Definitive Healthcare Corp. (DH) Earnings Call Transcript & Summary

Good afternoon. My name is David Grossman, and welcome to the Definitive Healthcare presentation. With us to my right is Jason Krantz, Jason, is both the Founder and the Executive Chairman of Definitive. So Jason, thanks for coming today.

Happy to make this as interactive as you like. If you have any questions, just let us know, raise your hand and we'll get you involved in the conversation. But why don't we just start out, Jason, why don't you just take a few minutes high level, describe the business for those that may not be as familiar with the business.

Sure. First of all, thanks for coming. Appreciate it. So when I started the company in 2011, the thesis was all around a lot of people were talking who wanted to sell into the healthcare market, but they didn't know how. So what we went out to develop is this concept of healthcare commercial intelligence, which is data and analytics on every single healthcare provider across the U.S. that is utilized by companies, help them sell and market more efficiently in healthcare. Now healthcare is a very complex market to sell into -- first of all, it's huge. It's about $1 out of every $5 spent in the U.S. But the interconnectedness of government agencies and pharmaceutical companies and providers and payers, makes it very complex to sell effectively within this market. Our data helps them do that. So the companies that we sell to are life sciences companies, healthcare IT companies, healthcare providers themselves, like hospitals and physician groups and then a whole diversified set of companies which is anybody from waste management to financial organizations, banks, anybody that sells to many industries but also understands us selling to healthcare is important, and you need specific data to do that effectively.

So why don't we get the elephant out of the room first. So obviously, Generative AI has dominated the conversation in Wall Street elsewhere. And that's been an important part of your model since the beginning of time because you've been using algorithms to segment the data, curate the data, right, as well as apply [ turbulence ] to data. So it's usable, right, by your appliance. So why don't you talk a little bit about how generative AI may enhance the value of what you can provide for your clients. And then also, what are some of the competitive threats now that the technology is more broadly available?

Sure. Let me go on backwards order. First of all, we've been using data science and artificial intelligence for a long time. And this is to as you said, create new insights from our product, from the data that we have, generate new ideas of how our clients can utilize the data to drive their business performance. From a risk standpoint, I see this all as opportunity. Because in order for AI to work well, you need great data for it to set upon. And that's what we've built over the last 12 years, it's a highly proprietary set of data covering the entire healthcare ecosystem that doesn't exist anywhere else. So if people want to use that data for AI, great, we can create new opportunities for us as a result. So how we're thinking about using AI is there's a couple of different paths that we're taking. The first is internal. Internally, we are pulling together a curated set of all of the information that we know on our products, the case studies, how we answer questions for clients, how we sell to clients, and we're putting this together, putting that on top of it in order to help our scaled sales organization become more effective. So what we do is very complex. There's a lot of difficult questions to answer from clients both before the sale happens as well as afterwards. If we can make our team more effective using ChatGPT and the AI and the large language models, that's a big win for our organization. The second is put it on our data and making a client facing. So as our product and our company has grown, our data has become more complicated and clients have a harder time figuring out, all right, I know you have a lot of data but where do I find it in the product or I think there's some data out there, but I'm not quite sure where it is. So we see sort of a ChatGPT large language model as an opportunity for -- to sit on all of the data that we have, all this proprietary data and allow our clients to query this in order to get that information that maybe they didn't even know we had but is beneficial to the problem that they're trying to solve. So I think this will drive up usage over time. It will allow our clients to get more value out of our product and drive our continued growth. And that's just the beginning of the step that we're done.

And do you see any incremental usage by your customers though? Is the, I don't know, call it, do-it-yourself, but they try to apply their own intelligence to the data that's available? Or are the data sets just really underdeveloped and insufficient to really drive much value without having a third party involved?

