Using AI To Improve Cancer Outcomes, With Jeff Elton

We are seeing more and more solutions to problems as technology continues to evolve and advance. With the growth of artificial intelligence, we are adding tools to gather more data and insights into even the most complex problems. This episode’s guest is proof of that, using AI to improve cancer outcomes. Jeff Elton is the CEO of ConcertAI, whose mission is to accelerate insights and improve outcomes for patients through leading real-world data, AI technologies, and scientific research. He talks more about this to J.R. Lowry along with the other uses of AI in the business. Jeff also shares his leadership style, leading a thousand people and creating a great place to work in. Having been a CEO prior to ConcertAI, he then shares how his current role differs from when he was at QUE Group. For more insights about AI, the biotech space, leadership, and culture, tune in to this conversation with Jeff to not miss out!

Check out the full series of “Career Sessions, Career Lessons” podcasts here or visit pathwise.io/podcast/. A full written transcript of this episode is also available at https://pathwise.io/podcast/jeff-elton

—

Watch the episode here

Listen to the podcast here

Using AI To Improve Cancer Outcomes, With Jeff Elton

CEO Of ConcertAI

In this episode, my guest is Jeff Elton. Jeff is the CEO of ConcertAI, whose mission is to accelerate insights and improve outcomes for patients through leading real-world data, AI technologies, and scientific research. Prior to leading ConcertAI, Jeff was the North American Life Sciences lead at Accenture, where he also led their predictive health intelligence activities.

Earlier in his career, Jeff cofounded and led another Startup QUE Group, and also spent time at Integral, McKinsey, and Novartis. Jeff is on the board of the Massachusetts Biotechnology Council and also taught for a number of years at Boston University. He and his family live in the Boston area. Jeff, welcome. It’s great to have you on the show. I appreciate your time.

Thanks. It’s nice to be here.

It’s good to see you again. It’s been a while since we have talked live so I’m looking forward to catching up on what you have been doing since McKinsey days. Let’s start and talk about your company ConcertAI. Can you give our audience an overview of that?

ConcertAI, we are in a domain that sometimes gets called real-world evidence or real-world data but it’s become much more of an advanced AI research company. We have the largest collection of research-grade oncology and hematology data of anybody in the world. As part of that, we have data that would be called multimodal. It means we have a full exome, transcriptome, digital pathology, medical claims, electronic medical record, derived data, and social determinants of health data.

What differentiates us is people used to say, “Here’s data and it’s hard to get data. I will see how many insights I can get out of that data relative to my question but there are always limitations.” What we have decided is the question of interest usually requires highly specialized data. We try to make sure that we have the ability to syndicate and integrate all the data types that may be required for that question and by having data that is multimodal.

It also has you eliminate confounders and one of the problems in healthcare and life sciences. You get so far and then you find confounders limit your interpretation, therefore limit the actions and that’s one thing we are trying to get around. A lot of our insights are a lot more actionable, as our name might indicate. We start with data and we are very much into advanced AI approaches and how that can help glean additional insights. We don’t necessarily stay with AI at every step of our workflow but we have it in about every single part of our operations.

The reason why that’s important is sometimes if you don’t have tools to allow you to get what would be considered new and insights not bound by the literature or previously conventional ways of doing your analytics, you may be missing many things and delaying the time that you get to something meaningful.

Since the work we do is in cancer and oncology, we are constantly thinking about accelerating the insight that’s actionable or the biomedical innovation to the patients at all particular points in time. The way I think about it is we are a very advanced research company using a lot of digital and AI solutions to do it founded on deep, rich and broad data.

You have been in this space a long time, since before I knew you back in the early-2000s. Was there a particular spark or moment that led to the creation of ConcertAI?

What’s interesting is the group I’m working with is we are a long-term collaborator with the American Society of Clinical Oncology. We do a lot of work there and during the pandemic we did. That was a group that I knew before I was at ConcertAI. We work with a lot of community-based clinical providers and many of them were people that I worked with before I was in ConcertAI.

Back in other parts of my life, we’re dealing with a lot of biopharma innovators. Many of them are people that I have worked with either side by side when I was at Novartis or in other cases, most of my life. The idea and the concept of ConcertAI that came together with our investors was a very natural act for me. It provided almost a point where many threads converged.

I did not anticipate its timing in a few other things. The time and setup here came about at the same time that my kids were going into college and I was probably having the largest cash outflow I have ever had in my life. They wanted me to leave probably one of the higher cash-producing roles I’d ever had in my life. One may question my sanity at the moment but it also was such an unusual opportunity and one that I could sit down with my family and say, “Here’s a situation. You guys are going to be in on this too. Are we all in or not on this?” They were all in.

