Entrepreneurs Rx_John Cordier: Audio automatically transcribed by Sonix

Entrepreneurs Rx_John Cordier: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

John Schufeldt:
Hello everybody, and welcome to another edition of Entrepreneur Rx, where we help healthcare professionals own their future.

John Schufeldt:
Hey everybody! Welcome to another episode of Entrepreneur Rx. Today, I'm excited to have John Cordier with me, who I met in a bowling alley, believe it or not, I hadn't bowled in 30-plus years and met him there at a VC conference. John is the CEO and founder of Epistemix, which is hard to believe but really got me interested while sitting there watching, you know, watching other people bowling and me scoring like a 76. So it was that interesting that I remembered it from that. So, John, welcome.

John Cordier:
Yeah, thanks, John. Really glad to be on.

John Schufeldt:
Thanks, now, did you end up bowling that night, or were you just wondering?

John Cordier:
Yeah, I did end up bowling. I think the top score I got was maybe like an 89. So, yeah, I'm right there with you.

John Schufeldt:
It was funny, you know, I remember bowling as a kid. Like, I'd always at least get over 100. I think the highest I got, I think it was about a 76. I mean, I just totally sucked.

John Cordier:
The, let's say that the slope of the lanes where, if you weren't on, you're definitely in a lot of gutter balls, and I think that was the same not just for us, but for a few others.

John Schufeldt:
I'm sure there's a metaphor in there for entrepreneurism and startups.

John Cordier:
Probably, I mean, if it was 79% of 300, if that was the likelihood of success of the people in that group, we'd be Midas list type folks if that was the case.

John Schufeldt:
Oh my God, yeah, I'd be. I'd be banking on unicorns the whole time. All right, classic. All right, so, Johnny, you have a really interesting background, and what struck me about you was I told you what, what you, I'm sure, you pitched it to me, but what you told me about was something I had an interest in before, and it's cool that you're doing it, and then some. Give everybody a little bit of your background because it's kind of unique. And then we'll go into EpistemiX and how you are where you are.

John Cordier:
Sure, So my background growing up, I lived in 13 places before I was 18. Military family bounced all over the place, and through that, I kind of got a lens into all different types of culture, all different types of people, And understanding the social dynamics around health was that I eventually realized that that's how I viewed the world. So I went to college at the University of Pittsburgh. Initially started off in neuroscience, was doing a Philosophy and History of Medicine certificate, and got exposed to the sociology of medicine. And that kind of got me hooked into understanding the social context of how health emerges isn't just like a given thing for anybody. So I added on another degree in sociology, and after trying out a couple of other things, whether it was to see if I want to do education policy and do a Ph.D. in cognitive neuroscience, I actually landed back in public health, which when people ask me, How did you end up in public health? I said, Well, between neuroscience and sociology, public health I think is squarely in the middle. I get to pull in all different things that I'm intellectually curious into the work that we're doing today.

John Schufeldt:
But how about the entrepreneurial path? When did that get you?

John Cordier:
Yeah, I started my first business when I was 14. I actually was, I saw a lot of houses in the neighborhood that were up for sale and their grasses weren't cut. And so I said, Well, I'm going to go to Howard Hanna and say, Well, I bet you I can help you sell your houses quicker if the lawns looked okay. So I ended up through middle school and high school running a lawn business with me and all my friends mowing lawns for people that moved out. But the serious side of entrepreneurship, I never thought I would end up being a CEO or a founder of a startup company. But when it hit me, I was 26 years old. I was a graduate student in the School of Public Health at the University of Pittsburgh, working with the dean on developing the software platform. And what we realized and we were nudged by one of the people on the board of advisors who runs our 35-hospital health system. He said, Guys, look, the software is great, but you're not going to be able to have an impact and software's not going to be an impact after the ten years you guys have been working on it, unless you get it commercialized. And it was really that moment of understanding the translation of the deep tech and the deep science that we were working on and the way to have it make an impact was connecting it to a market need. And it was really that that first summer I started going through the different innovation center opportunities, working with the Entrepreneur Institute at the University of Pittsburgh. I got involved with the Schwartz Center at Carnegie Mellon University and really just started gathering up all the tools that I felt could be helpful in eventually getting this company launched. So I was a graduate student that was crazy enough to take on a commercialization project, and here we are five years later.