No. Well, we hope that they put their own intelligence on top of our data. So we like that. So we want to be as integrated with our workflow as possible. So we have -- today, we integrate with all of the CRM systems with HubSpot, with Veeva, with Salesforce. We do a lot of APIs where we're pushing data to our client site and then they're using their own BI tools and other tools to analyze it further. So we want to be the engine that drives all of their decision. And if they want to put new AI models on it internally, that's great. We like that.

So you mentioned Veeva. A lot of -- one of the questions that comes up frequently is to distinguish what you're doing from Veeva, from IQVIA and there's one third one. I forgot. There's one other...

Symphony.

Yes, Symphony. How do people really distinguish what you're doing versus what they're doing?

Sure. Well, let me start with IQVIA because they're -- they've actually got a product to market. So what we've done over the last 12 years? We've built a proprietary view of every single healthcare provider. All the physicians, of course, but are focused on all of the facilities across the U.S., every surgery center, every clinic that's popping up, this is highly proprietary and doesn't exist anywhere else. What IQVIA and most other competitors have done is they've taken a view of the world based on claims data. IQVIA has done an amazing job of building a fantastic medical and prescription drug claims database, they have claims going across the globe. When you take this, though, and you try to create a view of the health care ecosystem just off of claims, what ends up happening is you miss a huge amount of number of the facilities that exist in the world. So when you overlay claims on our data, about 45% of facilities never show up within that. And the reason is claims are not always issued by a facility. They're issued by the physicians that work at those facilities. So what our data allows them to do is to roll that back up to the facility level and get a true view of the entire healthcare ecosystem. So what this means for IQVIA, they're in every big pharma. They are in every midsized pharma. But we've been -- over the last 12 years, we've been able to sell next to them and supplement the data that our clients get from IQVIA as well as other sources and make that data better because we provide the whole view of the entire healthcare ecosystem. Now Veeva, what Veeva is trying to do is compete directly with IQVIA. So they're going to go build an amazing prescription claims database. This is what they say they're going to do. They're going to build a great medical claims database, and they're going to go head-to-head with all of the core IQVIA use cases. So our competitive differentiation to Veeva will be similar to what it is in IQVIA. And we are in and with IQVIA, we don't have to be [ or ], which is important.

Right. So is the premise that what happens is that the claims data is missing the fact that a doctor could be prescribing from one facility versus another? Is that where [indiscernible] falls in?

The facilities just don't exist in the claims anywhere. And then we've -- over the last 12 years, we've painstakingly figured out how all these organizations are affiliated together. That doesn't exist anywhere else. We've looked at that longitudinally, so we know how that's changed over time, again, doesn't exist anywhere else. So we have a big competitive mode in what we provide that is highly complementary to what an IQVIA or Veeva would provide, which doesn't mean we don't compete with them, but we can be an and as well as an or.

Got it. So one of the challenges over the last kind of probably 6 to 9 months has been the end markets, right? It's been really challenging. I think you had some incremental enterprise churn last quarter, but a lot of the weakness has been -- I guess, enterprise could be any size company, right, because you have some enterprise customers that are smaller [indiscernible] providers. So net debt, what, if anything, has changed over the last 90 days or so in terms of the end markets? And how should we be thinking about what are some of the dashboard items that you're looking at to really identify -- is this something that we're going to be living with longer than we thought or we're looking for green shoots that would suggest the worst is behind us?

Yes. I don't think the macro environment has changed meaningfully in the last 90 days. I think we still continue to see macro headwinds, just longer sales cycles, CFO is not willing to make the purchase decisions. On the good news front, our top of funnel demand is better than it's ever been. We will do more demos to clients this year than we've ever done in the history of the company. And people want and need the product. However, the budgets and the CFOs are just elongating sales cycle. So stuff is sitting in the pipeline longer than it ever has been before. Generally, the macro headwinds are really across the board. As you pointed out, there's a few areas that have been a bit more severe, like small biotech has been a bit of a trouble area just with the financing crisis that they have, healthcare providers has been a little bit more difficult. I think some of these -- it's hard to say when this all comes out. It will. Biotech will be a great financeable market once again, and we will thrive when that happens within that part of the market. But what's important about Definitive Healthcare is we sell to all parts of the healthcare universe. So over time, yes, we're going to feel the pressure of an overall macro headwind, but the ups and downs of biotech, for example, doesn't impact us so significantly. It impacts us on the margin, not head on.

