point I'm trying to make is that, you know, technical skill is definitely there, but the business acumen is really super important, you know, for me. And also, you know, when you're developing a solution, you are developing solution in a holistic way. Not the narrow focused and a problem. Right. Problem never happens in isolation. I'm looking for someone. Can you see the problem beyond what's happening? Can you connect your dot upstream or downstream? Right. If somebody is causing the problem from down, upstream, downstream, getting impacted. So if someone who has that view of that I think, and if your data is scale, you can go, you're going to make impact much, much faster. So that's some of the things that I'm basically looking into that that's brilliant.

And Vanessa, do you have anything to add to that beyond the sort of the technical skills and the business skills?

Well, I would say that for me there, it's super important, like the soft skills. So communication is a big one because we don't work in isolation. You have to work with a business to create products that they're really going to be impactful and add value. I look at somebody that is very good at collaborating. I don't. I have a team that collaborates beautifully together and that I'm always, when I'm hiring, I'm making sure that the new person coming in is going to be in that with that same attitude. I need someone that is willing to work in a team environment and not by themselves and not looking just for themselves. So those soft skills and being able to talk to others, care about others, collaboration, communication to me are crucial. The technical skills. And the other one that's a big one is thought process. Also, I'm looking, when I'm hiring somebody new, I'm not just looking at what, what they have done because maybe they have not had a lot of experience, but I want to see signs of wanting to learn, wanting to be curious, as Vijay was saying, and be creative on your solutions. You may not, it might not be the right solution, but you have to think out of the box. So all those pieces are the ones that to me, they make somebody very different and very successful. And Python, maybe you're not super strong in Python, you can get strong by using it, but you have to be, you have to have that creativity of how you're going to use it to be impactful and solving problems. So those are the, the pieces that I always look for, somebody that is willing to learn and good at getting feedback that is very, very valuable because that's what's going to help the, these team members grow.

So we're going to go to audience questions in a few minutes time. I think I've got time for maybe two or three more questions before we get to that. So one thing that seems to be an important thing where you're trying to get new skills, you can either train people as we talked about, or you can hire new people. So I'm curious as to how you make the decision about, like when do you hire, when do you just train existing staff? I know there's talk of like a recession, a lot of people, a lot of companies are cutting down on hiring. What's the sort of calculus you use for this? Yeah, Vijay, do you want to go for this?

Yeah. So I think in my point of view, and that's how I do is I have a mix of resources. I do try to develop from inside the resources. There are people willing to come in and each side has its own, you know, plus and minuses. When I bring someone within the company, one thing I they bring with them is the knowledge of the industry, the knowledge of the company, the business process knowledge. Right. Technical skills can be taught. You know, we can upscale, actually train that. So there's a mix of people from inside, you develop them, you certain strength they come in, then you bring from outside the expertise that you don't have inside. Right. Very simple, you know, any highly technical area that you want to bring and there's a time crunch, there's no time for training. Right. And experience of a piece that you basically bring. So I would say you balance between the two, inside and outside as well. And what I found is the people who are inside, most likely they're going to stay longer because they're willing to change, they want to take the career. While when you bring from outside, you're taking your chances, whether that person fits into the culture or not or things of that nature. So I'm always giving a chance to the people who are internally to develop that. But I'm happy to go and fill some specialized skill that cannot be found inside the company.

All right, so once you've got your high performing team, your high impact team, how do you make sure that you're keeping your staff happy and you're retaining them? Vanessa, do you want to take this main thing?

I think it's going to be culture. You have to have a culture about learning, developing, making their job fun. There are days, of course, and things that we do that we are not our favorite things to do, but keeping that balance of data teams in general, they're people that they want to learn and do and be impactful. So I try to always keep that in mind and making sure that they're not just working on what it's needed at that moment, but letting them explore, develop. We are in an innovation team. So to innovate, you have to have that support of having the time and the resources to go there and do hackathons and have it as part of your culture of keep them doing that. So I would say that the main thing is keeping a good culture on your team, keeping that collaboration and communication between us and really caring about your team, caring about the people. You get a whole person, you don't get just a skill or a tool. You're really getting a person. Trying to get to know my team, trying to get, to help them grow and get them get where they need to get to.

Right.

It's like that's, that is, I, I think that's what makes a team strong and make our job more fun too. Like we spend too many hours together doing, doing things, we want to have a good time doing them as well. So I think for me that's how I keep my team and that's that, that's what I'm always hoping to do better and better. So, so they, they're fulfilled with what they're doing.

Yeah.

It turns out that sort of caring about your staff actually works and makes them happy.

It's not just something you say, you really have to do it.

Brilliant. Vijay.

