Embracing AI Organization-Wide: Insights from Theragenics

Stewart Gandolf
Hello. Welcome to our podcast. This is Stewart, Gandolf. And today, I have Michael Krachon, and we were just giggling offline about how hard Gandolf is to pronounce as well as Michael's last name.

So, by the way, I've been called Stewart. Nobody gets either. It's like Stewart, Steven, Stanley Gandolf. They just don't get it so at least you have one common name. Michael. Anyway, Michael is the President of beyond having a hard to pronounce. Name like I do, is President of Therogenics. Welcome, Michael.

Michael Krachon


Thank you, Stewart. It's a pleasure to be here.

Stewart Gandolf


Yeah, this will be fun, I can tell you. And I are like minded in many ways, CEO, to CEO and the sales process and the marketing process. So, this will be fun, Michael, for those people who don't know Theragenics can you start by just giving a little sense of your company and what your role is, and kind of where you guys are, and then we'll jump straight into the topic.

Michael Krachon


Sure. So, Theragenics is a medical device company that focuses on radioisotope production. And our core product is a radiation oncology product for prostate cancer called brachytherapy. We've been around for about 40 plus years making this treatment. And over those years we've probably treated almost 300,000 men for prostate cancer. We make a product that's about the size of a grain of rice that's implanted into the process and provides a very targeted treatment. Very targeted radiation treatment for their cancer. So, it's very individualized, very targeted, very precise as to what that is, and as part of our manufacturing and marketing, we end up creating that product direct for each patient that we treat. So, it's an individualized treatment almost like a drug prescription for radiation for each patient that we work with. So, we're very used to working at customized solutions and working directly with our customers and with awareness of patients on the other side. So, we've had to over the years, do a lot of marketing. Do a lot of sales do a lot of manufacturing optimization to make sure that we're able to get product exactly where we need it when it needs to be there.

Stewart Gandolf


Okay, so that's a that was a great opening. You know your story well, which I would hope and expect.

Michael Krachon
I

Been doing it a long time.

Stewart Gandolf


Yeah, the brachytherapy. It's funny. A friend of mine who's a neuroscientist, very knowledgeable on just about anything to do with healthcare, had prostate cancer. And he chose brachytherapy, which I think is interesting. And I didn't realize it was that personalized. So that I love doing these podcasts. I learn new stuff really, every day every time I do it so well that really lends itself. And I can see why you decide to embrace AI organization wide. You have plenty of complexity as it is. And so that's our topic today is really about embracing AI, not just in a little way, but in a big way. And so you know, I'd like to you know, to the extent that it's relevant. Tell a little bit about like how you came to Theragenics and your AI journey from there. So whichever order you want to start.

Michael Krachon
We'll start with the Theragenics journey because that's a little bit longer. The AI journey is kind of a nascent opportunity that we've kind of dug into recently. So, I've been in, came out of school as a biomedical engineer, did product development work and spent a long time at a large company called C.R. Bard, also in the brachytherapy world. So I've been working with radiation oncology, cancer treatment and specifically prostate cancer treatment for about the last 25 years. Over that time kind of graduated from product management to sales, training sales support, and when I left C.R. Bard I took a head of sales and marketing role with a company called Isoray, and that enabled me to kind of leverage a lot of the international sales that I've been working on some of the sales training and the relationships I had with clinicians and facilities to try and help grow the procedure a little bit more spent about 4 years at Isoray, got to the pandemic, life changed a little like it did for a lot of us at that point in time, and I ran into an old friend of mine, who was the prior CEO at Therogenics.

We started having conversations about what we could do to add a little bit of new world to where Therogenics was. They had been around at that point for about 35, 38 years, and had been kind of on autopilot for a while. So, I was able to come in. We talked about it a lot and talked about. How do we turn Therogenics into looking a little bit more at the future and looking at new ways to bring new products in and kind of recapture some of the growth and excitement that they had had in past years, so came over to Therogenics, spent a year doing kind of business development stuff, and ended up. Now the President of the business that goes direct to hospitals, clinicians, and consumers. We call it our interventional business, and it supports not only brachy, but some of the other direct line products that we have. So, with that in mind, I started to put the right team figure out how we re-engage Therogenics in the market. Try and find ways to get growing a little bit more. And last summer, after many of us had started hearing more and more about AI, and where that fits in, and the power and the value that it might bring to different centers, realized that I don't know a whole lot about it. I moved into management and wasn't using it too much in my day-to-day. And I said, if I want to stay relevant, if this organization wants to stay relevant, we're going to have to change that a little bit. So, I started last summer looking at ways for us to start integrating it into our organization, and as a proxy for that forced me to integrate it into my day-to-day activities, because if I can't use it, I it's hard for me to push the organization to use it a little bit more.

