What Happens When Patients Finally See the Price First

I used to think I understood healthcare.

Then my patients asked me one simple question…

“How much will this cost me?”

And I didn’t have an answer.

Not because I didn’t care…

But because the system wasn’t built to give one.

In this episode, I walk through a truth most people never hear—why prices feel random, why even doctors don’t know them, and what happens when we finally flip that model upside down.

There’s a belief that patients can’t shop for healthcare.

I don’t buy that.

And once you see what’s really going on behind the scenes… you won’t either.

This isn’t just about cost.

It’s about control.

In this episode, you’ll learn:

• Why healthcare pricing feels unpredictable—and why that might not be an accident
• The hidden system that makes simple care feel complicated and out of reach
• What changes when patients can finally see the price before they say yes

Connect with Rahul Naidoo, the CEO and Co-founder of Superscript.

• https://www.superscript.nyc/

Chapters:

00:00 – Opening Thoughts on Complexity in Medicine

00:23 – The Question Doctors Can’t Answer

01:46 – Why This Conversation Matters

02:01 – From Computer Science to Healthcare

04:47 – Quitting Jobs With No Plan

06:25 – Why Healthcare Was the Target

07:40 – Are Prices Really Random?

08:28 – The Idea That Changes Everything

11:25 – What Problem Is Actually Being Solved

13:38 – Why Pricing Is So Complicated

17:33 – The Hidden Insurance Puzzle

20:02 – Can This System Actually Work?

24:17 – Where This Gets Really Interesting

27:37 – The Bigger Vision for Healthcare

31:48 – Building the System From Scratch

35:04 – Growth, Scale, and What’s Next

37:28 – The Real Role of AI in Healthcare

42:40 – What Doctors Think About This

45:02 – Why Practices Are Losing Money

47:31 – A Different Way to Think About Insurance

51:53 – The Philosophy Behind It All

54:50 – Final Thoughts and What Comes Next

Transcript

SPEAKER_00 0:00

In medicine, the complex is often simple, and the simple is often complex. Taking care of patients is complex, but over the years, oftentimes it gets simple. But then patients would ask me a simple question. I'd recommend a surgery to them. They'd say, How much is it going to cost me? And I look at them, I go, that's a great question. I have no idea. So then I would send them to my billing people. I had great billing people. And I thought that was the end of it. And then I found out later they gave them just an estimate. They didn't really know either. And what I found out later in my practice, we started signing the checks, the partner students signing checks. And I was always signing these checks for $50,000, $7,500. And you're going to patients. There'd be a stack of them this thick each month. And I asked my manager, I was like, what is with these checks? And she said, Well, those are the people we collected copies for that didn't have a copay. And I said, Well, why do we collect them? She's like, Well, we don't really know. So it's better to collect it at time of service because you can't get it later. And I was like, Well, that doesn't seem to make any sense. But today we are lucky we have the man who's making sense of it. He's going to make that simple again. So we have, and I'm going to mispronounce your name. I apologize, Rahul. Nadu, is that the correct pronunciation? But that's that's so we have it there.

SPEAKER_01 1:15

You was so close.

SPEAKER_00 1:16

Okay, thank you so much. And I will just add that this podcast is educational purposes only, not medical advice, and always take the advice of a physician. Rahul, thank you so much for being here. I'm very excited.

unknown 1:27

Dr.

SPEAKER_01 1:27

O'Leary, thank you for having me. This was an exciting conversation.

SPEAKER_00 1:30

Absolutely. And one thing I noticed about you're an interesting fellow, and I was reading about things today, and I thought you are actually the Hercules of healthcare because you're taking on these huge tasks, and Hercules had 12 labors they had to complete. So I think I'm gonna put this down as your first labor, and over the years I'm gonna give you 11 more labors. But anyway, but that I think that's very exciting. But your background is fascinating because you studied in Cape Town, you're from Cape Town, South Africa. And your school, you excelled at English and debate. And looking through that record, I would have thought you would have been an attorney or a diplomat. How did you get into computer science from there?

