Startups vs. Corporates: Innovating AI in the High-Stakes World of Healthcare

Jonathan Morgan [00:00:00]:

Welcome to the Strategy Gap, a show for leaders who want to bridge the gap between strategy creation and strategy execution. I'm Jonathan Morgan and along with Joe Krause, each episode we have conversations with strategy and operational leaders on key issues in strategic planning and strategy execution. If you're looking to actually accomplish your goals, this is the show for you. Welcome in everybody to another episode of the Strategy Gap. Joining us today for our conversation is David McMullin.

Jonathan Morgan [00:00:31]:

David has more than two decades experience in corporate development, strategic planning, commercialization and operations across multiple sectors in the life sciences industry. In addition to serving currently as Chief Business Officer of Anumana, he's also the Chief Strategy Officer at nference. Prior to joining Anumana and nference, he served as a Chief Business Officer of Aldeyra Therapeutics, a NASDAQ listed biotech company. He ran U.S. internal medicine and was the head of global commercial operations at Shire, a $1.4 billion rare disease and specialty pharmaceutical company. And prior to his work at Shire, he was the Vice President of Global supply chain Management and Strategy at Novartis. He holds an MBA with high distinguished, have high distinction as a Baker Scholar from Harvard University and a Bachelor's degree in Economics from Brigham Young University. David, welcome to the show.

David McMullin [00:01:22]:

Thanks. Thanks, Jonathan. It's great to be here with you and Joe.

Jonathan Morgan [00:01:25]:

Yeah. So before we get into the heart of the conversation and we'll talk about a lot of great things around strategy creation, execution, I wanted to give you a bit more of a chance to talk a bit about your background and specifically Anumana. So will you mind giving our listeners a quick overview of Anumana and what you're doing there?

David McMullin [00:01:42]:

Anumana is a software as a medical device company. It's a startup. It's now three and a half years old. Its genesis story is founded on this premise that we can use AI to make standard of care tests smarter and do so in a way that can actually transform clinical care of patients. So we're focused on cardiology and we use, we go and we look at like what are the standard tests that are being used in cardiology today and where are those tests providing information that with the rich data that's available at a medical center, could be improved using AI technique. So we've started our journey with using the electrocardiogram or the ECG as the backbone of the algorithms we build and then to turn them into tools that a clinician can use, we have to create medical grade software that houses those algorithms. So that's the journey that we're on. I said we started with electrocardiograms.

David McMullin [00:02:38]:

We're also doing algorithms that use other modalities in the cardiology space, even into the interventional setting. So in a nutshell, that's what we're doing here at Anumana. We're just now a series B company, We've got two FDA clearances, are just starting our commercialization journey as we build out the pipeline.

Jonathan Morgan [00:03:00]:

Very cool. I think it's certainly an interesting intersection of a lot of different aspects. You've got medicine, you've got your role, that's some aspect of strategy. Then you've got a big hot topic, of course today is AI. So how do you think about the intersection of AI and medicine? Obviously there's, I think, a lot of good things that are coming out of AI. There's a lot of negative aspects about AI and probably there's a little bit of concern for most people that think about how does AI incorporate itself into the medical practice? How do you think about that intersection between the two?

David McMullin [00:03:29]:

We all know that like AI right now is almost like become a buzzword where everyone's talking about it. That has. It's a good side as well as like a downside. The good side is a lot of people are generally aware and even experiencing or aware when they more aware when they experience AI in their daily life. The downside is there is a lot of fluff out there. There's a lot that's said about AI, probably by people who aren't really in a position to talk about it. And that's very important when you come to healthcare. Because with healthcare, obviously we're dealing with the health and well being of patients.

David McMullin [00:04:02]:

And so the technology needs to be medical grade. I think fundamentally there's some very practical things that are being done already to ensure that AI, when it's applied into the healthcare space, is done the right way. And one is like with our technology, it's defined as a medical device category. Now the FDA has issued guidance on AI software as a medical device. The regulations are being written. So like for our technology, the first product to get cleared required what's called de novo. So the FDA actually wrote a new section in the code of federal regulation that specified that for this technology to get on the market, it needs to be clinically validated. Some of the worries about AI have to do with things like a bias in the algorithms that you create.