Yeah, I was going to add, you know, so think about why should, why should somebody work for me? As simple as that. That's a question for me. Why should they work, you know, for me, Every team member, you know, if you think about their career growth, right. Are they learning, am I going to give as a manager, a leader, give them opportunity to grow and learn? If they can see that, that it is not only they're working on something, but they are growing, are they getting opportunity? Right. So what I'm talking about is the ladder, right? So if somebody's joining in one role, can they see their next role, maybe next to next role? If that visibility is there and they understand that they have a future ahead, most likely they are going to stay around. Right. And so it's very important that they have that visibility of their career. The second element, I would say that it's really important that they have a high engagement to be retained. If data scientist is not doing a data sciencey thing, you think how long the person is going to stay, right? So you want to be making sure that they are getting what they enjoy and what they have been hired for. If we as a leader, we hire someone for one task, we give them something else, basically it's not going to be happening. And I think the appreciation, the whole idea is can they see the value of their work? Are they appreciated by the management, by the leaders? That's another element of that for them to be engaged and be proactive and coming every day happy to do that. So I would say those are some of the elements that I Would keep in mind. And sometimes money is not everything. Appreciation and inspiration and career growth as they grow, of course the money would get better. So those are some of the elements that I'm looking into in order to retain the people. Mentorship for me is a big one. You know, are they getting growth out of that? So those are some of the elements that I really focus on.

Brilliant, so engaging work, feeling appreciated, career progression, all very important things. All right, so with that, we've got five minutes left for audience questions. I don't want to revert. I got told off before. So let's move over and see some questions. All right, so question comes from Chris. So Chris asks organizationally how much is inside the data team and how much is in the business or in it? So how do your data teams interact with the other teams?

So I think the way the data has been emerging, and definitely it has been true at Merck as well, is that we have data teams that are federated in nature and that is organically, that is how majority of the organizations are doing well. Data teams are federated, are embedded as part of the business unit where locally they are helping the value. Getting the data value out of that same started with other technology companies as well. As the organization basically mature, it becomes more data aware. The data value is basically realized. I would think that you can start centralizing it. What I have seen and sometimes the teams are part of the business, sometimes it's part of the it. I preferly, you know, when I look for something where if it is part of the business, then one thing is sure that there's a whole accountability setup. Right. If you're part of the business, then you're definitely working, you know, deliver the value because you're directly working with the business. Both the models, whether you're part of the IT and whether part of the business both have pros and cons. My personal choices would be being part of the business because then you are going to be directly delivering the value for the business overall.

Next question comes from md MD is asking how can I make a project which is unique and has real world value? So yeah, I think as particularly for people who are wanting to get into data, how can you show off your skills with something that feels realistic?

So I would, I would say start with the data and with, with what you want to solve for. So think about a project or a problem that it's really that like maybe, maybe you have the data to say what is the shape of the, of the clouds and if they're going to look like a, like a bunny or. Well, was that really impactful? Is it solving for anything? So think about what problem you're going to try to solve. Then you can see if you have that data available and what data you can use and then you can see how you get there. So for me it's all about adding value. You could be adding value for somebody just to pick a movie like when you go to Netflix and are watching, or you can add value by helping a company to retain their clients. So depending what you're looking at, I would say pick something that is solving a problem, that it's really problem and use be creative. Creativity is going to be what is going to get you there. And then there's all. There's many ways to solve a problem. You don't have to solve it in a. In a specific structured way. You can, there's always different ways and that's what's going to make you more successful, I think.

Brilliant. And one thing to add to that, yesterday we had a session on creating a portfolio, a data science portfolio. If you're interested in this sort of thing and want to show off your skills, then please do check out the recording of yesterday's session. So yeah, it's one to look out for once the recordings become available. I think we've got time for one more question and then I'll show you off to Martin. So Jurgen asks, how diversified are your data teams and are there also different cultures included? How do you communicate with teams localized in different cultures? Either one of you, whoever wants to.

Take that, I can take that. So I can tell you that we have a very diversified team and I really focused on that. And Merck as a company, as a culture is big and huge on that. So data leaders are basically measured how diverse their teams are. So there's a whole metrics that is basically done. And I found is that having a different again culture, I don't have to explain diversity brings a lot of values. Sometimes we don't say it, but it does bring a lot of value. Different cultures piece to that. And being a multinational company within Merck, I mean there's absolutely no way to escape. And we basically promote and I've seen some really wonderful ideas the new thinking coming out of the diverse team. So very big on diversity point of view, not diversity in terms of demographic, but diversity in terms of thoughts and people from different background. Right. We hired in some cases a chemical engineer with not much background in data science, but they bring a very practical experience being in the chemical industry. Or in pharmaceutical manufacturing, big time. So diversity of thoughts, demographic in other country wise, region wise. I think all those elements are very valid. To me, it's a big element of the hiring process.

And I would like to add to that. Richie, definitely for us and business all over, diversity, it's forefront. So you have to be thoughtful about it when you are doing your search and then thoughtful about it of how your team looks at certain moment and what do you think it would need? As Vijay was saying, it's not just diversity, just like getting people from different areas or countries. It's more about cultures, about culture, age, thought, like thought process, backgrounds. So having a mix of different people, it's going to make you a lot richer. And the ideas that they're going to come to pass and the ways about how to learn from each other, it's very impactful. So we're always thinking about that. We're a very diverse company and we're looking to be even more in the future.

Fantastic.

Yeah.

I mean, this is such an interesting topic. I'm sure we could do have like a whole session just on like diversity and data. Maybe next Radar conference. We'll do that. All right. So with that, thank you very much, Vijay. Thank you very, very much, Vanessa. Thank you to everyone who asked a question. Thank you to everyone who showed up. This has been like such an amazing conference. I'm going to send you off to the closing session now. Martin, our chief operating officer, is speaking. He's closing up. Hope you've enjoyed the last two days. Thank you once again and goodbye.

Thank you so much. Thank you, everyone.