It sounds like you have got outside investors. Are your big investors individuals? Is it venture capital money? Is it corporate money?

I want to come back to your point about corporate. We have no corporate money but I want to come back a little bit about why we don’t have corporate money. We have a set of institutional investors who are our foundation investors. It’s a group called SymphonyAI. A gentleman named Romesh Wadhwani who originally had Symphony Technology Group has built probably about 30 companies through the years located in the Bay Area.

We also have Sixth Street Partners, which is a TPG spin-up company. Given your role in financial services, you probably know them. Sixth Street has been a very good partner and they invest a lot of things in oncology and healthcare-related activities. Both of those groups are what I would call more private equity-based and then we have Maverick Ventures out of the Bay Area.

They do a lot of work in healthcare and life sciences in very early-stage. They are much more on the entrepreneurial startup entity. AllianceBernstein is also there and then Declaration Partners out of New York City. Declaration is David Rubenstein, who had been the Vice Chair of Carlyle Group, one of his new vehicles.

All private equity and independent have caveats that they don’t. They invest in things in adjacencies but not directly counter to what we do. We are in exclusive positions in their relative portfolio. It’s one of the reasons why we have had offers and have talked about corporate money being biopharma. You could look at one of our peer/competitor firms would be something called Flatiron Oncology in New York City, which ended up having investment dollars from Roche and was ultimately acquired by Roche.

As we deal with so many healthcare providers and because of multiple deals across multiple biopharma innovators, we thought that our neutrality and anything that could contribute to trust, I have no conflicts to disclose. I have no exclusivity agreements or entanglements. I have no conflicts. We neither wanted the reality nor the optics given the nature of the work we do in healthcare. There’s a view. Whether it’s real trust or imbued trust, it doesn’t matter at the end of the day. Oftentimes the perception is the reality you operate in there.

What’s the state of the business? You have been at it for years. Where are you in terms of the number of people, revenues and things like that?

We are a private company so certain things. I will give you two references and the experts in your audience can back in and run their reverse multiples if they want to. We finished a C round in February of 2022 that established a $1.9 billion valuation at that particular point in time. That may be one of the last rounds we need to do. It’s a five-year period when by the time we close, we are expecting to be EBITDA positive and cashflow positive, which is in our comp. Very few, I’m not sure any of the companies in our space have been able to accomplish that.

We have made some decisions and part of it is the macroeconomic environment we have paid attention to because buying patterns and sensitivity to cost and structure are very high. We were taking a look at our growth. We have never not grown. Every period, we have grown independent of whether there was a pandemic or anything occurred.

We are incredibly thankful to our customers for the fact that we have been able to grow period over period. We also looked at it and said, “We could run a high net operating loss.” It had been traditional in our space where it gets shared points and used points and you will make it up with volume later on. We think the world’s going to pivot here a little bit and we’d rather pivot before the world makes us pivot.

We ran our scenarios, looked at everything we were doing and said, “This is going to take as much effort as it took to build the company but we think we can pivot our entire operating model and change how we build things, deploy things and what geographies we work in and force ourselves in 2023 to become EBITDA and cashflow positive”

When we get to the end of 2023, we will have this substantial majority of the raise we did in 2022, still very much in hand and be cashflow positive. We have about 1,100 people, which sounds like a lot of people. About half of that workforce is ex-US. We do have small teams in Europe and Japan and a pretty good-sized Bangalore operation, which we have had since our founding.

We are not a move toward. We don’t use the term offshoring. We have product leads, data science lead and AI teams that are located in Bangalore. We find India for technology-driven solutions. It’s a first-tier direction in place to begin doing that work. For us, it was much easier to mature and take advantage while their cost structure differentials. It was also about getting access to the talent we needed for doing that.

About half of what we have in place there, probably under that 45% is associated with a lot of what we have to do in data. A lot of what we do with data, we do interact with a lot of personal health information, fully identifiable data so we have very specialized workforces, technical environments and things in doing that. We have a scale that we can almost any program for any biopharma or any large provider system, we have the capability of taking on almost better than anybody else in the industry at this point.

That’s a huge run-up from the start a few years ago. It will be 1,100 people so congrats.

I know you know it. You feel it in different ways but the team has a lot of passion. The fact is that there’s a mission behind the business build given the space we work in.

We won’t have time to go too far down the AI rabbit hole. Apart from the obvious of using artificial intelligence to infer insights about treatment protocols out of the data, what are some of the other ways that you are using artificial intelligence in the business?

Anybody who’s reading that does work in healthcare will know that unfortunately the majority of the data, even though we quote use the term electronic medical records, most of what’s in those electronic medical records are physician notes, nurse notes and appended PDF documents. Those things are not machine-readable until you convert them into another form.

