On the 12th episode of Enterprise Software Innovators, hosts Evan Reiser (Abnormal Security) and Saam Motamedi (Greylock Partners) talk with Bijoy Sagar, Chief Information Technology and Digital Transformation Officer at Bayer. As one of the largest pharmaceutical companies in the world, Bayer positively impacts billions of people through technology innovations across healthcare, agriculture, and biotech. Today, Bijoy shares how Bayer is deploying digital farming practices, his perspective on AI, and the best methodologies for partnering with startups.
Quick hits from Bijoy:
On Bayer deploying drones to optimize farming practices: “What does it actually mean to do digital farming? We have drones over 73 million acres where we're collecting data real time on the field. We have satellite data coming in. We actually acquired a company to get the data so you can actually predict how much soil moisture is there in one square meter of the land. And then you can actually use algorithms to predict how much seed you plant there and how much do you water? What kind of resources do you need there so that you can actually grow, get the best outcome for the farmers in the most sustainable way? And so that's what digital farming is in a nutshell. This may not be the most critical digital topic somebody would think about until you think about the fact that this is the food you're eating every day.”
On building mission-driven teams: “You have to have everybody be mission driven. We spend a lot of time purposefully looking at ‘how do we build the teams together, how do we actually get them to be mission focused?’ I don't start a single presentation without first referencing our mission and purpose, ‘Health for All, Hunger for None.’ I always remind them you are here every single day because there is a patient at the end of the journey, there's a farmer at the end of the journey…You cannot go solve problems of tomorrow with the tools of yesterday.”
On the technology frontiers of the future: “The best is yet to come. And what do I mean by that? Some of the really complex pharmaceutical problems such as protein folding would require a 1000 qubit computer right now. We are playing with 40 qubit, so it's gonna be amazing but we're not there yet. So I don't want people to sort of feel like yeah, this is the pinnacle of digital; I don't believe that. I think we will look back six years from now and say my God, those were primitive days!”
On step changes coming to medicine: “Cell and gene therapy in pharmaceuticals is going to completely revolutionize the way we do drug discovery so that we can actually do amazing corrective solutions for diseases that have no treatment today, such as Parkinson's.With better models with stronger AI, [in the future, we will have] the ability to do protein folding and visualizations. We should be able to treat Parkinson's disease and cure it. I'm happy to predict and we'll see if I'm true or not in 10 years that Parkinson's will be a curable disease, not even necessarily a manageable disease.”
On the framework for engaging with startups: “Look for non weaknesses in the way of solving a problem. I always tell people, if band aid and bailing wire is the cheapest, best, most reliable way to solve a problem, that's fine because there is a role for a band aid and bailing wire, otherwise we wouldn't be making those things, right? So I wouldn't look for a startup to invent a problem to solve. As you say, if you only have a hammer, then every problem looks like a nail.”
Recent book recommendation: Sapiens by Yuval Noah Harari
Evan Reiser: Hi there and welcome to Enterprise Software Innovators, a show where top technology executives share how they innovate at scale. In each episode, enterprise leaders share how they’re driving digital transformation and what they’ve learned along the way. I’m Evan Reiser, the CEO, and founder of Abnormal Security.
Saam Motamedi: I’m Saam Motamedi, a general partner at Greylock Partners.
Evan: Today on the show, we’re bringing you a conversation with Bijoy Sagar, Chief Information Officer and Chief Digital Officer at Bayer. As one of the largest pharmaceutical companies in the world, Bayer positively impacts billions of people through technology innovations across healthcare, agriculture, and biotech.
In this conversation, Bijoy shares how Bayer’s deploying digital farming practices, his perspective on the future of AI, and the best methodologies for partnering with startups. Bijoy, first of all, thank you again for taking the time to chat with us. Super excited to talk to you again and I feel like every time we chat I either learn something or I’m inspired about some new technologies.
Bijoy Sagar: I’m looking forward to it. Thank you again for the invitation today.
Evan: Will you maybe chat a little bit about your role at Bayer and how you’re helping the company digitally transform?
Bijoy: I’m currently the Chief Information and Digital Officer of Bayer. Bayer or Bayer depending on which part of the continent you are sitting in, both works. In this role, I see myself in two or three different ways just to simplify it. One is, of course, leading a vertical line function, which is about 10,000 people inside and outside the country.