SPEAKER_01 2:10

You're exactly right. Growing up, I always thought I was going to end up in a humanities-leaning profession because I was my spike was really in this make making my mouth run, you know, making the words come out and uh sometimes sounding articulate when I did that. When I arrived on campus undergrad 2015, um, I knew one thing in my heart, which is that I wanted to build. And I didn't know what it meant to build yet. I didn't know where I wanted to build, I didn't know how, but I knew that I wanted to do something that mapped closer to the weird and wonderful and less to the conventional. And the only pathway I could see to doing the weird and wonderful uh for the rest of your life um was to build. And so that's what I did. Um I jumped into computer science, and computer science was for me just the pathway. Um it was an awesome degree. I learned more about discrete mathematics than programming, um, but it was really interesting to kind of sink my teeth into what is and isn't computable, um, the theoretical foundations that computers are built upon. Um today, when people ask me about you know what they should study, um, my advice tends to be that it's less want to learn technical knowledge that you're going to directly apply, and more you want to learn new ways to think. Um, because the new ways to think are really what opens up the world when you eventually leave school. So, computer science did that for me. And so when I left college, I knew that I was going to build something. I had no career momentum in any particular direction because candidly I didn't want a career. I wanted to do something. Um, and I didn't know what that meant at first. So, out of college, I spent two years at a company called Royband Sciences. Um, in particular, I joined a small team within Royband that was supposed to make digital health startups. You could think of it as like a startup incubator team, and that was perfect for me because um it gave me a crash course and what it actually meant to build a company. Uh, I was there from the initial seven people all the way through to 32 people, and um I saw a company and a product go from idea to being in the hands of over 2,000 researchers in the clinical trial space. At about the two-year mark, uh, my best friend from college, his name is Ritt, uh, he came to me and said, Hey, Ruble, this is the year that we quit our jobs and just built something. Um, we always knew that we were gonna do that. We were attached to the hip from the end of sophomore year onwards. And you know, we like to say sometimes that in our journey, what we ended up doing is almost more random than that we ended up doing it because we knew we had to do something. And so when he said, Yep, it's time, I said, You're right, it is time. And so we quit our jobs and we quit into, well, nothing. Uh we had no plan. Um, so the very first thing that happened was we changed our LinkedIn bios to say we're building in stealth. And some investors reached out to us to say, Oh, what are these two Harvard undergrads building in stealth? Uh, we want to know. And of course, we weren't building anything in stealth yet. We had no idea, and so we told them, you know what, we we're ready for conversations in three weeks. Um, right now we're very heads down, but in three weeks' time we're gonna take external calls. And then it was me and Rick going, Oh man, okay, we have three weeks to put something on a pitch deck, so that we have something to tell them. Um, now what we did know leading up to that phase was that we wanted to do something in healthcare. We had spent about two or three years meandering through lots of different idea spaces from restaurants to e-commerce to eventually healthcare. And we picked healthcare for two big reasons. One, we wanted a space where we could build the biggest company in the world. We're not in it to do anything small. Um, there will be no second prizes. And second of all, we wanted a space where, well, if you're gonna wake up in the morning and do something really hard, and we knew it was gonna be hard, it had to be worth doing. And so we wanted a space where there was a clear path to building something positive, some. And in healthcare, we saw that intersection. We we we we believed at the start, and we still believe that today, uh, that if you build the right thing in this space, you can do a world of good, and also build the biggest company in the world in the process. And that really excited us. And so we we said healthcare is the space we're gonna look at really, really carefully. And when we quit our jobs to get going, uh, something had just happened in healthcare that was really interesting. The price transparency legislation had passed in 2021. And the price transparency legislation basically said that for the first time ever, hospitals had to publish their internal rate sheets known as charge masters. And this was very interesting to us because, you know, two years out of college, we're not much of anything yet. But if you'd asked us to self-describe, we would have said we're consumer technologists, first and foremost. Um, we like technology, we've always liked technology, we like we liked our apps, we liked being able to see a price and click a buy button. And when we looked at healthcare, we saw distinctly not that. Um, exactly as you opened our conversation, Dr. O'Leary. Um, in every other category, you see a price, you click a buy button. That's how you move through the world. You have that absolute certainty, you have that absolute convenience. And then in healthcare, from the patient perspective, it feels like you're charged random numbers. Um, frankly, uh, you have no idea why they the numbers are what they are.