So you talked about the funnel being as good as it's ever been, right? I mean that's I think what you just said. I think you said that in your call. So one of the things, I think, that you have changed is your go-to-market and your sales motion, right? And so do you want to walk us through like what you think are the most meaningful changes and how that's impacting your ability to get deals closed? Because it sounds like the funnel is doing just fine. And how are those changes in the go-to-market motion going to help you maybe close things more effectively in the down tape that we're in right now?

The changes we've been making, it's really been an evolution over about 3 years, which is to verticalize our go-to-market team. So we now are organized in 3 subsegments: life sciences, healthcare providers, and diversified, which includes healthcare IT and diversified companies like banks and financial institutions. So why we believe that, that is important is it allows each individual within those specific groups to really understand what are the use cases of our end clients, how do we make them tick, how do we get them excited about the product, everywhere from bringing them on as new clients to renewing them and upselling them. So we expect that the benefit of that verticalization we've seen that over the last couple of years. We made the final move this year, where we pushed all new business and upsells under a single leader. So what's great about this is now that leader is only thinking about their end market. They're thinking about what can we do on product, why are we not closing deals? How can I make runups better in order to be successful within that market? So we have no doubt the benefits will pay off there. At the same time, within some of our more complex markets like life sciences and to some extent, providers, we're starting to supplement these teams with more subject matter experts, which is some of the sales and marketing investment that you're seeing, we believe that this will pay off over time as we are able to have those people go in and really understand, dive deep into the problems that the clients are having and design solutions using our data that meet their needs.

And if you think about your customers are facing some level of duress, right? And so you talked about modifying the sales motion, adding vertical specialists, subject matter experts. What about modifying the product in some way to make it more -- I don't know about usable, but maybe more accessible and affordable in a period of duress. Is there any modifications you can make to the actual -- the product itself?

Yes. I think what we're focused on is how do we drive more value for our clients rather than how do we change pricing or anything like that. And this comes in 2 ways. We talked a little bit about what we're trying to do with large language models of making our client service team and our client success team better at answering questions and more suggestive of ways that our clients can benefit from the product. So that's very important. And then just constant innovation in our products. So the things we're doing with data science and with new data that we're putting in, these are things that can drive value for our clients over time. And what's great about Definitive is we've always been focused on how do we balance growth and profitability. We're a very profitable company. We generate a lot of cash flow. That has given us a significant advantage during a market like this where we can continue to invest in the new product features and the new data that's going to drive continued growth over time. So for us, it's all about how do we deliver more value, not how do we change the price in any meaningful way.

And if you had to think back, like what do you think are the most compelling value drivers that have changed over the last 12 months?

Over the last 12 months, we are constantly innovating. So we've done a lot of good stuff. We have started bringing data science with Atlas AI, much more front and center for our clients. So we've always sort of developed different insights and module -- or different insights that we bring into the modules that we have. We've made these far more accessible. So we've -- where we -- maybe we try to charge for some of this stuff before, and now we're including it in the base subscription. So that's a really important way for us to drive more value for the clients. And then the second is more integrations. We've invested heavily on how do we get our data into the workflow of our clients. So we rolled out earlier this year a new Salesforce app, which gets our data right into the Salesforce workflow, which has been fantastic, very well received. We rolled out a Snowflake app. So we want to make sure that our clients don't have to go through extra work in order to use our data the way that they want.

Got it. And if you look back, since you're the founder of the business you've been around since the inception and I don't know if you've ever seen a downturn that look like this, right? But when -- if you think of the cycles, the business cycles that you've seen in the past how have they played out when you look back over time, like how have those cycles played out for Definitive?