Then you have the chief evangelist, cheerleader, coach, agent provocateur of digital across Bayer. That’s the second piece, and then of course, the external face of Bayer for all things technology, digital as it comes to, and that would be my third role.
Evan: Bayer is a really interesting and complex business. I think it both does more than most people think, and serves more customers in more ways, but also is using technology in a lot of ways that most people might not fully understand or appreciate. Do you mind maybe sharing some of the ways that Bayer deploys technology in ways that might be surprising to the amateur looking in from the outside like me?
Bijoy: I’ll maybe tell you a little bit about what we do and then I’ll tell you what the technology is because I think that would ground people that are say listening to it. We have three divisions, and the largest division is crop science. We support in that way the growers of the food we eat ultimately if you want to think about it that way. The second large division is pharma.
This is about the bigger drugs. If you are in cancer, for example, or women’s health or cardio. Then you have consumer health, which is all the brands that you have heard and familiar with, Alka-Seltzer, Claritin, aspirin, et cetera. I look at the problems differently depending on which customer segment you’re serving. If you are in the pharma business, how do you do digital farming? What does it actually to do digital farming? The first piece is how do you do responsible farming in a sustainable way?
What we have is, for example, drones over 70 million acres, something close to 73 million acres where we are collecting data, real-time, on the field. We have satellite data coming in. We actually acquired a company called Climate Corp to get the data so that you can actually predict how much soil moisture is there in one square meter of the land.
Then you can actually use the algorithms to predict how much seeds do you plant there, how much do you water, what kind of resources do you need there so that you can actually grow, get the best outcome for the farmers in the most sustainable way. That’s what digital farming is in a nutshell if you want to think about it. Now, this may not be the most critical digital topic somebody would think about until you think about the fact that this is the food you are eating every day.
Now, on the other hand, if you are in pharma, completely different problem there. Our mission is health for all, hunger for none, which I think all of us can sign up for, I believe. There’s not anybody in the world who would say, “That’s not something I’m willing to sign up for.” In the pharma side, it’s a question of how you actually do clinical trials at scale.
How do you do in silico drug development before you actually do physical drug development? How do you actually design the molecules in the appropriate way actually using algorithms before you go and make them in the lab? That’s transitioning because it reduces the time to market in a significant way. Now, if you’re a cancer patient, this is life-changing because it’s taking the time away from something getting to the market. These are the things where the technology use is really exciting for me. It’s not about the technology, it’s really about people on the other side.
Saam: It’s just interesting how significant and in some ways different to each of those divisions are. If you think about crop science which is not a field, I think most of our listeners would historically have thought of as being an industry that’s digital, yet to your point, has a lot of interesting opportunities for technology impact.
Pharma where biotech is one of the core pillars of technology over the last several decades, and then consumer health. How do you structure your team to be effective at managing very different business goals and driving technology and innovation against those different areas?
Bijoy: I have three CIDOs who are reporting to me who are directly embedded in these three divisions who sit at the management team at these divisions. They’re really business leaders if you really want to think about it. They drive a lot of that intimate conversations every single day. You have to do that because you have to live that business. You have to live and die by the business to the point where your success is tied to the success of that division.
Then I have the horizontal organizations that support these three vertical pillars if you want to think about it. What works is these horizontal and vertical coming together. An example of how that comes together is I have this digital hub strategy where, for example, I have a digital hub in Holland. I have a digital hub focused exactly on cyber security in Tel Aviv. I’m in the process of building a digital hub in the Americas, and that’s where this converges. You’re solving algorithm problems, you’re actually solving data engineering problems, but they sit together, and you can get a lot of convergence at it.
Evan: One thing that’s really interesting about how you talk about this is the purpose of technology is to go help our customers and their customers and the world, you go back to your mission. How do you drive a customer-centric culture in your team to help people think about how technology can be applied to best support customers instead of only internal employees?
Bijoy: That’s the most complicated question for a number of reasons. I’ll tell you why that is because if I spend most of my time thinking about people, because at the end of the day, and you know this, technology is a people business. It’s not a technology business at the end of the day. What does that actually mean? Within IT, I define digital and IT slightly differently.
