Takeda’s CTO on Building Technology Roadmaps

By Kaitlin Milliken |  April 6, 2021

Additional Resources: 


[Sponsor Message] 

This episode of our podcast is sponsored by Cantina Consulting. Cantina crafts innovative experiences, products, and services for a connected world. They help organizations solve complex and transformational initiatives, making them indispensable to customers. Learn more at


Kaitlin Milliken: Hey, you’re listening to Innovation Answered, the podcast for corporate innovators. And this is a very special season where we take you inside the C-suite of big organizations. In each episode, we’ll interview a different C-level leader to talk about how they connect to innovation teams. I’m your host, Kaitlin Milliken from InnoLead. 

Today, we’ll sit down with Leo Barella, the Vice President and Chief Technology Officer of Takeda Pharmaceutical Company. The Tokyo-based organization is the largest pharmaceutical company in Asia. And — with over $30 billion in revenue in 2019 — Takeda Pharmaceutical is in the top 20 pharma companies in the world by that metric. In February of 2021, the company finalized a licensing agreement to develop, manufacture, and commercialize Novavax’s COVID-19 vaccine for use in Japan. Here in the US, Takeda has major sites in California, Illinois, Minnesota, and Massachusetts.

Leo, who is based in Cambridge, Massachusetts, joined Takeda Pharmaceutical two years ago. Since then, Leo has worked to build a digital technology blueprint and implementation process for new technology at the company. During our conversation, Leo shared how  he creates roadmaps for the future. He also discussed working with data and setting the right priorities for his team.  

We’ll take a deeper dive with Leo after this break. 


Kaitlin Milliken: At InnoLead, we know that your team is looking for support and funding from senior leadership. But that can be really challenging without the right data to show that you are making progress. To help you out, our team at InnoLead surveyed 196 innovation and R&D executives to gather best practices on metrics and reporting. That includes what teams measure, how frequently they share that information with higher-ups, and the financial and non-financial metrics that your peers are using most often. You can download “Developing Innovation Metrics & Reports That Really Matter” on our website at

This report is only available to InnoLead members. Not yet a member? Become a part of the IL network at Now back to the show. 


Kaitlin Milliken: And we’re back with Leo Barella, Vice President and Chief Technology Officer at Takeda Pharmaceutical. In his role, Leo is responsible for creating a blueprint for the company’s technology strategy. Prior to joining Takeda Pharmaceutical, Leo also held the CTO title at Blue Cross Blue Shield of Michigan and Excellus Blue Cross Blue Shield. 

So Leo, to kick us off, can you talk a little bit about how your career path led to this role at Takeda Pharmaceutical?

Leo Barella: Yeah, I’ve always been impressed by technology. I’ve always been fairly advanced in the adoption of new technologies. And so I kind of felt that even from a young age to get myself into a job that would actually keep me connected to technology. I started out as a programmer for AT&T in Italy. Once I moved to the United States, I started to focus more on systems integrations in assembling more enterprise level solutions. 

So obviously, I was actually part of projects that were defining technology solutions to solve business problems, right? I’ve always been focused on the simplification of the systems that were actually used by a specific process. That actually really led me into the systems architecture, which is truly the field that I enjoy the most even as part of my role today. Enterprise architecture is really not to be viewed as a specific IT function, but more of any enterprise practice that really defines how technology can and should be used to basically implement a more agile and more efficient and effective process by the use of technology. Understanding the strategy of an organization. Focusing on: What are the key value streams for that organization? So how is the organization successful? How can that success continue and actually improve? And then really connecting to the data that is necessary to promote success and ultimately, in creating this data and digital strategy to enable that. To me the CTO, the “T” in the CTO is actually more about transformation than it is about technology purely. Because now technology is pretty much actually everywhere. And it’s just a matter of actually applying good data practice to technology to improve business processes.

Kaitlin Milliken: I was looking into Takeda, and I saw that the company has committed to assisting with importing and distributing the COVID-19 vaccine. I wanted to see if your position or any of the technology that you’ve worked with has been a part in preparing for that rollout.

