Hey, guys. Welcome to the podcast. My guest today is doctor Sanjay Singhvi. He is director at VMLY and R Health, and we discuss how artificial intelligence and machine learning will affect the role of the MSL. Really interesting conversation. Don't forget to follow me on LinkedIn. Check us out on MSL talk live, which is the 1st Tuesday of every month at 1:30 PM EST. And then follow me on TikTok and Instagram as well. So thanks for joining us.
Welcome to MSL talk with Tom Caravella, a podcast specifically designed for MSLs and all things field medical. Hey, Sanjay. Welcome to the podcast. Thanks for joining me, my friend. No worries. How are you? I am awesome. I'm awesome. And I'm excited to do this because I love this topic. I think it's gonna be great. And I appreciate you joining me. No worries. It is an exciting topic. Absolutely.
Yeah. So just to give you guys some background, you know how I like to talk about how these things come about. So I met Sanjay at a conference in Philadelphia recently, and he spoke on this topic and was amazing. I I literally listened to the whole thing, and I was like, man, this guy is super smart. This is a great topic. Haven't covered it yet on the podcast. And, you know, jokingly, he kept saying, you know, I gotta come on the podcast so you can make me famous. You can't tell me.
So I don't know about that, but here we are, and I'm gonna just let him introduce why don't you introduce yourself, and where you're from, all that good stuff, and then we'll jump into some of the content. Sure. Sure. So my name is doctor Sanjay Singhvi. I'm, one of the directors in, VMLY and R Health, and I tend to focus on product innovations and strategies. So AI, machine learning, all that interesting stuff is definitely in my orbit, and, I very much enjoy that.
Yeah. And it's, it's something that I don't know a lot about, and I I know that there's a lot of people that are curious about this topic. So Sanjay is an expert. And, actually, I wanna announce that, VMLY and R Health is the sponsor for this episode. So, just to learn a little bit about these guys. They're a global agency united around a single purpose of improving health all over the world by making science meaningful.
So for more information on these guys, definitely check them out, especially their digital platforms. And you can visit them at vmlyr.com/health/digitalsolutions. And look out for that link in the announcements on LinkedIn when you see me posting this so that you can make sure you're going to the right place.
So let's start with, and I think that for for novice people like me, I'd like to kind of know, like, when you talk about AI, artificial intelligence, machine learning, what do we need to understand about how that actual technology works? Like, do we need to know the nitty gritty and nuances? And what's the difference between all of those? AI, machine learning, deep learning, NLP, etcetera. Let's start there. Sure. So I would say in terms of do you need to know what these technologies are?
I'd say don't even bother, honestly. Let let let me tell you why. You know, a few years ago, when we talked about self driving cars or Tesla stuff, every article started about how many radars, lidars, ultrasonic detectors, cameras, CPU it has. But these days, you just wanna get in the car and you want it to do its stuff. Right? You we're caring less about what is the technology as it as long as it's compliant, safe, you know, validated, all of those things.
Another example, there are many there are some car parks in the world where you drive your Mercedes, leave it at the car park, and you go shopping. The car drives itself, finds a place to park, avoids everything, and when you're ready, it comes and picks you up. Now at that point, you don't care how it's done. You just wanna do your shopping, finish at the mall, come back, get in the car. And I think the same applies to machine learning, AI, etcetera.
So the focus shouldn't be on what these are, how they work, etcetera, but, really, what pain can it solve? And I think that's the important the the important thing to focus on. Because all of these technologies, we hear about stories about them across all of life, not just medicine. And I think that's enough knowledge for people to have. Yeah. These these technologies solve pains, and so how can they help me solve pains?
And, Tom, if you're ever stuck and you need to write, you know, a detailed presentation on this, I would say a quick Google search will give you better definitions than I ever can. Well, the you know, this is, like, one of those things that exists out there. We know it's there. We don't probably know enough about it. We just know that it's a thing. Like, when I was a kid, there was a show called The Jetsons. I know I'm dating myself, but The Jetsons was this cartoon. I I remember The Jetsons.