First of all, I think digital is entirely a purpose-driven part of technology. You’re there to solve a human problem, you’re there to be a force multiplier. You are there to disrupt an inefficient process to get something solved that you can’t solve without those technologies. IT traditionally has been a supporting function that enables the current business.
This is how I think about that differently and you need to bring the entire organization along on that journey. Because otherwise what happens is you have some people doing this cool stuff and some people not doing cool stuff and that does not set us up for success. You have to have everybody to be mission-driven. We spend a lot of time purposefully looking at how do we build the teams together. How do we actually get them to be mission-focused?
I don’t start a single presentation without first referencing to our mission and purpose. Health for all, hunger for none, I always remind them, you are here every single day because there’s a patient at the end of the journey. There’s a pharma at the end of the journey and you’re not here to solve the problems of today, which you are, but you’re also here to solve problems of tomorrow and you cannot go solve problems of tomorrow with the tools of yesterday.
The third point that I mentioned very easily is the best is yet to come. What do I mean by that? Some of the really complex pharmaceutical problems such as protein folding would require 1,000-qubit computer. Right now, we are playing with 40 qubits. It’s going to be amazing, but we are not there yet. I don’t want people to feel like, “Yes, this is the pinnacle of digital.” I don’t believe that. I think we would look back six years from now and say, “Oh my God, those were primitive days.”
Saam: I think this idea of balancing what’s possible with technology today versus where we are headed is such an interesting one, and you just gave a really good example of it. Are there any other tactics around how you balance that with the team so they’re both grounded in the realities of the short-term, but also thinking about innovation over much longer time horizons and time marks?
Bijoy: Yes. Look, there are lots of things that we are doing, but even with quantum computing, we’re taking some active measures today. Knowing fully well that these are not going to be the final solutions for the thing, but we need to get familiar with it. We need to figure out how to write algorithm and code properly for those days because you can’t just show up seven years late to the party. You really need to start doing that.
The team actually works on three levels. One, of course, if you’re running a big IT shop, your licensed to operate comes from making sure that things run stably. You have a modicum of stability and user satisfaction because people come to work today, they don’t come to work for tomorrow. The second piece is actually building the house while you are living in it. That second piece is the near-horizon journey. This is not where you’re waiting for the new technology. You know the technology; now how do you apply it properly?
The third one, of course, is how do you take these early steps to build up the skills, to build up the competencies so that we can use it when we get to it? You guys know this already. Most of the algorithms that we are playing with in AI, these are from the ‘70s. We need to catch up. [laughs]
Saam: I actually wanted to ask you about AI because one of the things we’re seeing on the Greylock side is it feels like we’re entering this next era of AI, is you have these large models getting larger and larger, having really nice transferability across domains, the ability to pre-train these models and drive really significant performance gains. What are you all doing in this area? I could think of tons of applications on the pharma side, on the digital farming side, on the consumer health side.
Bijoy: You’re absolutely right, Saam. Data is maturing, AI is maturing slowly, not there yet, but maturing. We also, with that, get data pollution. We get data tourism. What data pollution is, you get a lot of junk data that may or may not be helping the algorithm really predict the right things, so that happens. As the volume of data goes up, you also get that.
Then you get data tourism, which is large enterprise, data traveling from one location to the other because you just try to move this around, which does not necessarily yield better predictability. One of the areas that I’m really focused now on is how do you use AI ethically? Because if you think about it, the algorithms in the beginning of it, people are not worried about that. We’re worried about getting the algorithms, training the algorithm to work properly. There was a lot of that early industrial edge AI, if you want to think about it. Think Liverpool in the 1800s or whatever.
Now, we are getting to that second stage where we have to really think, are we using it responsibly? What are the guard trails around it, and how do you actually build it so that when we put these algorithms in production, they do what we want it to do and nothing more?
Evan: What are some of the unexpected ways that you think AI’s going to transform the industry in ways that maybe people don’t fully appreciate today?
Bijoy: I will take one from my previous example. In the previous life, could you do remote digital surgery one day using AI? Today, nobody is going to sign up for it. Why is that? It’s not just an AI problem. I think when we talk about AI, we misjudged this is an algorithm problem or a data problem. It’s also the infrastructure problem.