David McMullin [00:04:49]:

However, there's ways that we can get around that. The randomized clinical trial or with AI software. You also have the benefit of being able to leverage retrospective data to Demonstrate that what you build in a test environment can be generalized and that if there is bias, you can uncover it and determine it appropriately. That's probably the biggest thing that can be done in healthcare for medical device related AI. I won't speak to like business process AI, but the medical device related AI, it should be treated in a similar fashion as any medical device or intervention technology. You need evidence to demonstrate that it's efficacious and you need evidence to demonstrate that you've checked all the risks, understand what the risk factors may be when the product's utilization.

Joe Krause [00:05:38]:

It's super interesting because I mean a lot of the topics we talk about here is when we're creating strategy for the future, business strategy, what have you, it's what time horizon should we be looking at? And sometimes some organizations we work with are more stable and there's less change and they're, they're looking three, five years out with the role that you have and the, in the business that you're running, I mean ultimately we're talking things can change on a weekly monthly basis. So how have you managed to be able to have a long term vision at the same time managing the fact that that vision and that direction you're going could potentially move and shift more rapidly than other industries? How have you managed to keep the North Star in sight?

David McMullin [00:06:12]:

That's a great question, Joe. And it really is a part of the reality. I mean, we are at Anumana, it's a startup. So we're building an organization where we need to have that long term vision and we do have that. We have a picture of what we think the future will look like with our technology and it will look different than today. And then we break it down into steps in order to make that change, we have to roll out our technology. We have to do things like change the way products are coded in the marketplace. We have to do things like eventually change the way guidelines are written to associate the proper use of our technology.

David McMullin [00:06:48]:

That long term vision is important because it helps you identify and prioritize the work you need to do today. But because we're a startup, we also need to have quick wins and, and so we need to find within that path the discrete milestones that we can hit in a discrete timeline. Because generally when you go to investors, they want to see that you have a vision and they want to see the evidence to believe in that vision. And that can be like five years plus. But they also want to know that if they're investing in you, that you're going to translate that into some de risking milestones or value generation milestones that directly bring in cash in the next year or two. And so it's constantly blending those things. But we, I would say what's different here than when I worked at larger organizations, we don't really have like just separate strategy sessions and like budgeting sessions. It's sort of like all part of the same thing.

David McMullin [00:07:42]:

And we're constantly kind of adjusting as we go. And because we're a small organization, we can have a flexible, nimble process like that and it's easier to align a smaller group of people and adjust in real time.

Jonathan Morgan [00:07:55]:

Yeah. And certainly there's a lot to dig in around kind of that startup space and as you grow and that comparison from your career from a larger organization to a startup organization. And I think for the remainder of the conversation, we really want to dig into how you think about those selections and how you scale and grow and balance innovation and balance scaling aspects and doing those constructively together. But starting from that beginning, one thing we see a lot of organizations struggle with is the typical way to talk about it is there's a red ocean strategy and blue ocean strategy and which space you want to compete in. And I think even outside of that metaphor, as a startup, you have so many different things that you can do. Right. And in the context of your organization, within AI or within healthcare, there's so many things that you could have placed a bet on as, hey, this is where we're going to start. So whether you're thinking about it from your experience at Anumana or just from a general business perspective as a whole, how do you begin to think about how to make these strategic selections of how we're going to begin or how we're going to take that next step? When you have this world of opportunity in front of you from a business standpoint.

David McMullin [00:08:56]:

I mean, in my experience at large companies, I had the opportunity to work at like very high performing multinational corporations. And there I think I saw examples of what's best in class in terms of strategic planning. It's basically like one, you have a defined strategic planning process that you actually invest time in. You collect information, new information from outside the organization to inform what you want to do. You design it in a way where you wrap that up prior to the next year's budgeting cycle. You do it annually and one like follows the other. Now in my prior roles when I was at these larger organizations, it was a significant investment of the CEO's time and also the other executives as well as staff who are supporting that process and collecting the new information doing advisory board through the different sources that were used to inform the strategy. Now you need that same sort of thing at a startup, but you don't have the infrastructure or the resources.