Remember The Jetsons? Yeah. Yeah. Yeah. Like, back when we were kids, they would like you know, they put, like, this little pea sized thing into a box. They'd press a button, and then they'd have a whole meal would come out. And it's like, wow. Like, that's crazy. That'll never happen. And their cars were flying and all this other stuff. And, you know, now we're at an age where, I mean, who would have thought up the Internet and all of the stuff that we have now?
Who would have thought that you would be able to send video over your phone to somebody at the other side of the world, and they would get it instantly? Yes. And they would laugh at the cat making a weird sound. Exactly. So this stuff exists. We know it's out there. There's these tools. Yeah. We don't know how they work. We don't know how it's gonna apply to us. So my question to you is, how does this become relevant and relate to the medical affairs and MSL and field role?
And what do people kinda need to know about? So as I said, we we shouldn't bother, as you said as well, about how these technologies actually work and, you know, what's the nitty gritty and the difference between all these things. But, certainly, the applications of this technology is super important. And in medical affairs, with MSLs, field medicine, etcetera, these technologies or the machine, as I like to call it, is used in many different places.
So for example, drug discovery, patent analysis, identifying KOLs, analyzing publications, managing adverse event reporting, even searching, you know, your medical information database for things that are similar. Literally, there are so many applications, too many to mention. But what is common between all these things is that they usually tend to do 1 of 5 things. They're either solving a pain. They're reducing the cycle time for a particular action.
So you had an a sequence of actions that you took before, and you wanna reduce the cycle time. They're increasing the efficiency of something or reducing the risk, the exposure that medical affairs has or pharma has. Or most importantly, they tell you something you just did not know before, and they and most things usually fall into one of these these 5 buckets.
So the examples that I gave about, for example, medical information requests and stuff, that's reduction of risks, and it's also reducing the cycle of time to find, you know, has this MIR request come in before and so on. So Got it. There's a lot of applications. Now in field medicine and MSL, again, there are a number of areas where machines, can help. And one of those, areas is insights because we know that insights are tremendously important.
They're an immensely valuable currency that MSLs bring in. They drive thinking. They drive strategy, all of these things. But given their nature, insights are essentially bits of text written by humans. So they're written in many different ways, packed with knowledge, written in many different styles, submitted by MSLs all over the world, all coming to your organization.
And that sort of environment, that sort of pain is ideal to throw machine technology at, for example, to analyze those insights. So in essence, what machines can do there is give you insights about those insights or extract insights from those insights. So that's the way, I think, to think about, these technologies.
And, yes, they have many applications, including in field medicine MSLs, but this is just one example of how they can help because the data makes it very good for a machine to to work on and help with those 5 areas I pointed out. Okay. So then there's these applications out there. There's this really cool stuff that's either available or coming available. Is there anything that the MSLs can should be doing right now?
Is it just prepare yourself for it, be aware of it, get ready for it, or is there something that they can literally do right now with some of these resources, tools, applications that can help them in their their current job? I I I wouldn't say get ready for it. I'd say it's it's there now. And kind of in terms of this technology in the current time, it's, you know, it's happening now. So I'd say a few things that the MSLs should know.
The first thing you need to know as an MSL is that you do not need to know how the technology works, and I think we've we've been through that. You don't need to be an expert. Right. The second thing you need to know is that it can definitely help solve your pains, and it can support your work. So keeping an open mind is super important. Mhmm. And I think the third point is important because machines are not the answer to everything. Right?
So be realistic about where they can help, where can these technologies help, and where they can't. And this goes back to a few years ago. There's a lot of hype around AI and machine learning. You know, it'll solve everything. I remember a few years ago, we're at a conference where we're talking about, you know, in a couple of years, MSLs may not even have a role. But that that's just that was just hype. There's there's too much hype.
People got burned from it, and so there's a bit of a backlash. But I think now we're in a much more sensible phase of using these these technologies. But the other thing, Tom, you mentioned was, you know, what do MSLs, like, need to know in the current time? And that current time is really important because the field is evolving.