John Schufeldt:
Was this your grad project? Is that how it started or did you see a need and say, okay, I'm going to poke a hole in that?

John Cordier:
Yeah, it was even bigger than that. So the need was first identified in the early 2000. So my two co-founders, Don Burke, who was dean of the School of Public Health, and John Griffin ... who was leading the Public Health Dynamics lab. Don had led infectious disease research for the military for 23 years. And then prior to, then after that, he led infectious disease research at Hopkins. And it was when he was at Hopkins, he realized that the world of computational modeling and mathematics for decision support and public health, there really wasn't an overlap. And the big challenge was, if it's an emergent disease, an emergent epidemic, anything that's changing socially from a behavioral point in a population, there isn't, there aren't the tools to enable the medical community to understand, well, how do we respond and how do we react to this? And Don's whole kind of the big gap that he saw was we really need prospective tools that enable the clinical side and the policy side to be able to come together. And he formed a group that's called Midas, which stands for the Modeling Infectious Disease Agent Study, and in founding Midas, he basically set up a network of infectious disease modelers and researchers globally that we were the coordinating center at the University of Pittsburgh. And what a lot of these groups did is they started developing novel models, but instead of just going after a disease or a specific model, we said we were going to build a platform to enable any infectious disease researcher or any public health researcher to build models, to test interventions to actually lead to improving outcomes.

John Schufeldt:
So the example you gave me while bowling was really, really insightful. Tell people that example because I thought it was really cool.

John Cordier:
So what we ended up doing, we have a representation of every single person in the entire United States, every school, every workplace, every household. So when people think of the game SIM City from like the eighties and nineties, we really have a realistic SIM city of the entire United States. And the example here was how do you understand how to mitigate or prevent the spread of an infectious disease once it's already started? And so we began running models to test how do you intervene upstream to prevent the further emergence of COVID. Prior to that, we're actually running simulations. During the H1N1 epidemic, we were embedded within the White House. When to close schools, when to close workplaces, how does a workplace understand what policies they need to put in place to ensure the protection of their workforce? And we ultimately ended up getting a lot of traction and helped the global events industry return over the last couple of years. So from the Consumer Electronics Show to HIMSS to HLTH coming up this upcoming week, we've been able to run an entire suite of models. There's actually 240 different scenarios or combinations of interventions, and then through that, it gets distilled down into here's the best case, here's the worst case. You clearly want the best case. Let's look at those policies to put in place so you can gather people together safely.

John Schufeldt:
So you gave an example. I'm sure I'm not going to think of it, but you gave an example of, look, if you do X, and X was relatively small, you were going to prevent, here's a downstream effect of it over the next few years. Do you remember what that example was?

John Cordier:
So the example is probably something to the effect of more of the behavioral modeling that we've been looking at. So in the behavioral modeling, similar to the emergence of an infectious disease, we've been able to look at the emergence of things like the opioid epidemic or the emergence of things like improving cardiovascular disease in a population over time. So from the behavioral modeling side, we were probably looking at on the infectious disease side is how do you, nudging people from a behavioral standpoint, whether it might be signing up for a screening, signing up for doing a regular test for COVID, flu, RSV, whatever it might be, looking at those early interventions that can be put in place could have a disproportionate effect on the total size of what the epidemic could be. And one of the things that we've been able to do in advance of that is looking at, well, who actually cares about more or less disease happening in an entire population? Yes, there's the insurers. Yes, there's the healthcare providers. There's also the manufacturers and those that are producing their therapies, drugs, whatever it might be. So one of the things that we ended up doing was not only doing disease forecasting, but let's say you're Walgreens, and what Walgreens is trying to understand is, well, how many COVID vaccines are going to sell out of our stores? What do we need to be stocked with? And so there's both the disease forecasts, then there's the behavior of I'm sick, what am I going to do? I'm going to stay home. I'm going to go with one therapy or another. I'm going to get vaccinated. And we can start testing what interventions or what marketing strategies can be used to nudge people and move people into the direction that can lead to more preventative care. And in the Walgreens case, not only the prevention side is something that's good for them, but also how do you rightsize your inventory across thousands of stores?