The way I think about it is making better wheels so that the trains go faster, is different from going from trains to hyperloop. To do trains going faster with better wheels, you don’t need to change the tracks. You already have the tracks. You just have to build better wheels. That’s the journey we are on right now.
The next evolution for me is how do you go from those trains on the wheels to hyperloop, where you don’t have wheels anymore, or at least wheels the way we think of wheels, and we don’t have friction on the wheels and we have to then build the loop first. Then we worry about how do you actually have the vessel going through the loop? That requires infrastructure commitment. 5G, for example, we talk about 5G as something real, but we all know it is still infancy and 5G has a long way to go.
There is a lot of this edge computing. I think the next big leap for me in AI is going to be really getting this hyperloop infrastructure in place so that AI can do what it does in a much faster and inventive ways. This is also true for the quantum computing. Even if we had 1,000-qubit computer, how many algorithms are quantum ready right now? Everything that we are talking about-- Because if you’re running the same algorithm on the quantum computing, what are you gaining? In fact, you’re going to get more noise in the algorithm than anything else.
We have to get ready for that. We have to solve the cooling problem. To me, I think the next few years, yes, we’ll continue to evolve the predictability of the algorithms. We are going to play with a lot more data, as you said, we are going to get this to the next level. That big jump is going to be building the loop for the hyperloop.
Evan: If you went back 10 years ago, and you were talking to a farmer, you’re like, “Imagine 2022, we’re going to be able to use drones and we’re going to be able to map out the moisture and the soil on a square meter basis and give you everything you need to know. The toy transforms how you work.” They’d be like, “What are you talking about? That sounds crazy.” In agriculture farming or crop science, how does that look different in five or 10 years in ways that maybe even your average customer would might not fully appreciate?
Bijoy: Just as I would’ve been talking about 10 years ago, anything I would say today would be wrong without a doubt, because we can’t really see where these things are going. I will tell you a couple of things. One of them is cell and gene therapy in pharmaceuticals. It’s going to completely revolutionize the way we are going to do drug discovery so that we can actually do amazing curative solutions for diseases that have no treatment today, such as Parkinson’s and Alzheimer’s.
Saam: What is that?
Bijoy: For example, look, with better models, with stronger AI that I was talking about, the ability to do protein folding and visualizations, which we can’t do today. We should be able to, for example, treat Parkinson’s disease and cure it possibly. I’m happy to predict and we’ll see if I’m true or not. In 10 years, Parkinson’s will be a curable disease, not even necessarily a manageable disease. That right there is a journey that I’m willing to take a bet and say that’s probably what’s going to happen.
The second piece is as the earth climate patterns are changing, we need all this data we are getting to build more dryness-resistant crops. We need to actually develop a new set of crops that can be grown with less moisture in places which we won’t consider today arable for that reason. Because to feed the world and to continue that journey, you have to consistently evolve this so that could be that journey 10 years from now.
Remember I said it’s all mission-managed or mission-driven. It has to be not just a technology, but it’s also how you’re growing the crops or how you are treating, how you are curing, how you are designing molecules. That’s where I think that that big change is going to be.
Saam: One of the reasons why Evan and I have been excited to get you on the show is I think you’ve been a leader; Bayer has been a leader in working with startups to drive innovation. We do have a lot of companies listening to this podcast who want to work with startups, who are working with startups. Let’s spend a couple of minutes just around the texture of your relationship with startup companies. Maybe just to start, how do you evaluate, for a given problem, when you should think about engaging with a startup or newer technology vendor?
Bijoy: Look, my own personal belief is that you look for known weaknesses in the way of solving a problem or any problem. I always tell people if band-aid and bailing wire is the cheapest, best, most reliable way to solve a problem, then that’s fine because it’s not a problem. Because there is a role for band-aid and bailing wire, otherwise, we wouldn’t be making those things.
I wouldn’t look for a startup to invent a problem to solve. In other words, as you say, if you only have is a hammer, then every problem looks like a nail, and that is sometimes a startup issue.
Bijoy: Because startups are really focused on that one thing. They try to solve it, a company like Bayer or any other company of our size, not every problem is a nail. I look for what is that weak spot? What is the problem that we need to solve? Because the band-aid and bailing wire is no longer working. Today’s solutions are also not working. Yesterday’s solutions clearly are not working. That’s where I go say, “What’s a startup thinking in that space? How are you solving that problem?” That’s one way of engaging.
