Let's build the future - Conversation with Hiran Bhadra, Senior VP of Strategy at Belden

Hey, thanks Ken, nice to be here. And nice to have you and thank you Doug for recommending Hiran. I think this will be a really good conversation on a very interesting company from my perspective, of course, and that's Belden. You know, we call this the Digital Thread podcast, so we always like to start off really asking about what would you consider to be your digital thread, in other words the thematic threads that define your digital industry journey.

Yeah, excellent question, Doug. I think the first thing I need to feel grateful about is each of these organizations. You know, the so-called famous orange ball — Arthur Andersen in the early 2000s, then KPMG, and then Accenture, of course — these are fantastic platforms for learning. Although Arthur Andersen's legacy was with audit and tax, they had huge emphasis built on business consulting. KPMG similarly had huge emphasis on audit and tax, but they had huge emphasis on business consulting. And none of the leaders in these organizations kind of allowed the focus on people with curiosity on what is going on in the core business area; they allowed a lot of people to learn and embrace and build advisory practice. So these platforms, you know, when I started working with them, the advisory business often was less than 25% of the overall portfolio, but they were the highest growing businesses. A lot of funding went into it to bring people like us with engineering background and using the relationships they had to solve critical business problems. So I'm very grateful that they all focused on creating businesses that were adjusting to what was called Accenture. And I still tell my colleagues today, and I still have Accenture as a partner within my company — if you really want to learn technology, that is one of the finest organizations on the planet. And I'm very grateful for that experience because it completely opened my mind as to the whole possibility of technology. You know, these organizations also make you learn looking at the problem from what does it mean for the customer. And I think each of these organizations enshrined it from day one: you can have a great idea, you can have a great technology in play, you can have a fantastic story around how does it influence experience, but at the end of the day, is it creating value for the person or the customer that you have in mind? And being very, very value-driven about how we went about doing consulting — I will say that is a common thread that stood across all these three organizations. And finally, these are also organizations that embraced partnership. They all quickly learned that the complexity of the problems that are out there to be solved and the kind of trust that customers were placing on them needed strong partnership. So Doug, even if I think of your organization, it kind of is enshrined in that right? I've quickly realized that the kind of capabilities we need will never be always residing within the four walls of where I am, and that creates some openness to go and work with partners like you.

Let me break this into two parts: there's an internal engagement and there's an external engagement. The internal engagement is, you know, when I think back why I joined Accenture, besides the fact that pivoting to technology was a very conscious choice back then — I'm talking about the 2016, 2017 period of time — the concept of innovation in action, the concept of customer innovation centers where you conceptualize creating a digital replica or even a hybrid, a digital and a physical asset replica in an innovation environment and bringing customers to it and having a dialogue on what's the art of the possible — I thought that was a phenomenal internal engagement that I appreciate. I mean, think of it: I've been part of these design thinking sessions where the CxO of very large manufacturing organizations, like $50 billion plus sizes, have come in, spent two days sitting through in the middle of these innovation centers where they are looking at ... the concept of digital twin was still early then, but the thought process was very similar, saying if we have a data replica of what we do day in and day out, and we can have technology interrogate that data replica and we can get some insight, that'll be so powerful. So I will say that whole innovation experience that Accenture had established was extremely helpful. And then, of course, you then rotate and there are these specific engagements in which you apply yourself. The ones that clearly come on top of my mind, and given my background, which is around asset performance management as a topic — manufacturing globally and capital productivity as an issue is an extremely important topic for the discrete industry, for the process industry, for the oil and gas industry. And as much as it might sound a little contrary to what might be the popular perception, there are significant opportunities for improvement in capital productivity even in North America. So as a team, we just dived into that problem. And, you know, whether it's in paper, cement, chemicals, steel, mines — the kind of categories of industries we worked in — the issue is how do you use data to improve capital productivity of the huge amount of trillions of dollars of assets that were deployed? And when I look back, I can proudly say that often we think of overall equipment effectiveness as a metric. And there are many examples in each of these industries that we could move the needle by a few percentage points in how the assets became more utilized — whether it's a shovel on the mine floor or it's a truck utilization, travel in between distribution plans and the cement manufacturing sites, or the kiln in a cement company. All these real examples where we could apply that knowledge and move the needle. And from a business performance perspective, these mean hundreds of millions of dollars if you can make more chemical or more cement or guarantee quality, or you can utilize these fixed assets. You're not only creating a revenue base for the company which is significantly higher, of course you're contributing to the bottom line, but more importantly, you're making the company capital efficient because as they grow now, they're using lesser and lesser capital for the same amount of outcome. In our consulting language, we used to call it finding the hidden plant. And I think quite I can look back and proudly say we could find that hidden plant in many, many instances.