SPEAKER_00 7:04

They are random. To be fair. The charge masters are random. I mean, there is no sense from one hospital to another. You know, why why is one hospital in New York City gonna charge you four times as much as another one? Why does it vary by tenfold across the zip code? It is random. You're a hundred percent right.

SPEAKER_01 7:25

Yeah, it's true, because the the reason that one is four times more expensive than the other is so far abstracted from the buyer that yes, it may as well be random. It is. Um, and so we saw that experience in healthcare, and we said that has to change. There is no way that in two decades we're still going to have that experience. Um, and so our minds were spinning on okay, well, how do we create this consumer shopping experience in healthcare? How do we drag healthcare to the state where people see price and click buy button? And as we were having all of these very um uh dangerous thoughts, um, the idea that we kind of became transfixed by was that this was actually the solution to fix the category. Because when we looked at healthcare across the board, um, we could map almost every breakage that we saw to a fundamental misalignment of incentives that sits at the bottom of the market, which is to say that there will be no fair and true consumer market with real pricing and service delivery that is obligated to serve the consumer's best interests if the consumer is disempowered in the buying interaction in that consumer market. And so what we came to believe was if you empower consumers and you allow them to shop in healthcare the way they do everywhere else, you give us a decade of that behavior, and we will give you a completely transformed category. And so um that was it. The game was set. Uh, we looked at the price transparency data, we said this isn't enough, this is 2% of the way there, there's so much to build, and we got started.

SPEAKER_00 8:51

I love that. No, that that's a great explanation. And the other thing, there's a study from 1998 to 2021, the American Enterprise Institute, and they looked at cosmetic surgery prices, which is fascinating because that's cash pay. So that is exactly what you're talking about. You go in, and patients are very price sensitive, you know, for and they looked at the 16 most common procedures that were done, you know, tummy tucks, breast lifts, uh facelifts, all these things. And over that time period, it went up 38%. So it trailed the CPI at that time was 66%. So it's half of CPI. And medical services went up 138%, and hospital services went up 240%. So you're 100% right. If you can get patients, and this is one of these things that drives me crazy. People go, patients can't shop for health care. Well, they just did, and they're really savvy shoppers when they're given the price signals. And that's so I love what you're doing because we can't afford not to have people shopping for price care, and we've got to get rid of the rent seekers in there. So, no, this is phenomenal. And so, oh, go ahead. I'm sorry.

unknown 10:00

Dr.

SPEAKER_01 10:00

Larry, I'd love to add, um you made that comment, you know, people like to say patients can't shop for healthcare. Um, this hits at one of the cores of our belief, which is we believe that people are really smart and really sophisticated if you give them the opportunity to be so. Um, uh a trend that we've observed over kind of the rise of the information age is category that consumers do not understand, category suddenly becomes digitally accessible with price and information available at the click of the button. Over just three to five years, consumers become rapidly sophisticated. And then next thing you know, seven years from now, they're driving. Um, and we don't believe healthcare is going to be different. It might take a little bit longer because healthcare is a lot more complicated, but we believe that if you give people the access to the information, you will see the same trend play out. This one is not different. So that's it.

SPEAKER_00 10:50

And what's the specific problem, if you could just break it down for us, that you addressed in the whole healthcare unit? Because it's a huge universe, as you said. So, which specific part does Superscript address? For both patients and practices?