So keep an eye out because it's changing every, you know, every year, there is something new, something different, increase in inefficiency, or a new pain being able to solve, etcetera. So I think those three things are important, but keeping in the context that everything is evolving over time. Got you. Yeah. This stuff's kinda almost like when I look at what might happen down the road, and I'm jumping ahead a little bit.
I wanna get back to some of the applications, but it the stuff that I'm seeing almost seems kinda scary where a lot of this automation, although it can be really useful, it almost can can do too much. Do you do you do MS should MSLs feel threatened in any way by some of this stuff? No. Threatened? Absolutely not. But excited? Hell, yeah. Absolutely. It's an it's an exciting area. So, because I think as an MSL, you wanna think, look. There are these technologies.
I don't need to know the nitty gritty of them, but they can enhance my role as an MSL from the tactical and strategic perspective. So the tactical is, what am I doing on an everyday basis? And strategic is, what should I be thinking about with a particular KOA or group of KOA or the territory I have, etcetera. And the you know, it comes back to decision making. As an MSL, you're making these decisions. You are interacting with your KOAs and HCPs.
You have an immensely strong and powerful instinct about how those interactions are going, what you should do next, what are the needs of the KOLs and HCPs you're seeing, which is great. But what I think machines can give you is some additional data points that you can use to enhance your instinct. Mhmm. Because I think, you know, often in many situations, the best decisions are made when you can combine your gut instinct with some hard data. Right.
Machines can help generate that data and give you those that data points. So if you imagine the MSL is getting more and more of this, then over time, the decision making power, I think, that the MSL has will actually increase significantly.
So, in terms of what they should be doing, what they should be saying to an expert, how they should be saying it, even the kind of flow of the communication story, what should they be looking out for, how that fits into the broader kind of priorities of the organization, and so on. So I think the MSL will be much more empowered, and the decision making power will kinda be pushed to the front line. So I think that's something that, MSLs should relish.
Yeah. You know, the example I use is, it's like the Formula 1 driver. Right? They're going super fast. They have to have the skill and the instinct to drive, but the performance is very much enhanced when they have all this additional data and analytics analytics to help them guide them. So, you know, you took a a curve at 170 miles an hour, but the data says you should have actually started 0.3 seconds before.
And so can I apply that data with my instincts, start that, and then take the curve faster? So it's that sort of and then so the decision making goes much more down to to the driver when they have all this this addition. And by the way, Tom, I've used a number of car analogies. I know nothing about cars, just to make that clear. I I do watch an inappropriate number of video videos about the f Ford, f 150 Raptors, which is a beast of a machine, and that's it. That's my limit.
So I just want an interest of transparency. And those things are crazy expensive. Those raptors, they're like a $100,000 each. Like, you can't even get one. We can just pause this podcast and then fanboy over a video in a minute about some cool raptors. Well, for the purpose of staying true to the topic, we'll we'll stay gone. But the no. I love what you're saying, and I love these analogies because it really helps paint a picture.
I think if you are you're an MSL and you're doing your your your day job, and and you're focused on trying to make, you know, improve on your performance. There's the gut instinct, and then there's the ability to rely on machines to help you get to that next level. So there's a lot of this stuff exist in like the CRM system. Is this stuff that's online? Is it I'll be honest with you.
I just started using my phone more than I ever have before, not just apps, but just to, like, there's so many phone hacks. Like, if you go to TikTok and do iPhone hacks, there's so many things that can automate how you use your phone, how you communicate with people. You can type in 2 letters and then fill out a whole phrase with this, I forget what they call it. I'll think of it. But so is that what you're talking about?
Yes. It's it's it's using all those hacks, as you call them, but to to help with your work. And this does this is, like, stuff that's happening now. This is not in the future. But it's a change of the mindset of the MSL saying, I'm gonna use this, as you said, not just to improve my work, but actually also what what value am I bringing that KOL, that HCP that I'm seeing. Right? How can I help them even more?