So what attracted me was the relative size of the organization. And as you know, Belden is a sub-5 billion enterprise, which actually is a strength in terms of the speed with which you can execute. But more importantly, if you really think through the company's core and you think of the opportunity out there, the potential of value creation within Belden is just hugely significant. It's just astronomic. So in pure simple terms, you know, this company, if the world has, let's take the GDP in about 60, 70, 80 trillion dollars, Belden's contribution to GDP — given its core — the potential in the next five, six, seven years is to be in multiples. And we will see what the multiple finally figures out. So it's essentially a company with a very strong core, but what I was really fascinated by was, one, the leadership commitment and understanding that there was an opportunity to pivot the business model and create more value globally and for itself, of course. And more importantly, and this is where right, and there was a clear acknowledgement that this is probably the kind of pivot that needed external talent. And as they went through their own selection process, I think what drew me to them was essentially this value creation potential, but what drew them to me was my background in the interface of technology, including networking and 5G work that I've done as part of the whole digital transformation, as well as the fact that this transformation from being product oriented to both product and solution oriented has also been part of the work that I've done in the past. So that's created a very strong marriage, and I think that hypothesis has worked out very, very well.

Sure. I mean, since the early stages of any work that I've done, back in 2008 that we can even call remotely digital, and especially in the context of a manufacturing environment and even in a smart building environment, the value of the network was so ubiquitous that people used to think that, hey, what is a data need? Of course it's available. All I need is to create pipes that will essentially transfer the data from the point it's getting generated to the point where I need it to be. And hence there was a huge rush of investment into cloud and into analytics, and this whole world of big data started. The need for data scientists exploded, the need for data engineers exploded, and people invested billions and billions of dollars in platforms with the simple assumption that with relative ease we can make data move, and as long as we could find the data to reach its targeted destination, some magic models will crunch it and we're going to create insights, and the world would be a far more productive place. Guess what? 15 years or more since then, it hasn't happened. And when people try to diagnose as to why is it so — our quality of analytics delivery has improved, the amount of computing power in cloud platforms has improved — why are models and change management struggling? You really go into the core and you realize that the assumption has always been that if a manufacturing process has a physical self in which things are working, our current networks, our current technology infrastructure was built to essentially imitate that physical world into a digital world in a cloud environment, but it was not true. Which means that, and I use this term — it's a little technical, but I think it's also self-intuitive — the physics of the manufacturing process, if you have to imitate it well in the digital world, the current technology infrastructure needs significant upgrading. And there is complexity. For example, we have historians, we have MES systems which often sit between the sensor and the cloud. One of the first things the MES does, of course for obvious the right reasons, it averages data. So a signal can get generated in a few milliseconds, if not less, but systems that are sitting in the middle average them, they smooth them, and when they transfer that data into a cloud platform, guess what has happened? The actual physics of the process has got lost. Now one might say that why can't analytics work on smoothed data? And this is where people have realized the reason it cannot be fully successful is because the insights that need you to manage the physics of the process lies actually in the jitters of the data that got smoothed. So it's just not important to know what was in general the average temperature of a cement kiln or a chemical plant or a boiler; it's important to know the fluctuations of that temperature around the average line which creates insights. The moment this was realized, that we now need to create infrastructure that will essentially reflect the true physics of the manufacturing process in a computing platform, two things happened. One, network became the most important technology infrastructure for success of digital transformation. You need a pipe that will transfer high-frequency sensing data directly, safely, securely, and in an accurate manner. You cannot have a pipe that either cannot give you safety of IT and OT integration; you cannot have a pipe that averages physics; you cannot have a pipe that is not reliable, because the analytics platform needs reliability, it needs security, it needs clean data. It became important. And as IoT proliferated, it also became obvious that you need bigger and bigger pipes. And the concept of cloud computing became non-sustainable because bigger pipes mean more money, it means more technology infrastructure cost. So now networks have become the only solution to provide clean, secure, and harmonized data. It has also become the only logical destination for edge computing because you cannot take all data at the highest frequency reliably to a remote cloud. So these two confluence, this understanding that the role of the networks to provide data of this targeted quality and consistency and security, and the fact that we need to operate at the edge, has completely blown the lid of how people have started perceiving networks. And it's now concluded that you cannot have a digital twin successfully running in multiple sites of an industrial platform or multiple buildings in a smart infrastructure unless you think through the networking infrastructure that binds it.

Yeah, that's a very important question. You know, at the highest level, we have broken the problem statement into what is needed in a brown field setting and what is needed in a green field setting. The point that we are taking to the customers in the brownfield setting is that, let's say you are convinced about this whole aspect of transferring the physics of the manufacturing process to the point of compute. We have to come out with a solution to address the complexity of their current OT or operations technology environment, because I cannot go and say that my solution is you completely strip out everything you've got and replace it with new infrastructure. So the company is now focused on saying, for brownfield customers, multiple site where solutions need to be deployed at scale, how can Belden partner and show that using software, using the intelligent products that we have got in the world of industrial automation, and the fact that we completely are in the full portfolio of the path of the data from the signal to the sensor to the cloud, how can our product portfolio, our product capabilities around industrial automation plus software, provide them with multiple points of data generation, give them clean, secure, harmonized data without them needing to make any changes in their OT environment? Or let's say as much as possible, no changes in the OT environment is the proposition. So they don't feel that they have a technology debt that they now need to replace again with some billions of dollars. So we work through the technology complexity to provide the data and the network reliability, so then they can focus on applications and end-user adoption. Which at a technical level means that we have to now become a platform organization, which means we need to be able to say, irrespective of the pipe the signal takes to go from its points of generation to edge or a remote cloud infrastructure, we will provide you the platform that will orchestrate that process, that will deal with the underlying OT complexity that you the process. So which means that as a company, the journey we have now initiated is not only going from being product to a solution organization, it's also about becoming a platform organization that will orchestrate the products, the software, and the services that come along with it.