SPEAKER_01 11:06

Yes. So the end result of all of the technology we pulled and all of the talking that we do is a patient and in our vocabulary a consumer being able to see a price for a service, whether that service is a consult, a surgery, or an X-ray, before the service takes place and actually transact on that price. Click the buy button like you would for any other good in the market. And when you click buy on that price, you're done. There's no surprise bill after the fact, there's no change, there's no like, whoops, we thought it was different. Um, it is a price as true as if you clicked a buy button on Amazon. That's the experience we create. Um, to take a peek behind the veil, you know, what are we actually doing to the system to make that so? Well, we're solving a couple of problems. Um, the first one is that there's a big data problem inherent in making healthcare prices. And that didn't need to be true. Uh, if all the prices were cash rates, the doctor would be very, very capable of just maintaining their own rate sheet and saying, you know, this is how much it costs. The challenge is that two things have happened to healthcare pricing over the past many decades. One, there are many, many providers off price. And when I say provider off price, I'm referring to the insurance company. Uh, they set the rates as we know. In theory, they set the rates in negotiation with providers. In practice, that's not exactly how it happens. Insurance companies set the rates, right? And then the second is that over time, the healthcare goods and services have become increasingly atomized. Within industry, we talk about CPT codes. Um, and I know that you're familiar with CPT codes. Common procedural terminology, a single CPT code represents a piece of a treatment. And so now the challenge is that if you want to understand the price for a given service happening in a medical practice, and let's say that service entails eight of these pieces, eight of these CPT codes, well, in order to understand the price dynamics for that service, you need to be able to ingest the data of those eight CPTs across every single payer that that practice interacts with. Um you imagine every single CPT code and there are hundreds uh across every single insurance company that the practice engages with, and all of a sudden you have too many moving parts for anyone to make sense of that. So this is the first issue. The second issue is a data ownership problem, which is if the doctors, if the practices had their own rate sheet, you know, Aetna is going to reimburse XYZ for CBT1, W for CPT2, BCBS is gonna reimburse XYZ for CBT3. Um, then also the practice owners could probably do the job. They might have to hire someone to do this directly. The challenge is that practice owners don't even have their own rate sheets today. Um, insurance companies have their rate sheets. Most physicians are in a place where they signed a contract with an insurance company 15 years ago. They maybe saw the rate sheet 15 years ago when they initially signed on with an ed or a BCBS. They have not seen the rate since there's complete rate drift. They discover how much they're going to be reimbursed when the claim is returned, not beforehand. So that's the second piece. And then the third piece is what I would call a data coordination problem, which is to say that assuming you can identify all of the different pieces and you can get the different pieces necessary from the parties that have them, you then have to be able to put the pieces together into in a very precise configuration. It's like you're building a jigsaw puzzle in order to make a price for the patient before the service takes place. And to be a little bit more precise there, there are kind of two big buckets in this jigsaw puzzle. The first bucket is understanding which of those CPT codes are going to occur when the patient has the treatment. In some cases, it's very straightforward. If the patient is just going for a consult, that's probably going to be one CPT code. It could be the normal consult, it could be the severe consult, but it's going to be one CPT code. Um if it's a consult plus a blood draw, that's probably gonna be two. One for the consult, one for the blood draw. The second it becomes at all um I won't say complex, um, I think the word I was thinking about is the tip of my tongue, it's not there, but imagine a surgery, shoulder surgery, etc. That's gonna be 22, 24 CPT codes, depending on how it's performed, where it's performed the patient's shoulder. Now all of a sudden you need to have some sense for exactly what CPT codes will occur in that surgery. And given the fact that that surgery might be performed differently at different locations for different patients, you need to be able to account for the fact that that set of CPT codes is going to differ. Could be 20 this time, could be 22 next time. So now you have a pretty complicated data problem where you have to identify the right CPT codes and even make a prediction about which CPT codes are going to show up on this patient's claim. So that's the first part. The second bucket in this coordination problem is applying the patient's insurance plan to this whole mass of codes. Now, I think the right way to think about an insurance plan is to use slightly different language because insurance plans don't really function like insurance in the healthcare market today. As we know, um in other categories, insurance saves you when catastrophe occurs, and then you know insurance pays out an amount. Correct. Um, that's not what happens here. Insurance plans in healthcare are just cost-sharing structures. And so an insurance plan is best thought of as a number of cost sharing rules that range from very simple rules, we're going to make it a $30 co-bay, throw away the rate, it's now $30 for you, down to very complicated. You have to pay this amount before you hit your deductible, then you're gonna hit your deductible, and then a 10% coinsurance is gonna apply. And so you're gonna pay 10% of the remaining amount after the deductible. And if that sounds like Greek, that's because it's probably supposed to sound that way. Um it sounds that way to most people. So, what we have to do in that bucket is get that real-time insurance information on behalf of the patient, understand exactly what their plan covers and how it covers it, and then apply it to all the work we're doing on the rates side. This is a big, big, big puzzle. Um, the back end of our company, Superscript, is very complicated. We have a million moving parts. Um, but we have to hold a lot of technology to make a very, very complicated system simple. I like to say that there's almost a universal law of preservation of complexity. You can't delete complexity, you can just move it from one place to another. And so what we do really is we move the complexity to the front end where physicians and patients experience it to our back end so that the front can be simple.