While we can pull in things like lab values may be in EMR, a lot of what we are doing is using natural language processing models. Many of which we are developing ourselves are called large language models. Large language models read certain terminology classes. As we also have a clinical workforce, we have our large language models try to achieve accuracy and recall comparable to humans that are doing that work or transcribing network.

We do a lot of human-based transcription too so we can index ourselves versus human expertise. That allows us to get a lot more value out of those records, gain insights out of those records and bring the analytic tools in them. That NLP and turning health data into something that machines can interact with is a huge enterprise in itself.

When we go over to the other side, there were a lot of the models we start developing. We are in what would be called GPT-2 and GPT-3 class models or large language model structures. That’s very common for us. Some of these things do very simple things like in cancer, non-small cell lung cancer, one of the largest cancers doesn’t have its ICD code, which would be a normal coding of medical data.

To determine between small cells and non-small cells, we can create a model and it can distinguish in the record what it is, since it doesn’t have a medical code, which doesn’t make any sense to us but it is what it is. All the way over to we can build tools to identify patients who are at risk of progressing more rapidly than the standard of care might indicate they would do and that has its utility for beginning to do it.

We also have one of the most broadly deployed class one medical device to do clinical AI work in radiological interpretation. We are doing 510(k) software and medical device models, which are AI models and these have to go through a regulatory approval process. They are deployed and used as part of standard care. That solution is deployed in 900 hospitals in Europe, the United States, Japan and some other parts of the world.

It’s integral to everything we are doing. Your question had something very important in it though, which is AI has to be trusted. You think about trust in a methodology. Trust comes from a combination of how you train it. Was the data set that you trained representative and appropriate to how you want to use it? Were you transparent? Was the model a black box, do the model do or not do? You don’t want it to be black-boxed. We have to publish a lot of things which you think would be counter to our intellectual property interests but you get no utility if it has no use.

Therefore, that openness needs to be there. You have to make sure the whole basis of training remains in tech all the way over. I’m sure everybody is reading the headlines about ChatGPT, everything else and generative AI. The difference in generative AI is you don’t have these incredible experts and domain experts who are doing the training, oversight and supervision of this.

You have self-learning, self-training and model that do sometimes unusual things. If they don’t see data, they will sometimes impute data. If you read this ledger, it’s this term hallucination. Whoever heard or thought we start working on model-based hallucinations. We are spending a lot of time and we are super excited about this generation but how you use the latest generation in the right way and right use case to assure its outcome and that you are not getting unintended negative outcomes. That’s something we are spending quite a bit of time on. To be clear, much of that is not in production, it’s in our labs.

There were lots and lots of discussions in the press, hallway conversations and cocktail party conversations. AI is very much front and center. It has so much potential. From your vantage point, how do we avoid the downsides?

Part of what you need to do in AI is the class of problem you are working on needs to be a worthy problem such that you are going to spend the time understanding whether or not the AI is adding value to it. Let me give you an example. You are in the UK and some of this may be across geography. In the United States, nurse resources are very hard to recruit. Also, research personnel to do biomedical research and get patients into clinical trials that could be beneficial for them. Oftentimes, they came from being trained as nurses are exceptionally hard to find. Many of these people left the workforce during the pandemic, either out of exhaustion or other things.

AI models can read records like humans, find patients for clinical trial eligibility and augment what is super lean resourcing. That’s a great use because we may never get back to the required staffing levels and I may need decision augmentation. The number of radiologists to do radiological reads and the number of graduating radiologists is going down.

You have fields that are changing substantially but they are open. I’m using the word augmentation to say, “I’m not replacing. I’m allowing individuals with expertise to be a lot more productive by augmenting what they see and how they can work with tools that are appropriate around that.” When you come over, it’s going to be a long time before “a tool is making a decision in replacement of the human.” These are all things that have to be earned.

I’m not saying that models can’t be appropriate for making a deterministic diagnosis but you have to go through a very deliberate process and almost think about the development of AI models and the same way you think about the development of a drug where we run clinical trials and we get visibility. The same day and you may recall this even from the old days of drug development, there used to be something called Blackrock drugs where it had the effect we wanted but we had no idea how it worked. Those don’t group anymore because it turns out they did other things in biology that we didn’t want.

There are a lot of these principles that need to become how we operate but our strong belief is it needs to be transparent. We need to know what people are doing and how they are working on it. Otherwise, this is going to be like what happened with CRISPR, which is gene editing. It all went underground and weird nefarious stuff happened in the background. The nefarious people will not stop because we are putting breaks on other things in what would be considered the lit-up transparent world.

You can’t stop the use of these tools but it will be an interesting policy debate that we watch play out over the next few years about how to put guardrails around their use.

It should be, even about personal privacy and the connectivity of my data and how it gets consumed and used. All sorts of questions that we have not had to debate before will come up on this.