Yeah, again it goes back to this whole spirit that as you think of the full — I often call it data supply chain, data hydration. A typical manufacturing floor is projected by the leading investment bankers to have 20 times more sensing than what exists today. Even if you think it's going to be 10 times, people are using different kinds of capability to start measuring something: 'Oh, we didn't measure weather earlier, let's now start measuring weather. Oh, here is a very critical machine in the manufacturing plan, we were not calculating the ambient temperature earlier, let's start calculating it or measuring it right now.' The sensing technologies are a complete proliferation of different kinds of protocols. Our manufacturing technology systems are multiple industrial systems out there; you walk into any manufacturing floor, it will at least have three or four, if not six different industrial systems. Litmus is about dealing with that complexity and trying to harmonize data which comes from many, many different protocols. And how do we harmonize it? Which means if people on the manufacturing floor from a signal transmission technology are speaking Spanish, English, German, Hindi, Chinese, how do you harmonize them into a language, into all of them becoming English as an example, and making it available? And of course, I'm presenting in a very simplistic way, but we essentially dipped into Litmus's legacy of having worked on it over many, many years. And what it helped us was essentially bring an extremely important capability and acceleration into our own R&D process. Because if you go back to my previous answer, we don't want people to change the number of PLC versions they have or the number of PLC manufacturers they have. We are not the company saying please change all your PLCs to Allen Bradley or change all your MES systems to Aveva. We are saying that given the technology OT environment you've got, we will provide the harmonization of that complexity. And Litmus plays a very important role in addressing that capability.

Absolutely. I mean, if you think back of the earlier part of the conversation, the two important highlights from my experience were: remain completely focused on creating customer value, and the second is know that any one organization cannot have all the answers to what is required in an end-to-end integrated solution. It's the same concept, Doug. At the end of the day, we are trying to create more value in investments in digital transformation. And when we use terms like sensor to cloud, we have to decide what our core is and where do we need to partner. We have as a strategy, from an organization standpoint, we are very clear: we are the network and data infrastructure company. While any requirements on sensing, any requirements on cloud operations, we will create partnership. Similarly, even within the network and data infrastructure, we have acknowledged that technologies like 5G are very critical for success in a hybrid environment where our specialization in wireless and wireless over Ethernet and Ethernet capabilities need to talk and integrate with 5G capabilities, which is again a portion of the whole infrastructure where we have decided to pursue ecosystem partnerships. So it's essentially taking the solution view to the customer, unpacking that into all the Lego blocks that are required to work in an integrated fashion, and having a very conscious strategy of which part of the Lego blocks will come from our factories and our balance sheet, and which are the other Lego blocks which are important for the customer solution but they are not part of our core, but we need to be able to integrate — or have the capacity to integrate — so that the customer value is not degraded. So ecosystem partnership is an extremely important part of our execution plan.

It sounds cliche, but often the inspiration you draw from kind of often changes a little over time. I remember I was enthralled by books like 1984 by George Orwell or To Kill a Mockingbird. But if I have to really think of a book and inspiration, the inspiration-wise, like many Indians, Gandhi is in our genetic makeup. It's about patience, it's about remaining focused on the truth, and not let your resolve be impacted by a lot of noise that may be around you. And the Gandhian principles were enshrined in us within my family as I was growing up back in India. So the Gandhian spirit is very much what enshrines me. But recently, I'm almost kind of finished reading a book which is Oppenheimer, and the movie of course has become very popular. But when you read the book Oppenheimer, it just shows how you see the power of technology in the physical world that operates, but then as we are embracing AI and making it an integral part of our platform development, how do we execute technology in a responsible manner? I think it's one of the critical tasks for a CTO of any organization going forward. How do you ensure that what we are building and its intersection with the human aspect of what we are building, I think, is what will set aside organizations that will be called also-rans or organizations that will be seen as truly innovative — people who could bring the technology and the human factor in a very, very balanced manner. And that's why I thought I'll refer to Oppenheimer, not only because of recency because it just brings to sharp focus how that balance is so critical for the world we live in.

Oh, thank you, Ken. And on behalf of Belden and even at a personal level, I'm very grateful and appreciate the opportunity to both you and Doug. Thank you so much.