David McMullin [00:09:56]:

So what you do, what I have found, or at least one of one lessons in kind of this latest chapter of my career of doing startups is you can take the semblance of those kind of best in class processes and you can boil them down and get let's say 80% of the value out of it in a way that is very doable with at very small scale. You can do it by yourself if you need to. Or if you have like a small team, you can do it with three people, five people. Because the value really comes by carving out time to think and apply your thinking. Right. And so yeah, you can use different frameworks like red ocean or blue ocean or when I was in business school, some of my favorite strategy professors were Michael Porter or Clayton Christensen. Various people have come up with like various ways of thinking about business problems. In my experience, there's always like a mix of different types of frameworks that you need to find to apply to the situation that you're dealing with now.

David McMullin [00:10:52]:

And like, even in the organization of Anumana where we literally every minute is used, we have carved out time to do a prioritization process and we used external advisors. We didn't pretend like we would be able to see all the gaps in our thinking ourselves, so we involved external advisors. But we did it with like three key people because we, we couldn't involve too many people because we couldn't allow it to distract. But we did align, you know, the broader organization. And it's important to take this vision of the company and translate it into something that's tangible and understandable to every employee in the company. And basically that boils down to what are we going to do with our time in the next quarter, in the next year? And do I understand it well enough as an employee where I can associate what I'm required to do for my job description with what we're trying to accomplish as a company that I think is critically important as part of that process.

Joe Krause [00:11:55]:

You bring up an interesting topic that comes up once in a while on the podcast, which I live in the Morristown area, so if I squint, I can see a couple multinational pharmaceutical companies. I started my career at one as well and I have the same idea as you where I learned a lot in the time that I was There seven or so years at Merck specifically, a lot of what I learned I was able to, when coming to achieve it in startup mode, distill it down and saying, oh, their training program was great. We don't have the, the resources to replicate that here, but there are elements I wish to replicate. And so from your perspective, I always ask the question, it's like, was it best to start at a larger organization and use those learnings to now apply to what you're doing now, or would it have been better to be inverted where you started small and went big? How do you think, based on your experience, if you had a magical time loop here, how do you think that the big organizational learnings have helped you? Or would you rather it be the other way around if you could do it again?

David McMullin [00:12:47]:

That's a really good question. And of course, like, with my background, I might be biased and say, like, start big and go small. But actually I think the answer is it depends on the stage of the company. But I do believe very early on you benefit by having a mix of both in the organization. Right. So. And I do think at very, very early stages, at least, this is now a fifth startup. And all of the startups I have been part of, they have had founders that included individuals that were doing this idea like, very early on in their career.

David McMullin [00:13:17]:

Right. And sometimes they were paired with someone else who had more operational experience. And that very early stage founding, that's where I think just striking out of business school and trying to chase an idea, sometimes that's what's needed to take an innovation and put some meat on the bones, if you will, like, start to create something real out of something that's conceptual. It doesn't take long when you're raising money. That operational experience at some level is going to be demanded by the investors. I've never met an investor who just will give out money without having some assurance of what's going to happen with that money. And we tend to index on prior experience as being a good way to measure some level of expectation of future performance. Sometimes you can build that experience yourself.

David McMullin [00:14:03]:

So you raise like three, $50,000 and then you go to whatever, $200,000, and then maybe you go to $2 million, and maybe you can do that yourself because you're showing along the way that you can deploy those resources effectively and basically build your experience as you go. For Anumana, we were at a different scale where we initially raised $20 million. Right. And that was our first full round. We had seed before that. And then the next one we did 60. So we start to get to that skill. At least in my experience, I have found that you need a blend of people who have operational experience, who've managed budgets and organizations in the tens of millions of dollars and have an idea of how to deploy it and how to make the trade offs of accomplishing what you need to accomplish.

David McMullin [00:14:49]:

While especially in my space, ensuring that all the quality and compliance things are built up to have a really robust organization.