You know, they've talked about a certain topic a few times, and my machines are telling me that, you know, there are other people in the country that are also bringing up that topic. Now normally, it would have taken me a while to figure that out, but with machines, I'm getting that data instantly saying, okay. So I have these KOLs talking about the same issue. Clearly, that's something we need to address, not just with one KOL here or there, but as a group.
It may do with they don't understand some data or they have an issue with with with something, that you're communicating. So I think, you know and that is the data that you can use to get your gut instinct going and saying, right, this is what I'm gonna do going forward. Love it. So I just figured it out. So you follow me on this, guys. This is freaking changed my life. Okay? This is a little hack you can use your iPhone. So you go to settings, general. Right?
Go to keyboard and then text replacement. So when you go to text replacement, you can actually create a text replacement for a phrase. So let's just say you have a phrase that's and I didn't know about this until recently. So let's just say you want to say, on the phone, call you right back. All you have to do is type in that phrase and then put the shortcut. So maybe the shortcut is c b, callback. That's your callback phrase.
So once you set that up, if you're on the phone or if somebody's calling you or whatever, and you put them in, you can just send them a text and just put c b. And it'll be like, on the phone, call you right back. You don't have to type all that stuff in. So I use that for I have like 20 of these things in here now. And to me, that is kind of a piece of this, or am I just am I making this up? Am I just, like, stretching this here? Tom, I think this is for your Tom's TikTok hack podcast.
Your your your other famous life that no one knows about, but which we which we secretly do now. But, yeah, it's it's exactly these these sort of things. I mean and machine learning and stuff doesn't have to apply to you know, it's machine learning and other technologies. So, for example, you're running around a congress, and you are meeting people, seeing abstracts poster sessions, there's lots of interesting insights you wanna capture.
Yep. You know, on your app, on on your, management app, you'll use speech to text so that you just literally talk into your phone, and then it captures the insight. Now that's using a whole series of machines and technologies and all that to translate your speech to text and figure out how you talk and what words you mean and learn from that, etcetera. So it's happening, you know, all all in our lives, and you don't really understand how that works. You just know it works.
You wanna pick your phone up, say, I met this KOL. This is an interesting insight. Capture it. Send. Goes into your system. Right. Love it. No. I and listen. This is this like, a lot of these this stuff may seem like either insignificant or it might seem like, oh, yeah. Well, I already know about that. But the thing is, I know that we hear things about technology, but we don't incorporate it into our day.
At least it's that's something that it's I've been a late adopter in a lot of this stuff, which when I saw you speak and talk about this stuff, I'm like, you know, I have to get more automated in my life. Get more automated. Right. Tom, I'm like, you you need we need to separate podcasts in Tom's automation. Yeah. I need help. Lots of things. Yeah. Yeah. No. I think it's it's it's automation, but it's and it's just change of mindset. It's like, yeah.
I'm gonna embrace this, and I'm gonna run with it. What's your message to managers, and how do managers fit into this equation with their MSLs and with their teams? Like, is there an application? Is there something that leaders need to know in regard to this that maybe they're missing? Yeah. I think, yes. These technologies are definitely gonna impact impact the managers.
I mean, we talked about some examples of MSLs in looking at more data and making decisions, but managers and leaders have a series of tasks to do. Right? And so how can machines help those? And I think that's gonna be very important. So 2, we can talk about, like, 2 two levels or two examples. For example, managers sometimes have to go through a large number of insights that the MSLs have collected. They have to identify trends, do some analysis, write a report.
There is a manual component to all of those things, and that entire process can definitely be attacked, by the machine. So, for example, the m the machine can tell you, look. Here are the 1,000 insights from last month. These are the trends that I found in those insights. Now the manager still has to make a decision what's important, what's not, because the machine doesn't know your strategy.
But it's certainly speeding up, you know, reducing that cycle time, increasing the efficiency, and stuff so that the manager can click a button and say, okay. Here are the 10 trends, and these 1,000 insights. These 2 are important, and I'm gonna explore them bit of a little bit more, look at those insights, and and then generate the report. So that's an example how from a managerial point of view, those sort of actions Yeah. Can be aided by machine.