Yes. So what we've seen is as we go into a bad cycle, we see new logos get hit first. And that what we see is over time, upsells and churn take some impact, but not as much as new logos. And that's what we've seen during this downturn as well. So you've got that sort of delayed impact on upselling churn and a little bit more muted effect. What we see on the upswing out is new logos pick up first. So we start to see more excitement where people are -- have been looking at our product. They're interested. They just don't have the budget. But when those budgets start to free up, they go and add Definitive Healthcare because it will help them drive growth as they start thinking about growth. I would expect that's what we're going to see again at this point is at some point, we're going to see new logos start to materially tick up, and that will be a good sign for us that we want to start pushing hard on sales and marketing once again.

And how do you think about the cadence of the year? Do you think it's even possible that new logos could accelerate before we get to the end of the year? Or is this -- do you think we're looking pretty much into '24 based on...

I think it's hard to say, honestly. I think we have a good enough ROI that perhaps can pick up before. Also, the other hypothesis would be, you've got to wait until the next budgeting season at the end of the calendar year to really sort of free up that money. I think it's very hard to say.

And what happens to pricing in these cycles? Is it materially impact pricing? Or do you generally be able to hold price?

We think we have a significant pricing opportunity. So for 2 reasons. The studies that we've done show that we are actually under market on pricing. And then the second piece is because we've grown so fast as an organization, a lot of our existing clients have come in at much lower prices. And it's very hard to get people back to the price -- the market price as we're raising prices quickly over time. So I think what we're seeing in an environment like this is we don't have a downtick in price, we're just not able to get quite as much price increase as we would like to get.

Got it. So one thing that comes to mind with you and you mentioned is you're highly profitable, highly, highly profitable business. And I think shortly after coming public, you were operating prepublic in the low 40s. You brought them down after being public, you had incremental expenses and you made it known you were going to make certain investments in data, which you've done, right? But we're actually operating below that now, right? So obviously, business volumes impact the margins. So how are you thinking about where we go from here? Like what's your thought process? We're in the down cycle, our margins are lower than what we thought they were going to be. How are you thinking about how the progression is coming out of this?

Yes. I guess, first of all, the way that we think about our financial model is a long-term game. We have a fantastic -- a huge market. This is a compounder that we believe we can drive significant sustainable growth over time, and we can bring EBITDA margins over time back to the levels that we saw just after IPO. So we think that, that is all possible. Now the speed at which we do that depends upon the investments that we have in front of us. So we do want to balance growth and profitability, but we want to continue to invest to drive that long-term growth. So I think what you should expect to see is that margins will go up over time, but they will kind of steadily go up. We're not going to see sort of a quick return to 40% because we've got good investments to make, and there's other data sets that we want, there's new technologies that we want to invest in. But we'll continue to make those trade-offs very focused on the fact that we know shareholders want more EBITDA margin right now, and we want to make sure that we continue to make that balance that we got to drive growth, but we've got to continue to invest at the same time.

And one thing that I forgot to ask earlier, and I'm glad you mind just here is about the source of your data sets. Do you mind just taking a minute sort of explaining where those -- because you source data from a lot of different places and then curate it. So do you want to just take a minute just to level set everybody on that?

So we get data from 4 key sources. The first is we've built technology to automatically pull information in from about 200,000 or so places in the public domain. This is research journals, industry publications, websites so on and so forth. So we've built that up over the last 12 years. Second is primary research. We make about 700,000 phone calls every single year and gather data from about 3 million e-mail surveys as well. So we get a tremendous amount of proprietary data flowing in through that. The third is government and regulatory agencies. So it's a highly regulated industry. There's a lot of information collected by talent, municipalities, states, Center for Medicare and Medicaid. We pull that data in, it's about 25,000 different sources that we're regularly getting data from. And then starting in 2019, we started buying third-party data as well. We never bought any data before them. Third-party data is primarily medical and prescription drug claims. We launched a medical claims product in 2019. A few years after that, we followed up with a prescription drug claims product. So those are the 4 key areas we get data from. But the power is not only the breadth of different sources, but the data science that we utilize to pull this all together. Data is highly imperfect, especially when you get it from a lot of places. So being able to standardize it and fill in blanks with proxies when data is missing, building all of that data science out over time is really what makes us tick and what brings it all together to make it a useful data set versus just a lot of data.