Jonathan Morgan [00:14:57]:

Yeah, I think naturally when the conversation comes up around startup versus larger organization, the topic that many people are thinking about is kind of the innovation aspect of a small organization, the agility, the ability to create new ideas and move quickly compared to a large scale of an organization that has systems and process, has scaled up to that point and has that experience. And if I were to take our previous, my previous question around picking those ideas, if we were to fast forward into the future and assume that yes, we picked the right ideas, naturally that next step for a startup organization is okay, how do I scale? And better yet, how do I scale while still balancing that innovation? That got us to the point that we're at where we are now. So how have you thought about that throughout your career? Natural conflict point between scaling and innovation and making sure that you have a strong balance between the two of those here.

David McMullin [00:15:49]:

I think I first started thinking about it while I was at large organizations actually, and that was because there was a dearth of innovation and a large organization had so many processes which were able at sort of incrementally improving the business and growing the business, but not really designed to come up with breakthroughs. And when I was in a large organization, and by the way, like in full disclosure, a lot of my thinking was shaped by the teachings of Clayton Christensen at Harvard Business School. He had this very popular class calling building a sustainably successful enterprise and he was big on training the business students to when you approach a business problem, do your best to be explicit about the hypotheses that you're founding your decisions on, what makes you think that decision will work. The reason he would try to teach people to think that way so then you'll be more likely to see your potential blind spots instead of just sort of going by instinct and thinking that you've got it covered in a large organization. What that led me to is, and I did this several times at several different companies. One I would show when I was in a role of significant responsibility that I could hit my objectives and I could get done what needs to be done for Sustaining the organization and growing it. But then advocate that there's breakthrough innovation that's, that's an opportunity that's being left on the table. And in order to go after that breakthrough type of innovative thinking, I would advocate that I needed independent resources as an explicit investment in that.

David McMullin [00:17:23]:

And sometimes I literally would call it like a skunk works. And just to give you like one example, when I was running commercial operations at a multinational corporation, what I wanted to do was bring in PhD statisticians and AI type of developers who could leverage our data and do things with our data that our current processes didn't require. So we didn't hire people of that profile. So carving out those resources now in the startup environment, I think it's exactly the same as you bring in new profiles to build out to scale the organization, you have to start to standardize, you have to create swim lanes, you have to create repeatable processes, you have to put sort of these guardrails in place and that allows you to become bigger and stronger. But for the innovative people, it's like slowing you down. Right. And so what we've done at Anumana is built that up in parallel and we've kept the core R and D work as a standalone organization. It's small and it's got those innovative people in it.

David McMullin [00:18:26]:

They know that once the innovation gets to a point, it has to go through a design control process which is much more controlled. That's the only way you can create medical grade technology. But that little engine sort of discovery is still down there and able to operate. And it's got some messiness to it. Its purpose basically is to come up with ideas and patent the ideas or trade secrets, that type of thing. That's how it's creating value. But then it literally goes to different teams that we're scaling. You do sometimes find profiles of people who will start in one area and go to the other.

David McMullin [00:19:00]:

I think it's not uncommon. It's probably more common to find people who are kind of in one camp or the other. Either they're wired in a way where they want to work on rapid problems and they want to get it to a point where they can hand it to someone else to really operationalize it and turn it into something like medical grade technology. And the converses is the same. There's some people who want to really scale something and you need a mix of both. So at least in my experience and in my experience has always been kind of a carved out portion to do that exploratory, innovative piece.

Joe Krause [00:19:33]:

Another theme you hit on that comes up quite a bit, and it's always interesting to get different perspectives, is how to balance the necessity of operational excellence. Because especially in certain industries that you've worked in and healthcare specifically, those operational pressures are real. And as an organization, we help different companies bridge that gap between. We said we're going to do these things and here's what we actually were able to do. And one of the reasons when we asked, well, what didn't work in the past was they weren't able to actually carve out any time to think about the future and being strategic because they were so bogged down in the operational. And you mentioned one of the ways, you said, I need specific resources that will only be solely working on that so that we can make headway on that. So that's one interesting tactic to say like explicitly, I need these people that that's their job. So they're not obsessed, they're not sucked into the whirlwind of daily operations.