The second way, this is why I spend a fair amount of time, as you guys know, with startup ecosystems, whether it’s through Anderson or any other number, because I want to understand what you guys are thinking of. Maybe I haven’t thought about it, maybe I don’t know I have a problem. Then once in a while, I get super surprised by a new way of looking at a problem. I’m like, “That’s great. Let’s see if I can solve that problem. I didn’t know I had that problem. This is your iPhone problem,” as I always call it.
Nobody knew they needed an iPhone; nobody was asking for an iPhone until it showed up, and then they’re like, “Oh, this is really awesome. It solves a lot of issues for me.” That I think is the second way I engage with startups.
Saam: I think the iPhone analogy is an interesting one because one of the things I imagine, and I think I know with you specifically having worked with you on different startups is culturally in the organization. I can imagine there’s a different approach to working with startups. These companies are early, they’re working with you often to develop products develop solutions. Failure rates are higher than working with established vendors.
How have you built the culture, not just in yourself, but in the team around being comfortable with taking that risk, being comfortable with things failing for the upside of the potential innovation that these types of partnerships bring?
Bijoy: A lot of companies have this disease called pilotitis. That’s a disease where you do all these pilots and it ends over there and nothing comes out of it in one pilot after the other so you have to treat pilotitis. The only real treatment is few real wins where you can actually scale and you can say, “This approach actually worked,” have a success story. Then people would believe you. Until that happens, it’s going to be difficult.
The other thing I think, lower in the organizations, enterprise organizations, what gets people successful is the ability to do their work really effectively and predictably in the ways that’s been defined. Taking big risks doesn’t fit into the picture because, “Hey, I’m going to measure you on how many widgets you built today and how well you built it,” rather than how many widgets you’re trying to build it but your initial process is going to be clunky, and so your numbers will go down so people would say, “Why do I do that?”
Now, for large organizational leaders in the same thing that I read, they said IQ and EQ plays a big role, so your ability to take risks and manages, like something about startups, which startled me. Startup CEOs have a different profile, and it’s called job openness, but job openness means the ability to connect unconnected things in their head.
It’s not about solving the problems of today and doing it really well and getting success, but it’s about looking at two or three things which are not connected and say, “I see a pattern here. I know how to connect these things and create something new.” That’s not a quality widely present in large corporations anywhere, and we have to build that purposefully, because that’s where the disconnects that you guys see. If you don’t have that high level of job openness, this high level of interconnected thinking, or network thinking, that’s another word for it. You probably will miss out on that opportunity to connect.
Evan: I love this concept of a pilotitis, and I also see it on the startup side. I’m curious, for founders listening to this conversation, how would you advise them to avoid pilotitis?
Bijoy: Yes. Look, ultimate question is, are your mission-driven? As I was saying earlier, I always tell my team, how are you helping the mission? If the pilot is not tying to the mission, it’s a technology-driven pilot, you’re just trying too many things, then that startup CEO is going to say, “What am I doing here?” It’s okay to pilot something to further the mission because when you fail, you’re still learning something.
You’re absolutely right, I do see, for example, I love the idea of pivoting. In Silicon Valley, you guys are always pivoting, which is nice, and it’s flexible in general because I’m not saying something, you guys don’t know already. Is the pivot still taking you within what you know? Are you still doing it with the mission or are you just wildly swinging here and there?
What kind of revenue are you chasing? Are you chasing the revenue you want to truly grow? Or are you chasing the revenue because it’s going to plug a leak you have today? If you think that through, I think you’ll make the right decisions.
Saam: There are a lot of startups that are building enterprise software, but are actually not really familiar with how large enterprises work. For maybe up-and-coming enterprise software companies, what would be the piece of advice you have? What’s the one thing you wish, every enterprise software company understood a little bit better to make them more effective in working with much larger organizations?
Bijoy: Look, I think the top of the pyramid is what is the biggest currency a large corporation has as opposed to a startup? The largest currency you have is trust. It’s something where you would say, “I have established over the last 40 years, 50 years trust with my customers.” Whether you are an insurance company or a healthcare company or bank, or you pick your customer, their number one thing is trust.