But at a at a at a kind of more resource allocation level as well, it can help. So by resource and asset allocation, we're talking about, you know, how do you distribute, your MSLs? Who should they be interacting with? And so how do you allocate your resources? Because your resources are finite.
And, again, the analogy, not in cars, but saying, look, when you have, you know, tomorrow, the delivery company has a 1,000 packages to deliver to 800 people and it has 20 trucks, and it wants to press a button where the machine will then tell them the best route for those trucks to take, which package should be in which route. You know, this the the, journey optimization software, which all delivery companies use. And they are maximizing efficiency, minimizing fuel. It's a very complex problem.
Enormously complex, but machines can handle it really well. So the same thing, you can think, how can you apply that to resource allocation as a manager?
So I'm looking at what my MSLs, how they should be distributed, who should they be working and targeting and communicating, And not just looking at number of experts in a particular territory, but also the type of experts, the type of KOLs, the history that you've had of interaction with those experts, the type of insights coming back from those experts, the importance of those experts.
And in that territory, how many institutions are there, how many fellowships, programs are there, and so on and so on. You know, machines can assimilate this data much better than humans can and then help guide your decisions. Okay. In the next quarter or the or the next planning cycle, this is how I should manage the allocation of my MSLs in terms of the territory.
So those are just 2 there are so many more, but those are two examples of how machines can help you practically with the everyday kind of tactical stuff of analyzing a bucketload of insights that your wonderful MSLs have got you, and then also strategically thinking about things like resource allocation and actions and so on. So, from a managerial point of view, the you know, it's definitely once again, something exciting to look forward to. I'll tell you, it's fascinating stuff.
I mean, it I I just feel like the early adopters of a lot of this stuff, I think, are going to catapult themselves into a whole another maybe to a whole another level. And I know that that just kind of sounds obvious, but I think that being able to factor this into strategic planning could be massive for future potential. Totally. Totally. I think, it's, you know, the early adopters, absolutely.
But with this important realization that a machine's not gonna solve all of your problems, which is what happened in the first round of stuff a few years ago where I thought, yeah, machines will understand your strategy. They'll tell you this. No. It's like, you know, look at it in in in discrete problems, discrete efficiency gains you want, discrete things you want help with. As a manager, I'm sitting down every month.
I have to spend a day going through insights, figuring out what's the can I make that instead of a day, can I do that in 2 hours? Thanks to machines. You know, that sort of that's the approach that I think managers, and leaders should be taking. Yeah. For sure. So now what about I always have to kinda shift gears back to how it, like, relates to me and my business. A lot of the listeners on this podcast, are job seekers or people that are looking to break into industry.
Does this relate to them in any way? It does. And I think, obviously, you also have listeners now looking at iPhone hacks, Tom. There'll be out of context. I think it does because, I mean, we know MSLs what's unique about them is they have to have a variety of skills and wear many different hats to be successful at themselves. That that's given. You know, they have to have the science, the communication, the the managing up, managing down, managing sideways, managing KOLs.
I mean, all of these things, so it's many hats. And for people who wanna break into this industry or who look at jobs, I think one of these hats that they need to put on is simply saying, yes. I appreciate how machines can help. Yes. I can work with machines, and I am open to these technologies to improve my decision making power. So, sure, I'll have that gut instinct and all those skills and capabilities.
But, you know, if a machine's able to identify some data that I didn't know before and tell me that very quickly, I am flexible enough to change the content of my next discussion with that k o l which I'm having tomorrow, for example, because it's it's brought up something new. So it's really as simple as that. It's just another hat that you need to appreciate, identify, and embrace.
And I think to your point, it makes you more marketable being able to show that you either have some understanding or experience or even a desire to learn. But if the more you can so there's so much that we can do now that we have the Internet and there's master classes every time you turn around. There's just ways that we can educate ourselves on some of this stuff. That if you can add some of this stuff to your resume in some way, shape, or form, it makes you more marketable.
But the other thing too is and we're and again, getting back to the question, which is, how does it help job seekers? How does it help people looking to break into industry? If you understand how this stuff works, it can give you an advantage. And I'll give you an example. Everybody asks me all the time. Well, Tom, how do I get my resume through applicant tracking system filters, which is all artificial intelligence. It's all machines. Yeah. Right? That's all machine stuff.