And just on -- going back to the claims data, and we talked about this a minute ago. But just again, to differentiate what you're going to do with the claims data versus some of your competitors, just what the key difference is because you can leverage that on top of what you already have right on your proprietary data.

That's right. So if you think about our proprietary data is the hub of data, and then you start to add these spokes on, which could be like intent data or claims data or prescription drug claims data. When you link that to our data set where we have every facility, every physician and a map of all the affiliations between them, now you can start to see market share at the facility level or at the IDN level or our provider clients, for example, want to look at leakage. So they want to follow these patients, which are tokenized. So we don't know who they are, but we can follow that patient throughout their entire journey. We can see when they leak outside of the network and our provider -- clients can figure out who's leading, why are they leaking? Is it a certain service line? Is it a certain geography and they can develop strategies and tactics to stop that leakage over time. So really powerful analytics that you can do only when you link that to the proprietary ecosystem data that we have.

And is there any advantage in terms of populating trials with what you're doing versus what some of the others since that seems to be a big driver in the marketplace right now is populating and managing trials and the prescribers?

Yes. So we don't populate trials with patient patients, but we help them identify who are the best and most important investigators and sites to populate trials. Where clients are using our data that is differentiated is life sciences becoming more corporatized. And what I mean by that is it's less of going to each individual physician to sell or to find clinical trial opportunities. And it's more about how do you go to the IDN or a physician group and get many physicians on board at the same time, our data and our affiliation data allows them to do that. And that's the only place they can take that approach.

Got it. Before I go on, any questions in the audience?

[indiscernible] investments that you're [indiscernible] that are bringing that EBITDA margin down. What does that look like? Is that a discount on getting [indiscernible]

Yes. So if you look at our financial data over the last couple of years, you've got really 3 major things. The first is public company expenses. So that's just -- it is what it is, D&O insurance, more finance, that type of thing. The second is sales and marketing. So we have invested relatively heavily in sales and marketing over the last couple of years. And that is really -- it's more quota carriers, it's more subject matter expertise, particularly within life sciences and some of those supporting actors as well. And then the third is we made a significant data investment last year that's starting to hit the financials of this year. So what you typically see with that is these data investments tend to be a short-term impact on EBITDA and gross margin. And this one was to the tune of about 3-ish percent, give or take. And then what we do is we have to grow those over time. So you've seen this if you sort of look back over our history, short term, the data investment hits, generally, they're fixed cost, we outgrow them over time. You got it.

Anyone else? Any other questions? All right. So why don't we go on to M&A since that seems to be your charter in your new role here. Two of the more important acquisitions were Monocl, right, and the [ KOL ] space and most recently, Analytical Wizards, which is really an algorithm that helps optimize marketing spend, right? What do you think you've learned from buying these companies in terms of how it impacts the business? And how does it really inform your strategy going forward?

Yes. Good question. We like both those acquisitions. So we've we know we want to do more of that type of thing. In both cases, our acquisition thesis is about finding tuck-in acquisitions that can add a new capability to our data set. So how do we bring in new data that we don't have linked that to the data we already have to make it stronger or a capability where we can use data that we already have in new ways for our clients. So Monocl brought in data, enhanced our overall data ecosystem. Analytical Wizards is more of a capability where we're able to push our data into the tools that they have, combine that with third-party data and solve new use cases for clients. I think in general, we want to continue with that model. So high gross margin, okay, if they're unprofitable at first. I think the thing that we've learned overall is we'd like to integrate fast. So we've always tended to integrate fast. I think we need to -- as we get bigger, when we bring a new company and we need to figure out very quickly, how does this link in with the rest of our company? What are the changes that we want to make? And how do we integrate it even faster than we did? We're pretty fast in both cases, but I think we could have even been faster than we were. And I think the benefits of that for clarity of what is our offering? And how do we sort of manage this efficiently are very high if we can integrate quickly.