SPEAKER_00 17:39

Now, the other question I I was thinking about this this morning, and it probably differs as well if you have a fully insured versus a self-funded plan. Because if you have a, do you work with, you know, if there's a self-funded plan that has an independent third-party administrator, now it even gets more complicated because all of a sudden there's a bunch of vendors in the middle saying, you know, is this person eligible for this insurance on this date? Is the does that benefit cover this surgery? Is this surgeon, hospital, in-network, out of network, repricers? And that could be five different vendors that you have to go to. So do you how do you work with that? Is that more complicated for you than if it's you know a fully insured plan and you just have to deal with hopefully say United Health and maybe three other vendors that they have?

SPEAKER_01 18:33

Yeah, so it's definitely more complicated. The short answer there is for us, it entirely depends on whether or not the provider of the plan cost sharing rules can give us access to those cost sharing rules electronically.

SPEAKER_00 18:45

Okay.

SPEAKER_01 18:45

If they can, we can read it and then we can generate the price on behalf of the patient.

SPEAKER_00 18:49

Okay.

SPEAKER_01 18:49

The challenge, and the reason I'll say that we can make 92% of the prices, not 100%, is there's that long tail that exists. Yeah. It still hasn't digitized their planned cost sharing rules.

SPEAKER_00 18:58

Okay.

SPEAKER_01 18:59

And uh, while we are absolutely going to use modern AI technology to resolve that too these days, you can make a robot phone call, um, you can read all sorts of non-digital data, even a PDF, as you know, very, very straightforward. Um, today it's digital providers of cost-sharing rules directly.

SPEAKER_00 19:16

Okay.

SPEAKER_01 19:17

So some of these third-party administrators, easy to integrate with, they're they're very next generation. Others challenging, depends on which one we're talking about.

SPEAKER_00 19:26

Okay, that's what I was wondering about because I mean, some of them are obviously the TPAs are part of the large insurance companies, like UMR is part of United Health, and then other ones, you know, I could set up Dr. Jim O'Leary TPA tomorrow, and you'd be like, oh my God, this guy's a nightmare, you know? He's using 15 vendors. And so I was just thinking about all the complexities there. And no, I think it's phenomenal that you're taking that on. And could you tell us what is your success rate if you have, say, a thousand cost shares that you say we made a determination on what would be, you know, I'm you I'm sure you have a lot of statistical analysis of this. What would be your error rate? And I'm sure you've done a lot of modeling of this.

SPEAKER_01 20:13

Mm-hmm. So first of all, let me define what a success means. Because what we do with healthcare is we split between deterministic versus non-deterministic.

SPEAKER_00 20:23

Okay.