John Schufeldt:
Right, I mean, it's a little bit like the butterfly effect. I mean, it's just you make this small incremental change that the outcome of which translated across the population could be absolutely huge.

John Cordier:
Yeah, I mean, the one thing that our approach enables people to look at is the emergence of these broader population outcomes. And it really starts with understanding how a small change in behavior can lead to a disproportionate outcome. Because when you consider how connected people are, you're not really looking at a linear step, right? And so, exactly, so everything can be looked at as this compounding impact and health is such a complicated thing. But how, the emergence of health, and people talk about the 80% outside of the health system, well within that 80%, sure, you have education, you have the social conditions around somebody. How can you look at something like housing? And if you are able to move people into housing, what is the broadening effect that can have on somebody's health and opportunity? So our software is able to look at how these small changes in one person's behavior, if you're able to replicate that across an entire population, how that leads to really different outcomes?

John Schufeldt:
Did you run models as related to COVID with okay, if we had mask mandates right out of the gate if we closed schools right out of the gate, did you look at those models and say what could have been versus what it was?

John Cordier:
Yeah, so those 240 different scenarios, those all include what could have been if we did different policies with schools, what could have been if governments like the state or city level intervened in different ways. And so what you're able to do with this type of software is generate all of these data sets to test counterfactuals. So here's what actually happened, but what would happen if we had open schools earlier, closed schools earlier, had more people go into the workforce from working from home quicker? All of those things can be tested with this type of software to say, well, the next time around, here's what we could do differently. But better than that, to say if we understand what's coming or what could be coming, how can we know what plan of action we need to put in place before it happens? And that's really like the scenario planning exercise that this type of tool and software can be applied into the policy and planning strategic scenario planning type market opportunities.

John Schufeldt:
Yeah, I think I mentioned I met with a friend of mine who was a Ph.D. from Hopkins and Epidemiology who did disease modeling at ASU and had this and had this whole theater, for lack of a better way to say it, on disease modeling. And I kind of pitched this idea to her, I said, Wouldn't it be cool because people could buy insurance against the catastrophic loss that would come from this, I mean, just, Well, you've already, you're doing it, so you totally get it. It was just, it had to be frustrating and it changed course a little bit. Knowing what you know, I would think it had to be frustrating sitting here watching a number of deaths check off every day and saying, no, we knew this was coming and we could have done a much better job mitigating these deaths. Was that hard to stomach every night knowing what you know? Because you kind of have you know, you're kind of like The Wizard of Oz in some respects. You're looking behind the curtain.

John Cordier:
Yeah, it was frustrating, and also for everyone on our team who we were running simulations every day and understanding here's the number of deaths, hospitalizations, those that are going to be in ICU, those cases and recognizing that when we're running these simulations because we're representing each person in the entire population, that these are actual people and it kind of humanizes the data, and so, yeah, it's definitely frustrating. And one of the challenges that we had early on was, how do you communicate that? That's the type of, that's the level of analysis that's now possible compared to just taking a spreadsheet model and just saying, Oh, well, we think it's going to be this sort of trajectory and curve. And we, my co-founder, Don, he was given the spreadsheet out of the White House earliest October of 2019 of, hey, this epidemic we think it's coming. Here's what we think it's going to look like. And it was really eight cells in an Excel spreadsheet. And we're like, okay, this is clearly not the right way that we need to be making decisions specifically for the entire country. And so we got to work pretty early on developing models, and Neil Ferguson was somebody out of Imperial College London, part of the Midas group that he kind of got the word out on the U.K. side. And of course, everybody will say, well, the models at the beginning of the pandemic are totally wrong. Well, you hope they're wrong because you hope the behavior changes because of it. And this is something that we struggled with early on when we were going from, there's the national level, from decision-making earlier in the pandemic to the state level to the local level. And it seemed like at each one of those, the buck was getting passed on to make more local, local, local decisions, but those individuals, they're not all equipped with epidemiologists on staff, let alone understanding what the novel advances in the field are. So translating the policy and the political decision-making with the scientific decision-making is something that we've gotten much better over the last couple of years to the point that business decision-makers are able to trust the science because they can see themselves within the data. So going beyond just situational awareness of here's what's happened, here's what is to be able to ask the questions, well, what if we do this next? Or what if we test a different strategy? And that enables them to think about creating a different future or creating a different possibility for their workforce, for the health of their population, if their health plan for planning resources, if they're on the provider side or providing goods and services. So that's sort of the evolution that we came to over the last couple of years. But to get to the point of the question, yeah, it was definitely frustrating. And then overnight everybody became an armchair epidemiologist, so that's a separate thing. But yeah, we've overcome some of that.