And if you've seen some of my videos, again, getting back to, like, LinkedIn, TikTok, I talk all about how important it is to take your resume and match the keywords, phrases, and language to that that exist on that job description so that your resume or CV is inclusive of the keywords that are identified in the artificial intelligence. There's parameters that get set criteria.
So if you can tailor your resume to that specific job description, you have a much better chance of getting through the applicant tracking system. That's a perfect example of how to be able to educate yourself, get on board with the this technically driven society that we're in and we're moving towards to give you an advantage. Yeah. Totally.
And and as a job seeker myself, when you go to your you know, I think over time, people in interviews and evaluation process and stuff will be asked how confident they are of of working with machines and using that data to guide your decision and stuff. So if you are asked that, obviously, you you're not gonna sit down and give an answer. You know? In 1964, in Stanford, the first computer doing machine it's like it's more about what I know how machines can help me with certain things.
Pick an example, whether it's insights, whether it's arriving at the next best action, whether it's, you know, resource allocation, whether it's, getting some better digital KOL profiles. How can all these these these inputs help me as an MSL? And the more you illustrate that, the more more it becomes apparent that, yes, you're you're you're open to improving not just your performance, but the value ultimately that you bring in the KOL. Right. Right.
So so now so you're, obviously, you're an expert in this space. Right? Your company has these digital platforms. You're doing this, and you go to conferences. You're speaking about all this stuff. So now you you've established yourself in in this space. What requests are you getting from, like, MSLs and from companies? The type of requests have definitely changed in in the past couple of years.
And, traditionally, you would get, you know, it would be someone maybe high up in procurement, doing a request for information, reaching out, what are your capabilities, blah blah blah blah blah. Sure. We get those, but it's also much more we get more frontline MSLs and medical affairs professionals. You know, we get, emails and calls from them saying, look. I work in this team. This is what we do. We've got a particular pain point. Right? And they'll identify that bit that's causing pain.
So, for example, they might say, we're having difficulty with, the time it takes to run reports about our insights in the context of our strategic objectives. So it's a it's a specific thing. They say, do you have any technologies that can help me? Because then I can take that to my leadership and tell them about it. So it's not that they're decision makers, but they they're you know, we talked about empowerment and this idea that they see in my team, I've got this pain.
Can machines help me solve that pain? And so we're we're we're getting, you know, a lot of that from frontline MSLs, medical affairs professionals, because it just like, I think, you know, the TikTok hack you were talking about, Tom, you you have a pain. You did it. You didn't you know, you you went out there, saw a solution, solved it. Right? And it's it's the same thing. It's like, I've got some problems. Can machines help? Who can help me?
And then they'll take the leadership and say, you know, go through the person stuff. So I think it's that empowerment and confidence that we're seeing from MSLs and medical affairs professionals to embrace how machines can solve specific pains, and that's the way we we want people to think about it. Awesome, man. It's really exciting stuff. I, I don't mean to geek out on you guys here. And, but it's I think this was a great conversation. And, guys, like I said, check out Sanjay.
Check out the company. It's, again, it's vmlyr.com/health/digitalsolutions. I think you'll learn a lot. And thanks, buddy, for coming on. I really appreciate this. This was awesome. No worries. Thank you. It is a very exciting area, and I've sent you a video about a, yellow f 150 Raptor. Of course you should. Talk of it. Don't tell my wife. She's gonna be like, there's no way you're spending $100,000 on a yellow truck. It's not gonna happen.
But, hey, boys can be boys, and, you know, we can dream about these things. But Yep. Hey, man. Appreciate you. Thanks for coming on. Of course. Everybody, thanks for listening. Appreciate you all, and we'll see you back here soon. Thank you so much for listening to the show. And if you enjoyed it, please subscribe so that you don't miss an episode in the future and feel free to leave a rating or a review or a comment. Thanks again, and we look forward to seeing you soon.