Leo Barella: The effect of COVID is, quite frankly, expressing the digital preparation of any enterprise. There are some enterprises that basically have been extremely challenged by the workforce actually moving to their homes, and enterprises like Takeda, they were extremely prepared. And frankly overnight, we actually went from having hundreds of offices to thousands of offices, because everyone was actually in their home. So from the perspective of preparedness, I think we were actually really thoughtful. We’ve already made a lot of investment to kind of renew our architecture to become a lot more digitally enabled.

The applications also have been re-engineered, and re-architected to be more global. So in that effort, I think that we have built enough capacity to be able to sustain this new demand from a perspective of producing vaccine for others. I can’t say that it’s very business as usual, because it isn’t. It’s not really usual. But we were definitely prepared, so we actually have the capacity to be able to sustain the demand.

Kaitlin Milliken: Great. And I know we’ve talked a lot, so far about the architecture for being digital, part of your role included creating a digital technology blueprint for the company. Can you explain what that means and what goes into creating it?

Leo Barella: Digital is one of these words that sometimes is actually overused. It’s quite confusing, because everyone is actually interpreting digital in different ways. But from my perspective, digital is really the automation of data collection from a business process, right? No matter what it is. It actually can be related to a customer, in our case, a patient, or it can be a manufacturing process, or it can be labs, leveraging instruments to actually collect information from an experiment. Digital blueprints actually really start from the mission and vision of the company, and the strategic plan of how we plan on transforming Takeda to become more of a data-centric, digital-enabled company, and leveraging digital technologies to enhance our products, right. 

So basically, our strategy starts with the philosophy of we’ve been making what I call the analog products, which are basically the pharmaceutical products, for centuries now. And now what I should, at the point in which we want to, enhance the excellent results, obviously, that the product produced for our patients with digital technology, which means that we’re trying to actually combine our product devices, and eventually software. 

The digital blueprint now needs to be associated with each element of the value stream. So we actually start with research. How do we actually instrument that research process with better ways to actually search for research documents, right? So basically, it’s working with a scientist and understanding: How do I actually make your role within Takeda, the enablement of this strategy, more effective, so you can find better information? You can find it quicker. It’s more accurate. You can collaborate with experts outside of Takeda. So for every position that basically is a key element of the value stream of Takeda — how do I enable that role to be more efficient and effective, right? 

So it’s related more to the process. And it’s related more to the people that are activating the process and enabling them with better information, better technology, elimination of any sort of redundant steps or duplicate entries or confusion about what data they can trust. And then eventually, the technology becomes the solution. 

Kaitlin Milliken: Can you talk about the skills that you need to actually become a data-first organization?

Leo Barella: I’ve been lucky to basically create about five global data strategies for five global organizations. And that is actually the fruit of actually being an architect, and recognizing the fact that if you actually focus on data and you optimize at the data level, then digital becomes extremely easy to do. And it is actually enabling more efficiency. Now, the data component is fundamental because most companies don’t quite focus on organizing their data. They have the ambition that digital will simplify things. And quite frankly, without a strong data strategy and data foundation, you end up actually generating and creating more complexity that then eventually you need to manage in addition to the complexity you already had. 

Now, the philosophy around data centricity is one that actually is focusing on data that you actually generate. Data that you actually need in process, the accuracy of information that you need, how frequently you need to actually have that information. One of the ambitions that we have in a data-centric world is to also make every process a real-time process. So the fact that you can actually use data, and you can actually use data that is fresh, right? So I refer to data as actually being a vegetable or fruit, right? The nutritional value of an apple or a pear that is being picked from a tree is extremely high, as soon as you actually pick it. But if you actually leave it there a couple of weeks, now all of a sudden, it’s not as nutritious right? And so the same thing for data. So the more you can actually make it process react in real time, the better is the overall value of you know, in your organization.

Kaitlin Milliken: And when it comes to managing data and keeping track of all of the information that’s coming in, how does your team approach that?

Leo Barella: In order to have a good data program, you must actually have executive sponsorship. Our CEO, and our executive team has embraced this new philosophy of this data strategy, where we now actually have a team for each one of the executive teams at Takeda. That is represented by an individual that makes up what we actually call the Global Data Council. Now we actually have representation of the entire data community at Takeda. And what we try to do is now act as one, and actually looking at data as an asset that should be shared, not only inside of the data, but also externally. 