John Schufeldt:
Yeah, that became armchair, a lot of things. I stopped asking people in emergency department if they were vaccinated because I was like, I almost don't want to know because half the time I just shake my head and be like, Well, that's why you're here and I'm going to put you on a ventilator. And they still didn't seem to buy into it. When did Epistemix start?

John Cordier:
So we incorporated in 2018. We licensed the end of 2018. We licensed the software out of the university in 2019.

John Schufeldt:
Wow, so you literally had the ultimate and it probably wasn't an MVP by that point, but you literally had the ultimate product-market fit with COVID. I mean, in some respects the timing was either great or like, Oh shit, depending upon what your perspective was at that moment.

John Cordier:
Yeah, and so this is really all the way back to when Don, in the early 2000 said like, there's a need for this type of tool, and so the NIH funded us, the Bill and Melinda Gates Foundation, Robert Wood Johnson, DARPA, all were aware that this platform existed. And then the challenge was how do we get it into the hands of the decision-makers? And so our first customer ended up being a consulting group called RESULTANT. So they were hired by the state of Indiana to run all of the modeling and simulations to help the state with their policy and plans going forward. So we think April 2020 was our first customer, and then they started running models and simulations, and then from there we had the election in 2020. So they're like, Well, what happens if we do live, like people are going to show up and vote? We don't have vaccines out, but people are all standing in line and gathering inside, is that going to cause some like additional spikes? So we're even able to model down to the level of like you're bringing 200 people together at each of these voting places. So they're able to do all sorts of different analyses with the tool, all in the context of COVID, but really in the context of how our behaviors and how are our policies impacting the outcome that we're going to see.

John Schufeldt:
... be very frustrating to run the model, have the data, here's the support for it, and then have it ignored because I'm pretty sure your data was ignored by a few folks.

John Cordier:
Yeah, a few folks, and even to the point where in 2021, right before the Omicron wave took up, so we modeled that the CDC there, they've been using the ensemble forecasting group and the challenge with a lot of what those different techniques are doing. They're all very statistical-based, which like there's a time and a place for that, but when behaviors are changing so much, if you're not including the behaviors of people and decision-makers, you're really missing a lot of what's likely to happen so early on. It's like, Oh, look, there's going to be another wave, none of those models had it. So there's a bit of frustration there, but I think this is a good point of overlap with where there's actually an opportunity. So somebody who I met with recently, he's saying, well, you kind of fit into this synthetic data opportunity. And his take in, of synthetic data and healthcare, is that it's controversial, some people are all for it, some people are like, absolutely not. And he goes, whenever there's that sort of conflict, he goes, there's usually something within it. So my example to draw in from there is we actually worked with Drake University, their university president, and their executive team.

John Schufeldt:
What university?

John Cordier:
Drake University.

John Schufeldt:
Drake in Des Moines, Iowa.

John Cordier:
Drake in Des Moines, Iowa.

John Schufeldt:
My alma mater. I was on the board. Go, Drake. All right, plug.