Let’s shift gears. Talk a little bit about your role in leading the company. You said 1,100 people. How would you describe your leadership style? How does it connect to the culture that you are trying to create and the great place to work that is highlighted in your background there?

It is right there. I forgot about it. We got that in 2022 both in India and the United States at the same time. If you think about it, we are in a culture where we have data scientists, clinicians, cancer epidemiologists and engineers. Very quantitatively and rigorously databased people. We are also in an area where some of what you do doesn’t work but you have to do it first.

You are also in an area where everything has to be trusted. You can’t take shortcuts or if you know that there’s an expected outcome, you can’t manipulate things to get to the outcome or you make a lot of harm. If I put those qualities of what needs to be there, you will remember this even back from when we were both McKinsey colleagues. There was this obligation to dissent.

We had a different notion that challenge can sometimes be a negative word or challenge can become a mode of engagement and discourse to get to a superior outcome. If you level hierarchy in how you allow challenges where junior can challenge senior if it’s in the spirit of what you are trying to do, it’s that notion of obligation to dissent and challenge. I will add that when I was at Novartis, we developed a contact as an idea here, particularly for our drug discovery. If a program didn’t work, we would say, “That science wasn’t ready to be solved yet.”

That problem had not revealed enough about itself to be solved. It wasn’t the fault of the individual. We wanted the individual to do it. We wanted them to do the failure experiment first. If it’s going to fail, why it’s going to fail? Test that first. If you take those themes, there’s a culture behind it that indicates collaboration, intellectual rigor and intellectual sparring in a way to get to the best possible outcome allowing the iteration.

Failure is not bad. Failure may have been a necessary proceeding step. Things that you tried to do before will come back again. We are delivering things to the marketplace. We had ideas back in early-2018 or 2019 but we can do it in 2023. That’s one part of it, which is establishing that culture, establishing that value systems about how we operate and bringing them together. If you think about diversity inclusion criteria, we think of diversity inclusion criteria as being real.

At every possible level though, those terms get used but we also believe it in the sentiment of the perspectives and where the perspectives come from in the debate because it makes people sensitive to things that they otherwise, might not be. It’s one of the reasons why we allow our people to do research in areas like health disparities and things of that nature that don’t normally get funded even outside of our normal commercial lives and things of that nature.

The other part of it is my role oftentimes is we operate in three timeframes. We operate and try to meet the quarter or meet the year, which is on timeframe number one. I’m spending a lot of time thinking about what am I going to be doing by mid-year ‘24 and the latter part of ’24. I make my three horizon line, “Am I doing things that are going to make me successful in the latter part of ‘24?” Also, is that bridge from ‘23 to ‘24 going to get us to where we need to be in ’26?

This will be familiar too. Those three horizons and ways of thinking solving and operating are only going to come from the executive leadership team and people because day in and day out, most people either don’t have the luxury or don’t have the mandate or even the ability sometimes to go into it and bring that together.

The other part of it is when we think about recruiting people and doing things, I believe in a little bit of a symmetric contract with people. I go into with the spirit of they are selecting me. I’m selecting them. My ability to select them has no more power than their ability to select me. Part of my questions in my interview with them is to help them understand a lot about me to aid them to select me more accurately. I’m hoping they also do the same thing to help me do it.

That’s the pre-contract period. That same approach almost needs to be how people actualize their careers and opportunities in a post-period. If you come back to that, it’s leading that horizon and strategy. I’d probably be attributed with leading some of the areas where we innovate. This is a field I have been in for a long time.

I got a huge amount of energy. I’m excited about it. I love waking up every day and doing what I do. I start super early. I don’t know if you remember me and some colleagues. I was at 5:00 AM. Sometimes at 4:45 AM, I start as the working day guy. It’s because that’s when the world was quiet and I could dive in and get stuff. That’s a little bit of how I approach it.

Your points about the importance of challenge, failure, iteration and all of that. You talked a little bit about interviewing but are there particular approaches that you use to surface whether somebody’s going to be a good fit for your company?

It’s so interesting to say that because we have this as a debate all the time. We find that some things we do work well and some don’t work well. One of the principles probably won’t surprise you. Maybe you will wonder why it takes so long to be able to articulate this concept. One of the things we find is primary attribute number one is the level of personal curiosity and activation energy they put behind their curiosity.

Our most successful people are inherently very curious. They see and observe something. They see a relationship with something. They want to know more about it. They probe more about it. It’s interesting. We found all the things we said to predictability to be successful in our environment. Partially it was that motivated curiosity. Not just, “I wonder why that’s great.”

Jeff Elton: Our most successful people are inherently very curious. They see and observe something. They see a relationship with something. They want to know more about it. They probe more about it.