SPEAKER_01 20:23

In the deterministic case, the service takes place exactly the way we expect it to, and it will be the same every single time. It's the patient going for the consult and the blood draw. And so if we're pricing a deterministic service, what we look for is what we call an exact match in terms of the price that we make is exactly the price they chose up after the fact. Now, if the service is non-deterministic, it can vary. It gets quite complicated because what you wander into there is a question that we didn't realize we would have to answer when we started building. And it very quickly became one of the biggest questions at the core of the company, which is what exactly is a healthcare good? How do you draw the line between versions of one good versus entirely separate goods? Because this has not been done until we stepped in and said, hey, we want to make these goods understandable and viable for the front end, because the back end has always just been these atomic CPT codes. It didn't matter how you treat the lines. So I'll give you a simple example. Um, let's say the patient goes in for a consult and then an x ray is ordered. That is a clear case of good one, and then an add-on is added. It is a separate good that encompasses an entirely separate charge. And so that's a case where we talk about an additional test and service. And so in that case, um we would aim for an exact match on the consult and then an exact match on the x ray after the fact. Let's say, however, the patient goes in for a behavioral health therapy visit. The behavioral health therapy visit before the patient has the visit might be a normal ENM code or a severe ENM code. Now, it is actually between us and the practice to determine how do we want to handle that situation. Do we want to assume normal and then charge the additional if it's severe after the fact? But that's already a bit complicated because is the patient really going to understand that it was a more severe visit and now they have to pay normal? Do you want to balance between the two, which is what we're designed to do? Make a price that fairly represents both outcomes? Do you want to charge the more expensive one and then refund the difference after the fact? Now, the way SuperScript handles that is we balance. And then after the patient has had a single visit and we know which type of code is going to be billed when this patient goes for a behavioral health visit, we can then look at their historicals and get the exact right price every time. So it goes from a balanced case into an exact match case. The reason I call out all of this complexity is to say that in the exact match case, well, we look for exact matches of north of 90%. And we tend to get there. At launch, in our exact match cases, we start at north of 90% and we quickly get up to 95% plus. Because exact matches, we can just know. We can just be right. If we're making mistakes, it's typically because we had a wrong pay negotiated rate in our back end. We were mapping the wrong benefit, you know, a mistake like that. Whenever that happens, we go on risk, we pay the practice the full amount, we pay the patient the refund, etc. We make sure that all of the parties we engage with the whole, and then we just update our system internally so we don't make that mistake a second time. When it is that non-deterministic case, um, and an exact match is impossible because we're balancing between two rates, we look for a variance band, and the variance band is actually going to depend on the distribution, like how complex um uh that that that treatment or service is. And cool. At a high level, we look for within one standard deviation on either side of the mean.

SPEAKER_00 23:42

Okay.

SPEAKER_01 23:42

Yeah.

SPEAKER_00 23:43

That makes sense. No, because I was wondering because I didn't know how you could factor all that in and be deterministic on that, because how would you know? I mean, if you worked with a surgical group over three, four years, and you know, there's the same 10 surgeons, after a while, you could pretty much model what they're gonna do in the surgery because we all do the same things over and over. And you could say, okay, this is how you know this doctor bills, and you know, I never put in modifiers for my surgeries that would say it's more complex, I'm gonna charge more because I felt, and I know you can do it, and I know a lot of surgeons are gonna be upset when I say this, but I kind of felt like you're gonna have some easy cases and you're gonna have some difficult cases in the end, it's all gonna average out. I didn't feel it was fair to the patients. I in plus I enjoyed the complicated patients, it made me a better surgeon. So in the end, I was like, I was getting something to benefit from that. And you know, that's why I'm a poor businessman, but a good doctor. It is, but yeah, but I so I was like, you know what, I I don't want to add complexity unnecessarily. So there's things you could do like that that I wouldn't do, but uh, you would pretty soon get an idea of physicians and know quickly, because either you do that or you don't. You don't just wake up tomorrow and start changing. And it uh but I love all these ideas. I mean and to take it further is I was thinking about this, and where this really gets exciting is if you go into so I'll say superscript 2.0 or 3.0 because I don't know where you're at in your journey. It's if you go to a self-funded employer and you say who's working with a transparent TPA and they want to lower their cost, but they want to get good care for their patients, and then you could say to them, okay, we're gonna help you price medical services accurately and fairly up front. So you don't pay $100,000 for a knee replacement when you could have paid $17,000 and gotten better quality. Because I then, you know, in those scenarios, and then you could say, and then if you if your patients, your employees use that $17,000 service, they don't have any cost sharing because it doesn't really make any sense to have them do cost sharing anymore. So then you get out of like, I don't want to say get out of the cost sharing business, but you kind of you know move forward to like that next level. So now the employers all of a sudden have tools. And a lot of the uh like crowd health is doing this, they have their net cost share navigators, which works great. Because if I'm with Crowd Health and I'm gonna have a knee search, I can call them up and say, hey, I you know, I live in Denver, you know, is there's some good people there, like, yeah, because they have their database now of people. So I I could see you guys kind of moving into that and adding tremendous value because I think the cell funded employers are the ones who can really cut the cost of healthcare because they're paying the bills directly. And you know, they can really push back on the hospitals and say, listen, here's a benchmark. You know, you're 400% of Medicare. That's insane. You accept Medicare every day. Why do you want us to pay you four times that? You know? So I do have any like what's your plans for the next stage? I I know you you guys are a fairly young company, but what'd you what's your like in your craziest moments in the middle of the night? What do you dream up next?