John Cordier:
Yeah, so we worked with Drake. And what the team at Drake was trying to understand is, well, now that vaccines are available, this is '20 fall going into fall 2021. And what their question was, well, only about 45% of our students are reporting that they're vaccinated. We understand that we want to try to get to 70% to reach some level of protection across the campus. And then if we're bringing all these people back into Des Moines and what impact does that can have on the city? And so we ran through a number of scenarios with them, the earliest June and July, before students were coming back on campus. And if you think of the seasonal pattern of COVID flu, any respiratory disease in the US, that's the down season, and so you're at the trough of the wave. And what the folks at Drake said was like, Oh, well, do we even need this data? We don't know, like, come on, we're not seeing any cases, it's not going to come back. And we said, Look, here's the scenario. We're not just going to give you the data that says everybody can come back on campus and it's going to all be okay. So they were upset for a little bit, but then what they actually use the data for was as a benchmarking tool. So when students did come back on campus, they're like, Oh, well, here's what Epistemix said would happen in September. Oh, crap, that's happening. All right, we know what policies to put in place now and then they used us as a benchmark to say, all right, masking social distancing, what are we doing across campus? What's going on within the city of Des Moines? And they had an entire decision support tool to then say, all right, we know what strategy to put into place September, October, November, December. Did that for the entire semester, and they, all they were trying to do is stay below what we had projected. And there was sort of the resistance to taking in data, but then when people are able to see, oh, wow, this is actually being reflected, that had a really powerful impact. And that's one of our best case studies that we talked to other people about.

John Schufeldt:
Wow, I'm impressed. That's very, I mean, Drake's a small 5000 liberal arts school in Des Moines. I'm impressed that they did that. That's very cool. I'm proud. Proud of it on the board. The, and it's such a confined campus, too, it must have been a really cool model because it truly is a relatively enclosed campus.

John Cordier:
Yeah, and then you think of where the professors are at, they're interacting with the students or then going back to their homes, households. And what impact does that have on the city? So yeah, it was a really good model to see health, which is in this case we're looking at the emergence of COVID cases, how that's really connected to the behaviors of what policy decision makers at the university were doing. And they could see how their decisions impacted students, professors, and then those around. So it enables you to think about some of these social determinants of health or behavioral health modifications in a different way.

John Schufeldt:
Now, switching from the science to the business, how have you been able to monetize this? I mean, I totally get the value, but, and another people do, obviously. But how do you monetize it?

John Cordier:
So we've gone through plenty of iterations on this. So at the beginning, it was here's a licensing fee, and here's consulting services on top of it. We then said, well, by having the consulting fees be so much upfront, it's preventing people from interacting more with us. They're not getting as much out of the tool because we're not able to help train or see how they're using it, so we dropped some of the professional services fees and then we ebb and flow between licenses and professional services. We then went to what we did with the events industry. We said, Oh, we'll develop a full end-to-end solution for a narrow set of decisions that have to be made. We were able to get that product going, but what we've landed on to get to scale and really we see the power of this platform once it's in the hands of other individuals who are able to build the models themselves, are able to address questions that they're the subject matter experts in, they can build their own simulations. And that movement, we actually had a licensing fee, a subscription to the synthetic population that we have, and the pay-as-you-go compute usage model. And what we found is that there's friction between the user and the buyer, and so what we ended up doing is saying we're going to drop the licensing fee, we're going to provide the synthetic data that's a data product of our synthetic population of every person in the US, everything's especially located, and so people pay for the data product and then they pay us the go so that the CFO isn't committing upfront to, well, how much computer we actually going to use, or we're signing up for $50,000 here, but what if we only use $25 of that? Like does it get renewed? So to make it simple for the user and the economic buyer, we move to, you pay for the data, and as you run your simulations, you pay as you go. And that has enabled us to get into, it's helped us generate more traction that way. Then our long-term goal is in ten years we're at 10,000 users globally building, running their own simulations. There's synthetic populations not only for the US, we currently have some for Canada, for a couple cities in Europe, but globally to be representing populations, to understand behavioral dynamics, understanding the emergence of different disease, and how these different systems are all connected.

John Schufeldt:
Does the synthetic population change in real-time as a result of what's already occurred? So this is why I'd be a bad example, but I'll try to wing it. So of all the people that have COVID, X percentage of them now have chronic lung disease, chronic heart disease, does a model take into that account? So five years from now, ten years from now, the risk of the rate of heart transplants and lung transplants are going to go up.