So we have identified that there were a lot of duplications of requests for data. There were a lot of data sources or data that we actually even purchased two, three, four times from the same vendor for different purposes. So we now actually are organizing everything around a data catalog. The catalog is just a technology. It’s a catalyst for us to actually have a discussion around, “How do we best actually make use of data? How do we simplify our access to data?” We don’t actually look at the data program as built for humans. We look at the data program as built for automation, machine learning, and AI, where basically the human is actually being leveraged to build models, but not necessarily to interpret the data. As in, we’re seeing interpreting the data today, but in five to 10 years, basically once the model is actually built, the beauty of AI and machine learning is the fact that you can actually move a lot faster because the humans are training the models that basically would produce a lot more speed. And again, real-time ability to react to a process.

Kaitlin Milliken: Right. So the humans that are building these models, the folks working on those platforms, are they people that you’ve trained at Takeda? Are you hiring people into the company?

Leo Barella: It’s both. We actually have an inherent familiarity with obviously the proprietary aspects of the products that we make that are known by our employees. We also have more than 210 different partnerships with external organizations that basically are helping us not only develop our products, but also help us with the analysis of data, right? We also have a partnership with MIT, that is actually developing 10 plus different truly AI capabilities, right? So AI is another one of these terms that is being misused by many. It takes a lot of data, a lot of data, to be able to actually get to a model, both machine learning and AI, that people can trust. The only way for us to actually be able to truly manage an AI algorithm is through partnerships. There is really no way that Takeda will ever be able to actually generate enough data to actually make a model accurate enough to be trusted. The only way that we can actually react to that is either we actually acquire data from even our competitors in a way or we actually leverage algorithms that have been developed in partnership with other pharmas to then eventually get some traction relative to inside what we’re actually trying to generate.

Kaitlin Milliken: When you say using data and leveraging data from competitors, can you give a little bit more on that?

Leo Barella: Ultimately, our industry, it is actually more of the pharma industry, is built to save lives, or improve lives — in some cases improve, in some cases, actually save lives. And so when you’re dealing with a human life, the competition should not really be bringing a product to market. You’re talking about a very specific, you know, classified almost type of information because obviously we don’t want to actually necessarily share individual information, but if there is a resource that can actually be shared with others, where even our computation can actually build products based on information that was actually gathered by other individuals, right? When they’ve actually donated their information for “the science.” It’s one of the principles where basically we just want to make sure that collaboration is indeed an enabler for the development of our products, as well as products that come from other companies. 

We also have our COVID Alliance, which is actually an alliance that we actually have with other pharma companies that are leveraging plasma as a resource to develop their own products. Takeda is trying to really collaborate, specifically, around data to develop new science.

Kaitlin Milliken: So I know that we’ve touched on things like AI and machine learning, I wanted to focus a little bit more on technology. How does your team set technology priorities and decide what you want to explore?

Leo Barella: So we have two governance processes within architecture. The first process is actually called the Architecture and Innovation Council, which is basically broken up into two mega-domains. One is a technology domain that is focused on what we actually call the “information integration, application, hosting, or infrastructure technologies.” And then security. Within each domain, we actually have experts across the enterprise that collaborate on the definition of a roadmap, the second component is actually more related to the business. So now, for each one of our business units, we actually have a similar council or domain that is actually focused on research, development, manufacturing, supply chain, marketing, sales — each one of these domains actually has its own distinct roadmap.  

And what we have done is we’ve broken up our technology lifecycle into what’s actually known as the time model, which is a Gartner model, which is the tolerate, invest, migrate, and eliminate — where invest is actually the technology you want to always use in your solution; migrate, you’re now actually transitioning out; tolerate is where you have the technology, but you don’t want to necessarily continue to adopt it.  

Now, the other two stages that we actually have added to the time model, is what we actually call the proof and the watch. So basically, in watch is actually when this team of experts, for instance in the research organization, are actually looking around the market. They’re going to conferences that are actually looking at new technologies. They don’t actually have yet an understanding as to how this technology can be applicable to Takeda. But then the next step of the process from watch becomes prove. And this is actually already where we apply the technology for proof of concepts. 