SPEAKER_01 27:01

Cool. So, first of all, I love the way of thinking about this, and you're exactly right. That is such a cool angle for us. You're also right to point out that that's Subscript 3.0. So, craziest moments, what am I thinking about? I'm thinking about Subscript as a network company because ultimately a network is what we're building. And I don't mean network in the insurance uh landscape way, I mean it more in the technology company way. Healthcare has needed an information network uh for a long time. In fact, a company that makes prices for all healthcare practices is just a special type of information network. It's a pricing information network. More broadly, healthcare needs an information network for um proliferation of information about doctors, which doctors they're in the area, what treatment pathways are available, how different treatment pathways connect to each other, perhaps details about doctors and practices that patients might care about and select on. I'll give you a great example. In some of our earliest user discovery days, uh, we found that it was fascinating how many patients wanted to pick their physician on the basis of cultural axes that we hadn't even considered. Um, for example, we had someone tell us, I want a physician that hikes a lot. Why? Because he hiked a lot and he would see eye to eye with that position. He felt like they would understand each other. There's another person who said that she wanted a doctor who had a different view on sexuality because she felt like she was sex-shanked by her previous OBGYN, and that was something she was looking for, but she had no idea how to look for it. There are other information networks that kind of serve as a substitute for the lack of information network in this category. You see a surprising amount of medical discourse on Reddit, Instagram, and TikTok. In fact, here's a fun fact for you there are over 72 million hits for the search silent reflux on TikTok. And that's because when people self-diagnose a silent reflux and they want to figure it out and they have no idea where to go, they go to TikTok. Um, Wilder Streams, Suberscript is a robust healthcare information network. Pricing is one of the many things that we do, but really the information is not free and accessible. So the entire industry is lubricated. Everyone is communicating with everyone else. Now, how do I connect that to um the comment you were just making about us going to self-funded employers? Well, we need to hit a network state in order for us to provide the value that you were articulating. So, exactly as you pointed out a couple of minutes ago, for us to do a fair job pricing any given surgeon or any given physician, we need to know how they make their claims historically. Because different surgeons are going to construct the claims certainly, some will have the modifiers, some won't. And so if we use some kind of aggregate data set to price a specific physician, we're not going to be accurate. And in fact, we might misrepresent in either direction, and that's not very good. So something we learned very, very early on was um, well, the healthcare industry and specifically the the uh healthcare politicians were saying, oh, you know, aggregate data that's the solution, price transparency legislation. We got it. And we realized no, no, no, no, no, no, you cannot use the aggregate data. You need practice level data to make real healthcare prices. And so what Subascript is doing right now behind the scenes as we provide our pricing technology to practices, is we're slowly but surely integrating with more and more practices so that we can provide accurate prices at the practice level. So you can think of it as we are bootstrapping the network. Once we have integrated with enough practices and we have been able to make enough practices transparent, that is when we can go to a self-funded employer and say, Well, here's real prices for all of the medical groups in your geography. And now you can know, don't go pay 100,000 there, pay 17,000 here. Um, but we have to solve that network retract problem there uh first. So that's it. That's that's the the big vision, healthcare network. And the second we hit healthcare network state, um, all the things we can build from that point onwards, super, super exciting.

SPEAKER_00 30:54

Oh, that's phenomenal. Where are you at in your now? I know you started in 21 and you were kind of building the program for three years. And then where are you at in the rollout in terms of how many practices, how many encounters, and what has that growth rate looked like over the last year?