John Cordier:
All of that is possible. So when you, when we run the models for COVID, you're able to see by age, race, sex, household, and with co-morbid factors. What is the impact in that entire population? And then you can say, all right, here's the actual level of immunity. Here's the actual morbidities that we're seeing going forward. And then you're able to take that and say if you can make the connection between what's happening with COVID to any of the heart transplants, heart issues, cardiovascular disease issues, whatever it might be going forward, you can start from here's who we know has COVID, long COVID, whatever it might be, and then build on co-morbidities from there. So there's companies like MD Clone, Integra, and a couple of others that are really looking at patient-specific data and creating representative data sets just from that. What they're doing is mimicking a data set. They're not looking at how behaviors are changing and what impact that might have, but we can do is overlay all of that information onto our synthetic population and begin projecting that forward, and then include the behavioral interventions that might say, well, if somebody has this co-morbidity going forward, if we can intervene this way, we might be able to prevent that longer-term condition from happening in the future. So all of those things can kind of bring what they might say, precision medicine to the table through synthetic data. We really look at this from a precision public health standpoint.

John Schufeldt:
I wouldn't say it's precision data. It's precision medicine to the masses.

John Cordier:
Exactly, yeah.

John Schufeldt:
That's super cool. Man, if I was a hedge fund, I would grab this like none other. Because imagine what you could predict if back in January, or even better, back in October of 2019, you said, you know, I'm going to short the airlines.

John Cordier:
Yeah, honestly, I hope some of them are listening right now and would be glad to engage, so.

John Schufeldt:
Yeah, that's, in fact, I want to buy it. What would you tell, I mean, you kind of came up through an interesting path and you literally had a baptism by fire. What would you tell aspiring entrepreneurs? What have you learned through all this?

John Cordier:
Yeah, I think the biggest thing is get into the market as soon as you can, because you're not going to know unless you're working with customers what's working, what isn't working. You can spin up all sorts of things within your team internally, until you're getting external feedback, you're really not validating much of anything. So even if you think it's a little early, there are going to be people who want to champion this, sort of, whatever your project is, get in front of people as early as possible. And the next most important thing is making sure you have the right people on the team. And the best advisors I've had, have been people that have helped me look ahead and make connections to people that help them see around the corner. So from my position, I had no idea how to even start a company, what any of that looked like. Six years ago, those first mentors at the Innovation Institute at the University of Pittsburgh, they were my early mentors. They helped me find the next year and the next year and the next year. And when do you find people who've gone through it before? They really just want to help because they know how difficult it is. And the other thing is just being very open and honest about what the challenges are, because I experienced this with my own team, if challenges aren't surfaced, if people hold back, you can't really lean in and help out as much, and I found that to be frustrating with my team. And then I looked at my own behaviors and actions saying, Well, if I'm not doing this with my partners, my investors, that's just something that, the earlier you can overcome. Oh, it's not it's not good enough yet or oh, I want to hide this thing. It's, don't, avoid that. Just got to be open about where things are at and seek feedback early.

John Schufeldt:
Well, it's interesting. I mean, your minimum viable product was not minimal at all. I mean, it was it sounds like full-scale. How much did you iterate once it was already up or was it relatively, what you had was sophisticated enough to serve the needs?