They’re taking a step to probe a little bit in terms of where it’s there and it allows them to change how we do what we do. We can get that out of interviews. You start to understand how people did what they did and what they do in the role. If they were a product leader, how did they even think about the definition of the product? How did they even know if there was a market behind that product? How did they understand what competition?

How do they understand evolving use cases or potential points of disruption and discontinuities about how markets were going to be working or deploying that? A lot of that comes down to that curiosity. Other parts become value systems. Think about the level of PHI and other things. This idea is that they are motivated by doing something for them. If I have somebody that’s super motivated by doing something for the patient, they will never have a malfeasant action ever in the environment.

They feel they are so responsible that the quality of their work would never adversely impact a patient. A lot of our people have done work in patient care and direct patient care. It’s that depth that makes them wake up. Why would they want to join us? My chief revenue officer, for example, came to me and goes, “I shouldn’t even be telling you this. I’m going to lose all of my negotiations standing with you. I’m a cancer biologist by training.”

He went to school in London at the Crick Institute for Cancer Biologists by background. Of all the companies in the world I could work for, I came to the conclusion it’s ConcertAI. I’m in a universe of one. Now that I have told you that, you hit me over a barrel but he’s not over a barrel. He’s got a great contract and structure for doing but I use it as an illustration that it’s that matchup between people like his motivations are deep in what he’s guided his career to for a long time.

I have been thinking about this topic a lot. We have rolled out a new value system in the company that I work for. To some extent, that will drive what we are looking for in employees but there are things undoubtedly that aren’t captured by our value system that are important ultimately to success. I’m at some point going to sit down and maybe take one of those early morning quiet hour periods and write down at least a thought starter list of the attributes that I feel like we are hiring for or need to be hiring for. What are the hardest decisions that you find that you have to make as a leader?

Some of them are what I call a little bit more on the trivial scale but I will put them out there anyway. I’m excited about what I do but I have to make trade-offs. We have the fortune of having more things we think are commercially viable than we probably have the resources or the management bandwidth to take on. You want to make sure you are right. We are looking for some things we do to be synergistic so that’s part of it.

Jeff Elton: We have the fortune of having more things we think are commercially viable than we probably have the resources or the management bandwidth to take on.

People issues. I’m a very people-oriented human being and I value whom I work with. When you are in a company that’s changed as much as we have and grown as much as we have, you are in the hundreds of million categories in terms of what you are doing here. When you get into that particular range, we feel like we are on our third generation of being a company, even though we have only been around for a few periods.

Sometimes the people that are appropriate for Gen 1 are not the right people for Gen 2. They are great people and they did a great job of moving us from Gen 1 to Gen 2 but they are not the right people. Part of my responsibility to do this is to assure that the team, the leadership and what’s there is appropriate for where we are going, not where we have been.

I have to admit, my background is I have been in earlier stage entities. I’m an active Angel investor but sometimes when you get through something that’s going through a very high growth stage, you don’t realize how much you do have to change in and change out and be willing to come to move those operating models. Those are very tough decisions with good people and it’s not the right place for them and us to get involved. That is probably in my personally hardest category to get done.

Experiences that are hard. I’m a very customer-centric individual. I give my customers incredible amounts of gratitude. My company couldn’t be here and I couldn’t be doing my role and what I did unless I had trust and customers took a very long-term view on working with us to go through and tolerate periods of what I will call subpar performance because we are still building and getting stuff but they tolerate it. They gave us feedback. It wasn’t always nice.

They didn’t say, “I’m terminating your contract.” They are saying, “Let’s get this right together.” People are willing to work through it. This is the same thing in personal relationships, spousal relationships and things of that nature. Working through the things that are not working builds strength on the other end of it and it builds capability in your organization to know that those are survivable moments that you don’t like. Why you don’t like it? You probably are going to avoid more of them but also know what it’s like to cure, preserve, grow and strengthen out of those relationships that can become institutionalized as well.

This is your second CEO stint. Your first stint was back with the QUE Group. Tell us a little bit about the QUE Group and do a bit of the compare and contrast on the two CEO roles.

There’s a big difference and some similarities. I will tick off a couple of similarities first. QUE Group was also oncology-focused. QUE Group also had to do with bringing early days of molecular diagnostic information and informing treatment selection if you were to go into the search tool of your preference and look at patents and things there. We did work on integrating molecular data with clinical data to start identifying the treatment approach most likely to lead to a beneficial outcome for the patient. We were also trying to build out a clinical network, which is something that we have.

If I touch the meta headers, it’s going to sound like, “That sounds a lot what you were doing at ConcertAI.” However, this was probably back in the 2000 or ‘90s period. I’d have to go back but quite a while as we are going into that. We had capital but were less well-capitalized. This time, we had a much stronger foundation capital.