So now all of a sudden, you have a match between the need of the business trying to improve a process and the match of the technology that we’ve been watching for. And then we actually start to actually run what we actually call the proof of concept, right? Where proof of concept is not related to technology. But we try to actually solve a business problem leveraging technology, getting feedback from our customers, and then deciding if this technology is in conflict with one of our standards, or one of our invest technologies. Or as we basically form groups of experts across the enterprise that basically make decisions on the roadmap for specific areas of the business and also specific areas of technology… It’s not the same people, but they actually have full authority to the definition of the roadmap. And then once they actually make a decision, then basically we try to leverage, then invest technologies in all of these domains for every solution that we develop.

Kaitlin Milliken: And how do you keep those different groups connected? Because I’m sure that in some cases, the technology and one group could be applied successfully elsewhere.

Leo Barella: The leadership of all of these domains is actually coming from our enterprise architecture team. So within the EA group, we actually have stewards, right? That basically are the host of this discussion. They obviously form a circle within the enterprise architecture team, so they can talk to each other about the integrated enterprise roadmap, right. So that’s actually our integrated target state architecture is mostly referred to, but then they actually have monthly meetings with the experts. And basically, the experts actually have a representation. You know, we actually use, I don’t know if it matters, Microsoft Teams to make sure that these teams can actually collaborate interactively, asynchronously. So we’re basically if there is actually a group that is actually challenged with a problem basically you have the group of experts that can come together quickly, discuss the opportunity, discuss what solution we want to implement, and then and then continue on. 

Now, the other aspect of governance is actually the architectural review board as it’s probably known more industry wide, which is actually making sure that for solutions that we develop, we actually already follow the standards that basically were created and the roadmaps. They were created by this Architecture Innovation Council.

Kaitlin Milliken: So I have two final questions, which are very advice driven. Do you have any advice for innovators who are looking to work more collaboratively with their CTOs?

Leo Barella: So first of all, I don’t believe that innovation can or should be represented by an individual, even like the Chief Technology Officer, right? So it’s not, it’s not really a role of decision making is more of a rule of, of stewardship and collaboration, right? Same for innovation, innovation actually has to be embedded within the culture of an organization. And so to me, the Chief Innovation Officer or people that are actually working on innovation, just need to understand that there is a huge potential writing in sharing information and sharing experiences, so that we can all grow as an organization, rather than actually being a siloed approach.  

My role is to be a great equalizer, and making sure that I can actually broadcast and inform and communicate the ability that we have in leveraging technologies to to basically create automated outcomes, right. And if you’re actually looking outside of Takeda, I lead by the philosophy that again, we should really not compete on technology, necessarily, at the end of the day. The product is actually what matters. And if we can actually evolve the industry, where basically the entire industry can actually move faster, with more information, better information, then ultimately, we’re benefiting our patients.

Kaitlin Milliken: So are there any pieces of advice for teams that may be working on going digital working with data that you would like to share?

Leo Barella: So data is where you need to start. If you optimize your processes based on your data flow, you actually already start to monitor how data is actually flowing across your organization, across your process, and identify bottlenecks and where data is actually causing conflicts, right? Then the rest of it is easy. 

So I keep actually referring to the fact that there is a lot of data without digital, but there is absolutely no digital without data. So that to me is the concept that, focus on data and build your digital program after data. But if you actually started with digital irrespectively of your data or you don’t even speak about data when you’re speaking about a digital program, then you’re generating more complexity than inefficiencies. 

Kaitlin Milliken: Perfect. I think that’s a great note to end on. Thank you so much, Leo, for your time.

Leo Barella: Thank you. 


Kaitlin Milliken: You’ve been listening to Innovation Answered. This episode was written and edited by me, Kaitlin Milliken, special thanks to Leo Barella for sharing his insights. For more tips and tricks from inside the C-suite, be sure to subscribe to this special season of Innovation Answered. You can find our show wherever you get your podcasts and on our website, Thanks so much for listening, and we’ll see you soon.

Special thanks to our sponsor, Cantina Consulting. Cantina crafts innovative experiences, products, and services for a connected world. They help organizations solve complex and transformational initiatives, making them indispensable to customers. Cantina is made up of a community of strategists, designers, and engineers who rise to the needs of their clients. No challenge is too small or too large — whether it be building a digital business for one of the country’s largest retirement services companies, or making a connected dog collar which became one of Oprah’s Favorite Things. Their team has also helped a major movie studio reopen after COVID-19 shut it down, and designed a new approach with AR/VR to train fighter pilots. Learn more at