Joe Krause [00:20:18]:

But what other tactics have you employed or deployed within your roles in your various organizations to make sure there's enough time to be strategic and you're not just putting out fires day after day and you blink in a year has passed and no real innovation has happened. How have you managed that pull that comes along with it?

David McMullin [00:20:35]:

As the organization gets larger, you can start to sort of differentiate those roles, but when it's smaller, at some point it comes down to an allocation decision. I think I would say that's part of the secret sauce of an entrepreneur being successful. At some level, when you go down, you have to have executives who ensure that what you described, Joe, doesn't happen, that you don't just get caught up. And either end of the spectrum can be a derailer for organization. But you have to ensure that you're balancing both and talking about both. Now the easiest way to do that is to assign people, right. And different people to do different things and resource it that way. There's often where that's not possible.

David McMullin [00:21:17]:

And so like when we first founded Anumana, there were like three of us working on it and I had to just carve out time where I could like read published literature and explore ideas. Right now that couldn't be like my whole day, every day sort of thing. But if I didn't carve out the time and focus on it, I wouldn't be able to come up with a well informed storyline of what I think is possible in the future. Right. And the probably 2/3 of my day at a Minimum was like, okay, now what are the things we need to get done to hit the de risking milestones we need to hit and how do we get that done between the next, between now and the next board, for example, and just balance those things. But if you're not checking in on yourself and making it a part of the way that you schedule your time, you're going to trip at some point along the path. And then as you grow an organization, you have to instill that in other people until you get to a point where you can actually create swim lanes and say like, okay, this group is doing this specific set of activities. These others are, I guess as we've grown the organization, it doesn't have to be a large set who have to sort of bounce between those two, but it has to be the key people, sort of the executives and keep people from the next direct reports who are ensuring that that's the way that the company's run.

David McMullin [00:22:39]:

Let me give you a very practical example. Even at Anumana, the way that I forced myself to stay disciplined in that, in that process that was I committed myself to hold a town hall for all the employees a week after every board meeting. And in that town hall I would always update anything on the strategy front of what we were doing. And then I would update on the progress to date and the operational priorities from here to the end of the year. And then I had that scheduled. As soon as we were scheduling out the board meetings, we'd schedule that. And I set this expectation that also like put a stake in the ground where I was holding myself accountable by doing that touchpoint in front of other people and ensuring that it wasn't getting the tyranny of the urgent wasn't overwhelming my day to day and eroding my, my success.

Joe Krause [00:23:30]:

You make a good point because I mean a publicly traded companies, the, the quarterly earnings report is definitely something that people pay attention to, especially at the upper echelons. But in a startup, you have a board meeting, that board meeting is happening whether you like it or not, and there's a lot of pressure to get something done. So that point that you're making I think hopefully drives a little bit of what you're saying, which is you can't get complacent because the next board meeting is coming quickly. So the idea of spending a third of your time thinking about where you want to go and 2/3 of your time really working on those milestones to get there is I think, a good bit of advice for folks that are struggling with that Prioritization, where right now they might be saying to themselves, I'm spending 0% of my time thinking strategically. If you can bump that up to a third, that's a good place to start. It's not Mount Everest. You don't have to do an inversion. You can just ease it in.

Joe Krause [00:24:16]:

And I think that's really great advice because we get this question all the time.

David McMullin [00:24:19]:

Yep.

Jonathan Morgan [00:24:20]:

Yeah. And I think kind of one thing I loved about what your last answer was, David, is you talked about holding yourself accountable, giving those individuals insight into how things are progressing. And I think one natural transition point that happens, kind of tie everything together in our conversation today is at a certain point in an evolution of a company, the way you think about strategy changes. Right. In some aspects an organization is going to have this top down strategy. At some point they're going to have that, at some point they're going to have more an emergent strategy where, hey, we just have to learn and iterate and try things and we're going to have to adapt our strategy as we go. And then you might flex back and forth between those two. And in either of those cases it's important to communicate to your point how you're progressing and what those expectations are to the organization.