That was one of the criteria I had when my family sat down. We said we are going to do this in a materially different position. We had a couple of underlying acquisitions which had been done coincidentally. One of them was one I was trying to do back at the time of QUE. If there was a little bit of, “I have been here before but the deal is done,” we’re starting assets on doing it. The scope of the team and the investor group that I’m with and particularly SymphonyAI and the experience they have of building 30 technology-centric companies are incredible.

I have done a lot professionally. I have a lot of people that will work. I have to tell you, when I work at 30 companies that have had their ups and downs and near-death moments and to get the council out of my board group in doing it, these guys are outstanding. I soak up the guidance in addition to them being able to apply the stuff here.

It’s this notion of having the right team, capital, board and partners and even the right founding assets. We got to product and revenues fast. That was something very important to me to be a revenue company. It’s incredibly fast. Those things were all very different. I internalized my learnings. I will own everything and the insights. I will own what I don’t want to do. I will take all the feedback from my board. We are very oriented to making sure we get to where we need to go.

Go back to the beginning of your career. You got the trifecta of degrees from the University of Chicago. You had three undergraduate majors and then you got an MBA and a PhD as well. You have covered the ground well in South Chicago. Back then, what did you foresee that you would be doing professionally?

I’m on the margin of embarrassment or lack of self-awareness. My mother who has passed was a biomedical researcher and had discovered drugs at Merkin & Co. My father was a biomedical research dermatologist and physician and also did a lot of work on a dual board in pathology and dermatology. He did a lot of work in topical skin cancers and melanomas. Sitting around the dinner table, it wouldn’t be unusual to look at rather icky photos of different things on the skin and have an interesting conversation and things about that. I had determined, “I’m going to go my path. I can’t go into healthcare and life sciences. That’s what they are doing. I need to go do my thing.”

In a lot of the early stuff I did, probably even early consulting, I was doing technology work. I had the pleasure of doing some work at Dell Computer when there were 300 employees there. I saw the last days of Zenith Data Systems before they were acquired by Groupe Bull. It’s the whole early days of tech. I did that for a while. My PhD is in Economics. The expectation back when I got a PhD is a little bit different in 2023 but the expectation was I was going to stay in academia.

My committee and not the person who was my thesis advisor, Harry Davis, who was the deputy head of the business school and a gentleman named Edward Lazear, who’s out at the Hoover Institute at Stanford, were very good about working with me but the rest of the people at the institution said, “If you are not going to go teach, why are you in this program? Why would you get a PhD? Why should we put the effort into you if you are not coming back into our world and our orbit?”

The reason why I was there is I enjoyed the depth of thinking through theoretical constructs, taking on a large heady research project and going through. It allowed my mind to continue to mature intellectually. I loved the interaction I had with the faculty there and what I was doing. For me, I felt like I was maturing and finishing what I wanted to be intellectually and how I wanted to think and problem-solve in my quantitative skills and background. I did a lot of work in quantitative finance theory at the time and did some stuff with Myron Scholes, which is also one of the things that got me involved at the time.

When I finished graduate school, did that and graduated, I also said, “I got to get out of Chicago.” To your point, I have been there a long time in the South side of Chicago. There’s got to be a little bit more of the world so I ended up in Boston. I wasn’t supposed to stay in Boston but I’m still in Boston. It’s a little side story but I also decided I wanted to go. One of the reasons why management consulting was so appealing to me was there are a lot of fields and disciplines I could get myself intellectually interested in and I did.

I got captivated by all of them and then I got back on pharma and healthcare again while I was doing consulting and then I did another. All of a sudden, I said, “The idea that people trained themselves like crazy and went into a field where the reason why they are there is to better humans, take an ill person and make them well. Take a horrific situation and make it less horrific and ultimately restore them.”

The sentiment, values and behaviors of people doing biomedical research and everything else appealed to me. I understood how I grew up and what had been inculcated into me in my values growing up. I don’t think I had that level of awareness until I started doing work in that area. I go, “Full circle. I’m comfortable. It’s my decision to come back into healthcare and life sciences as opposed to my dad telling me, ‘You should go to med school.’”

At that point, I then said, “Everything I would do is going to be biomedical research and healthcare-related.” That’s why I continue doing it at McKinsey. That’s why I went into Novartis and I was in their discovery through early clinical development. That’s what I did when I was at Accenture. Everything I did was for healthcare providers and doing that. I’m an active Angel investor. Everything I did was healthcare biomedically related. It’s not a regretful move in the least. It continues to be the field that pours out.