So without playing your hand, like what are the areas of interest you most? Like what are the areas that look could be most value to you at Definitive?

Yes. So we love unique data sets. So if we can buy a unique data set like Monacl, that is sort of, first and foremost, what we're looking for. We are -- we think life science is a great market for us long term. So any tools that bring in data or can use our data to drive life sciences, sales and marketing effectiveness, we're interested in that as well. So those are the 2 areas where we spend most of our time.

And what is the primary value proposition to pharma for using Monocl? How is that?

Yes. So Monocl is a database of about 12 million experts across the entire globe. So details on the clinical trials they run, research publications, which pharma companies are currently collaborating with. The value proposition is really twofold. It's in the R&D stage figuring out who are the investigators to run your clinical trials as well as who are the people to help you design protocols and drive your research and development effort. And then once you start to commercialize who are the important influencers across the globe that you want to make sure understand your product and could be evangelists as to why it's a better standard of care than what's currently out there.

Got it. So we just have time for one more. So since you're the founder, this is a fair question. So if you could do one thing differently, make one decision differently you could have that back. What would that have been?

That's a tough one. I think throughout time, it's funny. I always run a very profitable business. And sometimes some of my institutional investors would say, hey, we should probably hire this person, I was always a little slow, kind of those people. So oftentimes, I find myself 8 months later saying, boy, I should have hired that person 8 months ago when I thought it. So I would have probably done that a little bit more throughout the career just thinking about what do we want to look like when we're a scaled organization and how can we hire slightly in advance of that was something that I always thought back, I should have done better off.

All right. Well, I think we're just out of time. Is there any -- we have time for one more if anybody has anything? All right.

So probably question has a long answer, but can you describe the data that we [indiscernible] the stack data translate into what I thought I heard to say, which was evaluating healthcare facilities and [Technical Difficulty]

Yes. So we -- it's a good question. So we deliver our data through 2 different ways. So we have a SaaS platform where somebody can go in and view a provider and look at all the details about clinical trials they run or the procedures or diagnosis of the patients that they're carrying for, technology, infrastructure, you, David. So we've got all of this data. So that's kind of how we deliver our overall intelligence. When we had the claims data on top of it, now we can look at things like you start talking about like standard of care, we can look at for a certain type of cancer patients, what drug are they receiving and in which order and what's happening for that. So you can start to look at this patient journey, figure out what sites they're going to, what types of drugs they're taking, what are comorbidities from a diagnosis standpoint, so you can start to figure out who are underdiagnosed patients for that disease. So there's a lot of different things that you can do with that. We try to do this as much as possible through a self-service delivery model with great tools.

[Technical Difficulty] this care. So a bit of understanding what are the costs [Technical Difficulty]

Yes. For somebody to take that like that's a bigger project, frankly, than I think you would use some of our data, but you probably would need serious consultants and stuff to really help them evaluate that type of thing. But we do have like we have cost structure, for example, of the entire episode of care, which is important for that analysis. So as an example, when you get a knee replacement, we don't just say Mass General charge you $20,000 for the knee replacement. We say we look at the overall episode. We say, you had rehab afterwards, that cost another $4,000. 1/4 of the people got readmitted back in to have it fixed. That's another $70,000 for a quarter of people and the overall cost of the knee replacement is now $80,000, not $20,000. So that type of analytics we have, and that is part of what would go into that type of decision. But just say you could do that all just from our product, it would probably be a little bit of a stretch.

All right. I think we're going to after we're over. I think we're going to wrap up. Thanks very much. Great to see you.

All right.