You need to turn your mute off.

Stewart Gandolf


Thank you. I was getting some background noise as, CEO. You can't just talk about this stuff. You actually have to walk the walk right.

Michael Krachon


You. You do and it. And it's funny because I'd spent so much time doing sales training I had, and organizational training. I had some real thoughts on how do you bring in a new technology and make sure that the entire team is using it? Right? You know there is, there is one thing about you know, putting a new tool in front of people and saying, here, go try it. There's another thing in integrating something in a manner that really encourages the team to use it and start leveraging it a little bit better. So, I wanted to kind of bring in that that new tool adoption phase, almost like we do with a new product when we send it into the field and get the entire team using AI in a different fashion to, you know, help us all be a little more efficient in what we do day to day.

Stewart Gandolf

So that you know, when we're recording this now, AI has come a long way, baby, very fast, but you're a little bit ahead of the curve there. So how did you know that was going to turn out to be a good idea? Because that's you know, you're brand new. You want to show everybody you're doing some new stuff. But did you have that much confidence? You did the due diligence process? How did that go.

Michael Krachon

I think it's both. I think one is just with how rapid AI became a part of what society was using, realize that it wasn't going away right. We had heard historically little bits and pieces of it, but last year it felt like something was dramatically different in that, right, you know it was coming ChatGPT had come out some of these other things, and it was going to be a part of what we had to do. And at the same time, as a you know, a smaller organization. I realized that we were stressed for resources, and we needed to find out whatever hints and shortcuts we could make to get things done, because, you know, we were not in a place where we could hire a lot more staff. We had to get a lot more out of what we had. So yes, it was a little bit of a risk, but I really think it was you know, a calculated one. And you know my, my background in engineering and getting used to new products and product development and setting things out made it seem like this was just a natural progression as to where we were. Gonna go.

Stewart Gandolf


Love it, love it so. You chose a specific vendor to be the center point of everything. Valkai. I'd love to know more about that, because that again, is something that a lot of people aren't doing. A lot of companies are. Some are doing it more enterprise and picking a single tool. Others just let you know it's like chaos. Everybody does whatever they want, and somewhere in between. So where? Where, and how did you choose your vendor, and what drove that process.

Michael Krachon
Well, it, it really drove it a little bit with my own experience of getting started. So last summer I said, I need to do this. I tried a couple of different tools out there. I realized I was really just learning any one of them, and I wasn't good at any of them right? So, I was looking for something from my sales structure. Looking at how we introduce new products, I wanted a system or a process to say, All right. This is the tool, and this is how we use it. And you know, and some of the other things enterprise wide. We had done. Theragenics had brought in an ERP system a couple of years ago and half integrated it. We looked at Salesforce and some other tools out there to help the sales team be more effective. And you know, I found that, you know, with some of these tools, when you engage the team with a you know, a real structured process. You found that they adopted it a lot easier. And so, with that in the back of my mind, I wanted to figure out a way. All right. How do I give my team kind of these set of tools and then give them the resources to focus and adapt it. And it's turned out, I think, to be pretty well. I've talked to a lot of my colleagues out there, and as you suggest what it is, it's the Wild West is how people use it. You've got one group using CoPilot. You've got one using ChatGPT, somebody's using Gemini somebody's, you know. They're all using it in different ways. But they're not. They're not able to translate those benefits quite as easily across the team. And you're not really getting reports on how people are using it. So, you can help those who are not using it start to incorporate it a little bit more. So, you know, those two pathways have made me feel pretty good about this initial decision, and I think it's a really useful way to try and bring something in is you know. Give people a structure to use it, and you know, then help them make sure that you hold their hand through the process.

Stewart Gandolf

Great. I love it. Very good, very systems oriented mindset on that. How did implementation go for you in the real world. Was it as easy as expected, or how?