Jonathan Morgan [00:25:02]:

But kind of going back to that again, another, another conflict between top down, full fleshed out strategy versus that emergent strategy. How do you, how do you balance and transition between those two as well?

David McMullin [00:25:14]:

That's a great question, Jonathan. My answer might be an articulation of my current reality, but I think I've come to a conclusion that it always needs to be a hybrid of both. I really think that that's the reality. I can't prove that to you, but I think that really does apply regardless of the size of the organization. There's just never a situation where you know enough at an organization level to purely have a top down. But there's almost never a situation where you don't face something that clearly needs to be top down and then the conversion is also true. There's almost never an organization that can't benefit from a bottom up of hearing what is working, what's not working and adapting. I think there's almost never a situation where that left to its own devices will get you where you need to go.

David McMullin [00:26:03]:

How do you actually implement that? That's probably the tricky part because I think most people come to strategy without spending two years in business school thinking about business strategy or whatever. Most people come to strategy as just a part of their job and as the organization gets bigger, you have more people who are just subject matter experts in some area and they're working in a function. Maybe they got a master's or a PhD in something else. Their regulatory quality, product development. They don't typically organically just know or understand or appreciate the difference between top down and bottom up. And usually when they think about strategy, it's only top down. And they actually need to be explicitly kind of coached and educated on the concept of a bottom up. And that is an explicit part of our strategy.

David McMullin [00:26:50]:

That's probably what I've experienced most recently with an organization. Pick your function. It could be marketing, it could be sales. You'll often have employees say like, well what's the strategy? When you show the strategy, you point to the last quarterly town hall and say, hey look, this is where we're going. This is what the objectives are for today. And then they'll say, well what about this situation here? That's an example where the top down approach is not going to dictate what needs to happen. We need to learn from the market and get feedback and then adjust and continue to adjust. There's probably that element of translating strategy is something where it's easy to talk about.

David McMullin [00:27:30]:

It's actually I think it can be hard to make your culture one that has a mature understanding of strategy and embraces it as part of the way it does the work as it moves forward. And again, my philosophy is pretty much always going to be a blend of top down and bottom up. And you need to educate the team so that they see the reality in that framework and they understand the difference between discovering something from the market and feeding it up versus the direction that's being given from the top down.

Joe Krause [00:28:06]:

Analysis, the top down hopefully allows you to give get a little bit more isolation on your focus areas. But then the bottom up makes sure that you're not missing anything and also making sure that people feel connected. That's our strategy. We, we're in this together versus this was something that was given to us upon high. There's a real benefit, especially in the beginning when you have people that are joining on the idea of this is going to be something big in a few years that's so critical and as you've lived through. So I think that's sage advice because that's the question is always top down, bottom up. But the answer is why pick one if you could blend the two together for the best outcome. So I appreciate that.

Joe Krause [00:28:37]:

But as we draw to a close, we always tend to ask the same question of all of our guests to impart wisdom upon us. And, and ultimately, if you were able to go back to the beginning of your career, your first role, and give yourself one bit of advice from yourself to yourself in terms of your, your journey in the, in the strategy formulation and execution space, what advice would you give yourself to change the trajectory or make things different for yourself?

David McMullin [00:29:00]:

That's a really good question. I think the advice that I'd give to myself is that keep learning and you can learn from everyone in the organization. What I've gained through experience is sometimes some of the best insights or just really profound business insights comes from the most unexpected sources. You could have a member of the team who does not have professional training in shaping and building organizations and yet has insight on what needs to be done for the organization's culture that are very, very profound. Like as one example, that is something that if we all embrace, we have an opportunity to become better in our roles. No matter what we're doing. Every one of us has something to give to the organization, but every one of us also needs to continue to grow from our experiences in the organization.

Jonathan Morgan [00:29:52]:

Well, perfect. Dave, I appreciate your time today and all the insights you shared. I know our listeners will appreciate your insights and perspective and we look forward to catching up again in the future.

David McMullin [00:30:00]:

Thank you, Jonathan. Thank you, Joe. This has been great. I wish you all the best in your endeavors.