John Cordier:
So April 2020, our first customer, that's pretty much what we had licensed out. Little improvements here and there, but pretty much the same product that we had licensed out of the university. They were able to take it out of the box, make use of it right away. But the challenge was the level of sophistication that that user had to have was pretty high. And so when we use the phrase and I really like it, like precision public health to the masses, in order for that to happen, we had to do a couple of things over the last few years. The first was, make sure that our platform delivery method was scalable. So of course software works, but if it's on-prem installations all the time as we're updating and making improvements, we'd have to go back and do that at every location. So how do we get to a more scalable model that as we improve, everybody else can get the same sort of tools? So making the migration from high-performance computing centers to running everything in the cloud with AWS, that was one thing. The next was, how do we enable people to visualize and better interact with the data? And so we used an off-the-shelf tool called Jupyter Lab, and we built everything so that people can build, visualize, send off models all from one interface. And so what that did is it enabled the user experience to improve. And now we can start, we're training non-data scientists on how to use the software. So there's people in public health, people, even in marketing who are saying, Oh, well, if I can just have these parts of a model, I can put it on my population, run a simulation and understand what's going on. So the out-of-the-box nature of where things were in 2020 compared to now night and day, and that's been a really, that is what will lead to the quick kind of flywheel effect of building out more of a platform where we have the users and then those who are benefiting from it. And over time, the same way that there's marketplaces for all different types of data exchanges, but at the end of the day, like you can look at Excel, it's probably the best example of this. There's people that build entire business models off of Excel. They sell their models, plug and play whatever you want. When you move into more of the data science machine learning side, it might be a little more sophisticated, but the core principle of the business model is the same. And so what we're trying to do is make sure that academic users, researchers, they can play with the entire sandbox, contribute their models, and then people on the market side can come in and say, Oh, I want to license this one, I want to lease this one for this amount of time. And we're creating both a model marketplace and the synthetic population marketplace as extensions of what we already have.

John Schufeldt:
Did you get, at your level and people probably didn't even know you existed, but did you get any backlash from the naysayers? And what I mean by that is, you know, I'd hear people say they look at these predictions and they say, Oh, that's crap. That's a conspiracy. That's you know, I just had a guy recently who's a physician, embarrassingly enough, said, Oh, you know, I read Robert F. Kennedy's book and, you know, he says, blah, blah, blah, blah, blah, which is this is all a big hoax. And did you get that at the data scientist level, or is that, where you guys are immune to that? Because you were really the wizard, you were the Wizard of Oz, that you were behind the curtain still?

John Cordier:
So I think there's like almost like a religious war on the data science side between stochastic modeling and just like statistical modeling. And so, yeah, there's a bit of that that we came up against. But then when we were actually getting what we had in front of a decision maker that had to make use of the data to inform a policy or take action, they're like, Well, I at least have been able to think about the problem differently. It isn't just here's the statistical model and this is the data, and I just have to react to it. It's like I can see now how my decisions impact what the data shows and that sort of thinking, it totally opened up the minds of these people, mainly in the events industry, where they were able to see, I make these decisions, here's how it impacts my entire business. My entire business is gathering people together and using Disneyland as the example of, if you're trying to gather people together, safety is the number one thing. So Disney is all about safety for showing up at their parks, it's a prerequisite. And so people in the business or consumer events industry, they realized if people don't feel that they can gather safely, we don't have a business. And using data to support that notion, they feel more credible, they feel more empowered in their role, and that's just one example. Our goal is to make sure that this software gets out. Other people are able to use it, and that's the experience that other people can have.

John Schufeldt:
But they also have some backup because they can say, here my decision, here's why I made that decision, here's the science that supports it. I mean, that's got to be a huge insurance policy for them because now they're not winging it.

John Cordier:
Yeah, I think for far too long people have just been winging the challenge that. The second slide in my investor presentation deck, or third slide, at this point, it's really decision-making at a strategic level for a long time has been, let's look backwards, and then let's just trust our gut, and now it's like, well, if this is what your gut is telling you, you're really making different assumptions. Well, now you can test out those assumptions of what the future might look like, and it's not going to give you the exact answer, but at least enables you to understand, I make these decisions, here's how it impacts what outcomes I'm hoping to see. And early on, when we looked at uses of our software over time, and we're coming out of a school of public health, so health has always been our number one thing. That's what, I guess another piece of advice going back to entrepreneurs, like make sure that you feel connected to the work that you're doing because on days that it's tough, you need to be able to dig deep. So like the health component and the idea that we can improve the way public health is practiced, that's a big driver for me. But people will use our software to create a better future for healthcare, create a better future from a finance standpoint. People certainly try to use the software to understand, well, how do we get this distribution of votes? All of those things are how the software is going to be used. But it gets back to, our first intention was, how do we enable people to make better decisions that are going to impact millions of people around the globe? And this type of decision-making and thinking really can make an impact.