Michael Krachon

It. It's actually in some respects been a little bit better. It? The initial part was challenging. Why do you need to use all these different tools getting it through the internal side. Why do we need to pay for something else? You know we're questions that were all there, but once we navigated those we did it in a, you know, kind of aligned rollout, and again I use some of my background in sales training to engage a couple of people get them going and then expand it one piece at a time. We didn't try and roll it out to the entire team at the same time. We wanted to get a couple of wins and then let everybody build on those wins. I think what you learned when you did a lot of sales. Training is, if you had salespeople who want accounts, they in their first 30, 60 days they'd be a lot more successful long term, but if you had 4 or 5 people coming in they wouldn't all win an account in those first couple of days. So, the ones who were not engaged. It became harder and harder to engage later, so we kind of took it. One or two people at a time got them engaged, rolled it out to the next one. Got them engaged, and I'll be honest. It was great to have a partner like Valkai, who was willing to work at our pace, and willing to kind of, you know. Take that one step at a time, instead of pressuring things to just do a full implementation. I think that that really made a big difference, because, you know, we got into some of the marketing people and they started using it. And they started talking about using it and showing the results of what that was. And then customer service got involved and then sales got involved. And now we've got quality regulatory engaged. We've got finance starting to be engaged. So, you know, we're getting to where it's a full organizational tool that people are using.

Stewart Gandolf

I love it. And you know, it's funny. I've mentioned on some other podcasts that we have something called AI Innovation Club every month. And what we see is there's definitely early, not surprisingly, there's early adopters. There's mid majority, you know, early majority, late majority laggards. Do you guys still have laggards. And is that something that you just feel they'll come along eventually and like, how is that part of it?

Michael Krachon

So, it it's funny some of the people we originally thought would be laggards. Are the ones who have become the leaders a little bit more because they've realized where it is, but the ones who are not using it quite so much. I have you know, periodic conversations with the team at Valkai to understand who is there and who isn't the ones who are not using it. We either have Valkai follow up, or we have an internal conversation to understand. And then we have some conversations about some specific challenges they're having that maybe AI can resolve instead of saying, just use it. We have ongoing conversations of what do you need to do? What are some of your problems? What are the things that are taking up a lot of your time, and then we work with Valkai to help try and develop some solutions specifically for that. So you know, I think one of the challenges about being a small organization is you've got to find the right partners, and a lot of small companies end up working with big companies that don't take as much time to help you navigate through these processes, and you know what I was looking for was somebody who was gonna help take the time. We took a risk with them as a little bit of a start, more of a startup, and they took a little bit with us, and it's been a really good partnership to make that work. But you know, we then communicated through that process and continue to communicate through that process to make sure that we see that adoption continuing to grow, we measure hours of engagement per week. We measure number of engagements per different individuals. And then, as we bring new groups in. We connect them to the people who are using it. Talk about some of the tools that are being used and build incrementally. So, we're now probably about four months into it, and we're starting to see some real advances. The marketing side is the easiest place to see it, because it helps us create some different materials and allows the team we have to really focus on the value. Add, instead of the bulk work to create some of these things.

Stewart Gandolf

So what? I'm guessing. Some things went easier, and some things went harder than expected. Can you think of any anecdotes that some of our listeners might be able to learn from in their own company.

Michael Krachon

So, I think there were a couple of things that it is. One is the open partnership to help us get our data into their system. So, one of the really exciting pieces is we use kind of the overall level of AI. But then we've got our own database. So, it pulls from. So, it's got our IFUs, our marketing materials, our clinical articles, and our different pieces. So, the easy part became customer service, who, when they would get questions instead of having to deal with their reference documents or anything else. They could just type in the product name. And the question about the IFU, and it would pop up, and then it pop up with a reference as to where the IFU, where it was pulling it from in the IFU so, customer service. Now when they get a customer question about maybe activities that we can do or product configurations or things like that, instead of having to bother two or three people to get a confirmation of the right answer. They get an answer direct from direct from our resources that directs them as to where they can, you know, provide that direction to the customer. I think the harder part has been with the sales force a little bit and getting them comfortable trusting the data that comes out and being able to help them identify new targets and pieces like that. So, we're starting to use that with, it's integrated in with some CMS data with some other tools to help us identify where clinicians are. It helps with, you know, day planning and things like that. But that's a little bit slower, because, you know, we've got a pretty experienced sales team that's used to doing things their way and getting them to adapt to something new. It just takes a little repetition.