I’m on the mass biotech council. I’m on the committee. I have done that for years with four rounds of strategy sessions and things of that particular nature. Sometimes there’s a path one needs to go before one has the level of self-realization that then allows one to be directed in a super comfortable way. I can’t have enough hours in the day to think through what I want to think through, read what I want to read and write when I don’t want to read. You know that when you got this back pressure of motivation.

Jeff Elton: Sometimes, there’s a path one needs to go before one has the level of self-realization that then allows one to be directed in a super comfortable way.

Your passion for it is evident. Not everybody gets that. You had to take a somewhat circuitous route to get there. There’s a little bit of, “I’m not doing what mom and dad are doing. I’m going to go forge my path,” but you came back to it. There’s no sin in that. You are helping people and you have been doing it for a long time. That’s great. As an aside, Myron Scholes works for our firm.

When I got out of my MBA and I don’t even know if he will even remember this but this was way back in the day, I had a class with him. I was doing something. It was back with the Black-Scholes option for pricing models and things of that nature. We had him who was super highly quantitative and very off on the edge of some of what he had been working on. I did something that wasn’t even clear if I had the quantitative capacity to do. It was exciting to work with him. He gave me this A+ and I’m sitting going like, “An A+ from Myron Scholes. I felt like I would frame that and put that on my wall.”

We had a conversation about it and I ended up then transitioning into the PhD program. I was always asking about finance and financial theory in the economics world. Economics ultimately resonated with me because I had more direct applications to the work. I’m a fairly hands-on person at the end of the day and want to start putting some things much more rapidly into models that can be applied and economics, at least for me, then a little bit closer to being able to do that. Economics is still a supernatural act for me. I’m still passionate about that as well.

Congrats on getting an A+ from a Nobel laureate.

I probably have it somewhere in my files.

You have done a couple of startups. You work for a big company, Novartis. You have been in several different consulting firms and rebounded to another consulting firm after your time at Novartis. Do you look back on any of this and say, “I can’t believe I did that. That was not a great fit for me?” Is it all thrown into the stew when you have gotten something out of each one of those experiences and you look back on them all as right for you at the time?

I don’t look at any experiences as being negative in the list and even things that are challenging on multiple levels. You know how firms work. There’s neutrality in politics that all intertwine, its little healthy and unhealthy DNA sometimes of the organization. At McKinsey, we had this incredible vantage point of working with executive teams. I don’t think my maturation about how an EXCOM looks at things could have occurred rapidly if I hadn’t been at that firm at the time. The set of colleagues I had were spectacular, the intellectual depth and effect.

Some of the partners that we have then continue to be close personal friends. One that you and I had in the Boston office lives a half mile from me and his wife. We are still friends. We see each other regularly. It was a social work and I learned a lot. I left to join Novartis, which was a white paper that I had worked on with a set of other McKinsey colleagues that established what I thought at the time as one of the leading biomedical institutions and concepts ever that had an extraordinary amount of funding attached to it.

We did the white paper. We advanced it and worked on the initial setup. The chairman of the company, their head of clinical development, and the head of research, all three asked me if I would leave McKinsey and come join. I almost eat my cooking, finish the bread, and this entity up. I’m sitting here going, “Where am I going to have the opportunity to spend the billion-plus resources to fundamentally transform and change how biomedical is going to be doing at that scale?”

For me, it was like, “I have not been in the industry. This is weird. I’m going to come in an SVP role.” Most people grow up in that environment with incredible trust. I have to tell you, at the end of that first year, I was exhausted. I thought I was a super smart guy carrying a PhD at McKinsey. When I got to the end of that, my wife and I sat at the end of the first year in our fireplace and I said, “I have never had a year where I learned how little I knew.” You take on that operating role. I had people in Basel coming up to me and saying, “I worked for you. What do you want me to do?”

Seriously, that was the first line. I knew it was a challenge. You are the smart guy. You are the guy from Boston. You have no background in this industry. What do you want me to do? I’m sitting here and the amount that I had to learn come down the learning curve but also it bridged for me. Council means to be actualized. The first thing I said at the end of that year was, “I would have changed my council at McKinsey if I knew what organizations could do.” We were smart but we didn’t always know all the details of what it takes tens of thousands of people and get that done when you are responsible on the ground and you pull it.

I learned a lot coming out of that experience. Coming out of that, I then went entrepreneurial again. I came back in. For me, there were no regrets. Accenture was tech-deep. I was tech-deep in AI and advanced analytics stacks. I got to build things to do predictions of heart failure patients before they decompensated to stop the decompensation.

For me, all these things are built upon them and I can be integrative of those things even as I go into my next role. No regrets. I wouldn’t say they all were hyper-planned. You would agree. The way I have worked and operated and always had an openness and built my networks created some of those opportunities along the way. If I’m super excited, I always thank people for thinking about me and I don’t even try to remotely get into conversations but have no regret about anything at all.