Now, if you are a startup company, you don’t have existing trust. You’re coming in cold so when you are navigating this conversation with a large enterprise company, be aware of what are you asking them to compromise on. If you know that you can work with it very well, what does trust mean? Trust means complying with rules and regulations. Trust means meeting customers’ expectations. Trust means fulfilling promises that you make. Trust means being accountable to governments in multiple geographies.
That I think is where the startups sometimes lose it and you and I have had these conversations in the past, because we operate in many different geographies, many different jurisdictions, and if you say, “This works in Silicon Valley,” doesn’t mean anything. It may or may not work in another geography for many number of reasons.
I think that’s the number when I would say disconnect between startups and large enterprises. Startups sometimes think, “Oh these guys are slow. These guys are not getting my vision. If only they would just buy into it and give me a chance,” but go into it thinking, I am asking them to play with trust.
Saam: That makes sense. It seems one of the common themes across our conversation is empathy. Being empathetic for the customer and really understanding their world and their problems. That is the nature of innovation. Both in startups and larger organizations.
Bijoy: You know this, if you’re in cyber security, for example, it’s not about technology at all. It’s really about trust.
Evan: I know only have like maybe 10 minutes, so one thing we like to do at the end of that episode is kind do a bit of a lightning round just to get a couple of quick hits. Maybe, Saam, we can transition to that and just want to get a couple of short answers. I know these are little bit meaty topics, but Saam, you want to start first?
Saam: How do you think companies should measure the success of a CIO?
Bijoy: Just to measure the success of any C-level executive. What is their values? What is their commitment to the business? How are they using their tool of the trade, whatever that trade is, to further the mission of the company? I would not measure CIOs any differently than I would measure a CFO or a CHRO.
Evan: Are there any common mistakes you see newly minted CIOs make that may be avoidable? Any advice you would give to someone stepping into that role for the first time?
Bijoy: Yes, so I think one of them is that they come into these roles sometimes thinking that they have to know the answer to every problem in that technology space, you don’t. Have the empathy and the humility to know that your team is going to teach you a lot and you cannot necessarily know the answer to everything.
Evan: Maybe to switch to a few personal questions. Bijoy, I’m curious, have you read a book that you really enjoyed recently, and what book was it and why?
Bijoy: I read a lot, because I spent quite a bit of time but the book that made the most profound effect on me, the recent time has been Sapiens
by Yuval Noah Harari. Absolutely, I mean follow-up books as well.
Evan: It’s a great book.
Bijoy: Yes, because he’s thinking about it very differently. I get a lot of inspiration from it.
Saam: Bijoy, maybe as we get to the end here, any other, just words of wisdom you want to share with, I think technology leaders that are trying to innovate and use technology to transform their businesses.
Bijoy: The only other additional point that I would talk about, which we have not explicitly talked about, is how important reverse mentoring is. If you’re a technology leader, spend a lot of time with the people coming into your teams and into your network who comes from very different backgrounds, from academia, straight out of school coming out of a startup company and learn from them. Because sometimes if you’ve been in these jobs for a long time, you get jaded by, “I tried this in 1996, it didn’t work, so it’s not going to work for the rest of my life.”
I get a tremendous amount of joy and personal satisfaction by spending time with the more early employees and how they teach me things that I had not thought of before. [laughs]
Evan: Bijoy that’s not surprising for me to hear you say, because I know you’ve always been very humble, very open-minded, curious, and I have to imagine that’s helped you become a great innovator in technology so I really appreciate you sharing your thoughts with us.
Bijoy: Thank you. I really enjoyed the conversation as well, and it’s been fun and my pleasure. [music]
Evan: That was Bijoy Sagar, Chief Information Officer and Chief Digital Officer at Bayer.
Saam: Thanks for listening to the Enterprise Software Innovators podcast. I’m Saam Motamedi, a general partner at Greylock Partners.
Evan: I’m Evan Reiser, the CEO, and founder of Abnormal Security. Please be sure to subscribe so you never miss an episode. You can find more great lessons from technology leaders and other enterprise software experts at enterprisesoftware.blog.
Saam: This show is produced by Luke Reiser and Josh Meer. See you next time.