Stewart Gandolf

Yeah, that's a classic. It sure is. Your sales team is used to doing things their way. They want to do it their own way. No, that. Never heard of that before.

Michael Krachon

Never, never, never.

Stewart Gandolf

So unusual. Now, that's a that's a that's a thing for sure getting the sales team. Not, I mean, you know, basically in sales. The culture is. When you're successful, you're successful and you figure it out. And then you're really committed to that, and everything else is kind of everything else. So totally.

Michael Krachon

And that next level of success is hard to sometimes envision. Right? You know, when you're reasonably successful and you're comfortable going and taking a little bit of a risk to do it differently. To do more, is a bit of a challenge at times.

Stewart Gandolf

So, it's really funny within our own business when it comes to that sort of thing. Even things like just researching prospects is so much faster, and the team figuring out what the hot buttons might be, and some of them are already creating agents to do some of the heavy lifting that's common for everything. So, it's like we'll get there.

Michael Krachon2

No, I.

Stewart Gandolf

Like the lead salesperson wins that way. Then everybody's gonna follow.

Michael Krachon

I think the interesting thing about the way we're doing it as well is the trust that's built up in pretty quickly. I don't know about you. But the first time I went on an AI thing you know, the first thing you do is you ask about yourself. Then you ask about the company a little bit, and the first response was, our company had been sold 15 years ago, so that immediately, and we haven't just for the record immediately. Kind of takes that foundation out a little bit more and makes you say, All right. Well, how do I really know these things? And when one of our clinical specialists started using it. She started doing comparisons between two or three different AI systems and found that you know what we were doing was significantly more relevant to her searches and her other things, and then with it coming back with references, just gave her an amazing confidence in the product. And you know, when you know one of your senior clinical people is that confident that rubs off, and when she's willing to talk about it, and just you know that again: Wins help everybody, and you know that kind of goes back to the sales side as well. When you see people winning, you want to take advantage of their winning in your world, as well.

Stewart Gandolf

Great. So, in terms of you mentioned sales or marketing, or just broadly, like what are some of the tangible benefits you've been, you can now point at and say, okay, these are really the things that move the needle for us.

Michael Krachon

Its ability to help us, you know, draft materials and reference them when it drafts them and allow us to focus on doing the last level things, its ability to help us with scheduling social media and content in different pieces like that. So, if we want to talk about a recent clinical study. It can put that whole campaign together in seconds allow our person to focus on refining it and going from there. And it comes with the references. So, when we go to quality regulatory, the references are all set up as well, so it saves a remarkable amount of time like that on some of the Q. And now we're starting on the QA. Stuff. The same sort of thing we can go in, you know. Do clinical references do other data references and collect the data, you know, faster than we could otherwise. I think the best thing that I was told when I was learning about AI from the group was, AI is really your own set of interns who do all the grunt work for you or all the basic work. So, you can focus on the real benefits of what's next so, or the real value added to the process. Instead of doing all the all the research that underlies it.

Stewart Gandolf

That's so true. I love that metaphor particularly like with our SEO team, it gets so technical and so complex. And some of this stuff to do is just really hard. So, I have a senior person doing this really hard, mundane stuff. Now, it's not like we're doing less for our clients. We're doing more, faster because we're spending our time on the valuable stuff, not the other stuff just needs to get done. And so much of the stuff that I'm talking about here is very hard to scale and very hard to teach, whereas the machine just gets it right away and makes it so much easier.

Michael Krachon

And it. And it doesn't have the same bias that a user might have. Right. So, when I came up through product management, if I were creating materials, I would go back to the clinical articles I knew best, right? So, you would pull those in. You'd add kind of the reference to them. This looks at all the data, and if we've got more things in, it'll pull a reference we may not be as familiar with, or a data point that may not be there and gives us an unbiased version of the data and may inject some things we hadn't thought about before. So, you know it's a real neutral perspective, and it doesn't rely on me remembering the four clinical articles or somebody else on that, it actually goes back and finds them so it's in that respect, I think it gives us a new perspective.