John Schufeldt:
Yeah, I mean, for no other reason, and there's a ton of them, but if for no other reason, it should allow you to sleep better at night because now I'm not relying on past experiences and intuition, It's like, okay, I've got all that, which is great, like you said, and I'm running the model and see what it comes out. I think for a while, I was talking to a politician and I said, I don't think people realize that some of these decisions they've made have left blood on their hands. And I know they didn't think that way, they thought of their political ramification or what the populace thought was the best thing to do, but at the end of the day, a lot of their policies kill people.

John Cordier:
Yeah, I mean, this is something that my co-founder, Don, and I, we discuss it all the time. There's that triad of health, dollars, votes, and at the political level, a lot of times those things aren't all congruent. And you're going to say, well, this decision definitely shifts more towards the political decision rather than the health decision or the monetary decision and being able to understand the implications of that. I think when that tool, when you talk about giving precision public health to the masses, can really even go a step further and say, we're enabling people to understand, we're represented in all of these data sets, and the political decision maker can't hide behind, Oh, I didn't know it was going to happen that way. And so now it's like, well, we have data to understand how we can help shape and form better futures for many other people, so.

John Schufeldt:
I mean, Epistemix says no dog in the fight as far as the political or the votes go, it's like, hey, here's the data, do with that what you want. But you're, right now there's nothing to hide behind because now I say, well, you know, actual data was out there. You did have access to it. You chose not to look at it, chose not to utilize it. Okay, but there's ramifications to that, whether it's your votes or someone's blood, there's ramifications to it.

John Cordier:
Yeah, and the other side of that is there's groups like Rand and Miter and a whole bunch of other research firms. They're all using different tools, and the challenge there is it doesn't enable people to have the same type of conversation. And so one of the things that, by having the synthetic population that represents every person, every school, every household, every workplace, it basically enables you to have the conversation around the same data set because everybody's there, everyone's represented. You can say, well, here's what happened in Atlanta. Let's look at what those policies, if we drop that in on Charlotte, or drop that in on Phoenix, or drop it in on Seattle, what might happen? And then you can test your assumptions and enables people to have a conversation from the same source of data and in the same type of modeling language. Because a lot of the challenges from different modelers, it's, they might have a brilliant model written in C++ or Python or something else. The decision maker, they're not going to go into the weeds at that level. I think the gap in understanding from policy decision maker, they have an entirely different career trajectory than the data scientist, and ... the top of their field. But how do you bring those groups together? And so we talk about this concept of campfires, so bridging the qualitative and the quantitative to come together on the same platform can then inform a policy decision-making much better than just saying, Well, our data says this, our data says this. There's no way to compare across models...

John Schufeldt:
Right, now we have a fixed data set. It's not fixed at all, it evolves, but we have a fixed data set that we all can use. That's amazing, well, John, congratulations. This has been really informative for me. And I know we have your pitch deck and we're taking a look at it, but it's really impressive. And like I said, while we were bowling, this is something I'd thought about years ago and I thought, this is so important and someone a lot smarter than me had to do it. So thanks for doing it.

John Cordier:
Yeah, really glad to be on. And somebody wanted me to give you a shout out Eric ... from Nashville, from Project Healthcare. So he and I met this past, I guess a couple of days ago from a connection at the Population Health Colloquium in Philadelphia. And he said that you were a speaker with one of his cohorts recently and just wanted to say hi.

John Schufeldt:
Thank you for that. Yeah, I think I owe him an email or two. Well, John, that's been great. Folks, we'll have everything in the show notes, including the transcript of this and ways to get hold of John and follow what he's doing with Epistemix and on LinkedIn. So John, thank you very much.

John Cordier:
Yeah, thanks, John. I appreciate being on.

John Schufeldt:
Thanks for listening to another great edition of Entrepreneur Rx. To find out how to start a business and help secure your future, go to JohnShufeldtMD.com. Thanks for listening.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you'd love including automatic transcription software, automated translation, automated subtitles, share transcripts, and easily transcribe your Zoom meetings. Try Sonix for free today.