You are a very high-energy person. What do you do to recharge your battery and keep yourself energized?

A couple of things. I do start each day very early and I still do read whatever is published in medical literature and healthcare literature. Having time to think and not just dive into sets of Zoom, Microsoft Teams meetings, and things of that nature, which are processed-based interactions. I do have to start that way. I do have to make sure that every single day I’m doing some form of workout.

Generally speaking, if you look at what I do in my free time, I’m kayaking, biking, and hiking. All nature and no technology for the most part. My body is moving it or I’m moving myself through it. That to me has been increasingly more important and those are the more restorative things. The things I don’t need are things that I have sought in my late 20s, 30s, and probably even 40s.

I would love to go and be in the center of a city. I would be captivated by urban life. It’s fine. I do it and I enjoy it. I do it because I want to be there with the people I’m with but my restorative time is getting out of there. Also, doing these other things and they have this Zen repetitive motion quality to them that is restorative.

I’m a lot like that too. I’m living in a city so it’s a little bit different. I was back in Boston for a wedding. I was driving around the suburbs and seeing green space. All of that was nice. I’d get out, run, hike, and do some of the same things. My thinking has always involved a motor. It’s not for me. I do the things that don’t involve a motor. It’s similar to the way that you were thinking about it. We are running up against time. One last question. If you could go back and tell your 22-year-old self one thing that you have learned the hard way along the way in your career journey, what would it be?

I have tried to do it because I have a 21 and a 23-year-old myself. You got a unique relevance. You happen to hit right in between the two of them. Part of it is don’t worry about the decisions you make in your career so much. It’s going to be a long career. Number two is don’t worry about whether it is a wrong answer or a right answer because discreetly and binary right or wrong is almost never the case in most jobs you bring to it.

You are probably not going to make a wrong answer because you are going to pick that up on your radar even in the culture and other stuff. Don’t overanalyze and oversweet right or wrong because all these things are recoverable. If you want to do something that looks like it’s higher risk and you want to take a period, throw yourself at it, build something, try something or do something, do that because you still have every ability to get back on the track you were on before. It’s not a problem at all.

Even when you and I were there and nobody talked about it, it was still possible to do that because people came back to McKinsey, Accenture and Novartis. I see that pattern a lot but it’s even easier than it was before. People even like it when you return to venues and things. You may never want to go back. It’s this idea that the world’s not as risky as you think it is and all of these things are probably important for you to feel that you have been able to do it and you will make other decisions differently.

The other part of it is your work life and home life. As much as people say, “I’m at home or I’m at work,” intense people do intense things halos onto the home. Home can strengthen what you bring. Understand those dynamics. Let them bring benefit to the other. If I bring my excitement and my energy about what I’m doing home, that extra two hours that I may have at work isn’t resented. There’s a motivational energy that comes of it. This is a difficult thing. We tend to be very compartmentalized there and humans aren’t compelled. We are systems. That’s probably my last one.

Somebody I interviewed on this show talked about not liking the term work-life balance, and preferring work-life integration because it does all come together. It’s hard to compartmentalize. Jeff, thanks. I appreciate you doing this.

Thanks too.

It was great to catch up and learn what you have been doing since your days at McKinsey. Congrats on what you have built. It’s amazing.

Thank you. We are still on the journey, so we are not at the destination yet by any means. I appreciate the opportunity to be here too.

It was great catching up with Jeff, discussing his career journey, what he’s doing, the company he’s built in ConcertAI, and everything going on in the biotech space, and how he thinks about leadership, culture, hiring, and many other things. There are lots of great insights in there. If you are ready to take control of your career, you can visit PathWise.io. If you’d like more regular career insights, become a PathWise member. It’s free. You can also sign up on the website for our newsletter and follow us on LinkedIn, Twitter, Facebook and YouTube. Thanks.

Important Links

• ConcertAI
• Massachusetts Biotechnology Council
• SymphonyAI
• Sixth Street Partners
• Maverick Ventures
• AllianceBernstein
• Declaration Partners
• LinkedIn – PathWise.io
• Twitter – PathWise.io
• Facebook – PathWise.io
• YouTube – PathWise.io

About Jeff Elton

Jeff Elton is the CEO of ConcertAI, whose mission is to accelerate insights and improve outcomes for patients through leading real-world data, AI technologies, and scientific research.

Prior to leading ConcertAI, Jeff was the North America Life Sciences Lead at Accenture, where he also led their Predictive Health Intelligence activities. Earlier in his career, Jeff co-founded and led another start-up, KEW Group. He also spent time at Integral, McKinsey, and Novartis.

Jeff is on the board of the Massachusetts Biotechnology Council. He also taught for a number of years at Boston University. He and his family live in the Boston area.