Stewart Gandolf

So that's again, my experience is consistent when you know I've been a writer for a long time, and I'm pretty good at it. But even when I go back and promote a new webinar, for example, and I'll take my draft and then put it through AI just to double check. There usually be 2 or 3 things that I thought, oh, I should have added that like that you know, it's like we don't have a checklist in our brain for everything we've ever learned in our entire life, right? It's really hard to recall that. And it's really helpful for those kinds of things. So, I'm still driving the ship. But it's nice to get some input from an expert friend. Do you feel like this is going to be a competitive advantage for your firm. This reliance, compared to some of your other colleagues and competitors.

Michael Krachon

I think it will start off at a competitive advantage because we've adopted it early. I think it's gonna be up to us to continue to find ways to innovate, how we use it in the different pieces. AI is not gonna go away and everybody's gonna have access to some element of it, I think. How do we customize it to what we need to do? How do we focus our team on how to use it best. How do we teach them how to use it best, so we can stay ahead of the curve. That's where we're gonna preserve our advantage. You know, because I it's like I said, it's not going away. Everybody's gonna use it in different fashions. Everybody's gonna have access to some of the same core data within it. You know, it's how do we get the most out of it? And you know, I think we do that by embracing it and using it, and learning from it, and innovating, as opposed to just, you know, taking the can portions off the wall and saying, All right, let me use this for this, or let me use that for that.

Stewart Gandolf

Well, you know, I think that it's funny. I don't know if this is true to you, but if you're trying to develop an innovative culture. You have to be innovative. So, this to me, would be part of that right. This is just one way you're expressing an innovative culture. And it's proof of that. And I feel like you're right, like people will start catching up. But that just means you have to move that much faster in terms of how to use it. And I, you know, based on my experience with humans, a lot of people will never fully adapt it. And then the other thing, too, is, you know, people worry about, you know, justifiably, jobs. But so far, at least in our company, it's not. It may keep us from we may be able to do a lot more with less, but that just means our people are going to be doing more. So, in the short term. I don't see any problem with that in the long term. It is going to be hard, for you know, for example, even very you mentioned entry level interns. So, if you don't need them, they won't. You won't need them. So, it'll be really interesting to see how that impacts. So, I guess the last question is, as somebody is thinking about this topic. Most of our listeners are experimenting with AI at one point or another, but you know, what advice would you give, based on your journey, either by choosing a single source for AI, or just where your wins have been? Or you know, because I feel like you're still ahead of the pack, broadly. What are some of the insights you'd like to share as we wrap up.

Michael Krachon

So, I guess what I what I like to kind of share is, you know, think a little bit, and have some kind of vision of how you want to use it as an organization in your head before you do it. If you can build on what individual experiences are. That's great. But figure out a systemic way of how you wanna push down the organization, and how they're using it. So, they can continue to leverage each other and build on the tools that are there, and you can help them identify what tools, as an organization, you need. As I said before, people are using. And I came from a conference where people were rattling off 6, 8, 9, 10 AI tools that they were using. And you know, it's really hard to get good at 8 or 9 or 10 things along the way, and it's really hard to teach an organization of a lot of people to be good at 9, 10 things along the way. So, identify where you want to get to with that in mind, then, you know, find the right partner to help you get there and understand how you can measure and look at the progress of the organization. I don't think this is different from doing any other process improvements that you might have looked at in an organization. Most places do it on the manufacturing side, where you call it 6 sigma, or something else, where you continually try and get better, AI and marketing has to be a little bit the same thing. How do we iterate our processes to get better leverage our tools a little bit better, and, you know, put kind of that oversight or that perspective into it. So, we can, you know, help the team get there. You know, my role is to, and most people in leadership, is to help your team be more efficient and more successful. And sometimes that's because you see things, and where you're going a little bit better than they do, you can nudge them into those lanes instead of having them figure it out themselves.

Stewart Gandolf

Great. Well, I think this is a good lesson on AI, and it's a lesson on good leadership. So, hey, my Michael it has been great. I appreciate your time. Good job! I knew it was to be fun. Thank you for your time.

Michael Krachon

Oh, I enjoyed it very much, and I appreciate you just looking into the brachytherapy and the radiation oncology world a little bit.

Stewart Gandolf

Yeah, for sure. Thank you.

Michael Krachon

Thank you.