SPEAKER_01 31:13

Yeah, awesome. So we spent the first, we're four and a half years in. We spent the first three years doing close to nothing commercial, um, just building that pricing technology uh heads down, uh, basically living under a rock. Then we surfaced from under our rock and said, all right, um, hello healthcare world. Do you want to use our technology, please? And uh that started what I would say this the second chapter of the business, which was us figuring out how to um build the adapter layer between our technology and the real healthcare operations. And in some ways, that was the harder part. It took about eight to 12 months of meandering in the desert to figure out the right product configuration to really make it seamless for practices and patients. Um quick like side note, piece of context you might enjoy. Our first swing at making a product for the practice was we wanted to make an open table iPad experience. You know, if you go to a restaurant today, the host host desk at the front has that little iPad where they drag your butt into a seat and they say, Oh, there's your table, go for it, right? So we thought we could give something like that to the front desk at a medical practice, um, except ours would make the price too. And so, you know, that's how we make the price transparent in every case. And so that's what we built. And for eight months, that was our thing. And a number of practices really liked it. There was just one big problem. That product form replaced the front desk admin's entire workflow. We were telling them, don't look at your EHR, look at this thing instead, which meant it took three, four weeks to train. After they onboarded, we were supporting it like crazy for weeks on end.

SPEAKER_00 32:44

Yeah.

SPEAKER_01 32:45

And we ended up drowning under our own service load because we were supporting intake, we were supporting scheduling, we were supporting the pricing, which is what we actually do here. We were supporting patient profiles, and we couldn't move forward because we were just doing so much. Recently, we re-scoped to something much lighter, lighter weight. Um, we serve it as a web browser. And when I say web browser, I mean literally like a Google Chrome. And so what we told Frontesk Admins is open your electronic health record, Athena, Experity, et cetera, in our browser and then keep using your EHR exactly the way you do. So there's no retraining. The only difference is the browser every now and then will just throw up the little pop-up that says this is the patient's price, click here to charge or click here to activate the POS device. And so that shift um made a night and day business, uh difference for the business, uh, for our ability to actually get the product out there and have people understand it and actually use it. So, with that context, last year um was really the time we spent with that open table iPad product. We went from doing tens to low hundreds of patient encounters to doing a couple of thousand as we deployed that open table iPad experience to on the order of 10 to 12 locations. We then made the swap to that browser experience, and about two months after that, we hit 50,000 patients uh served, uh, which is very cool. As I speak today, we work with 34 practice groups. We are live in, I want to say about 45 locations, and we are in a period of frenetic activity right now. Every single month we are launching more locations than the month before. So over the month of March, we're aiming to launch about 24 locations over the month of April, um, 35. And then May, June onwards. Let's see, let's see if we can break 50. Um, I will also say that we kind of already figured out sales at the end of last year, which was very exciting for us. We we didn't really know how to get in front of people, tell them about what we're doing, and for context, our contracted revenue at the start of Q4 last year and the end of Q4, 5x difference.

SPEAKER_00 35:07

Oh, that's great. That's great, Dan. And when would you see yourself being cash flow positive?

SPEAKER_01 35:13

Hmm. That's a great question. Are you an investor? Because our VCs will love to ask us that one.

SPEAKER_00 35:20

Really? No, that's funny.

SPEAKER_01 35:22

I think we're going to be cash flow positive probably on the order of two to three years from now.

SPEAKER_00 35:27

Okay.

SPEAKER_01 35:27

Let's say two. Let's say two if we're really disciplined.

SPEAKER_00 35:29

Yeah.

SPEAKER_01 35:30

Subscript is presently in very much a burnt-crow phase. Um, and part of the complexity there is we had to do so much RD setting up all of our infrastructure before we could get going. We have a pretty large system, um, and scaling requires us to build quite a lot of infrastructure. I mean, as we speak right now, um we're about to get into Q2, and the company is full-time focused on building what we call deployment automation and acceleration technology. Take, for example, the pricing on the claim side that I talked about a couple of minutes ago. For us to price value at any medical group, we have to ingest two to ten years of historical medical claims and then do all the analysis to understand which claims represent which treatments, what rates are associated, build the distribution, so on and so forth. That's days of work for a pricing and data engineering team. We have line of sight to the technology that makes that automatic, and in fact, we're building it right now. But you can see how you know, after three years spent setting up the infrastructure to even do the pricing, we're now setting up the infrastructure to do that quickly enough that it becomes a very, very efficient business. So I say two years because right now we're more focused on growth than that.

SPEAKER_00 36:39

Sure. No, that that makes sense. And how does AI factor into this? And what do you see as the best role for AI in healthcare?

SPEAKER_01 36:48

Cool. So AI is something we don't talk about nearly enough. Uh, we should talk about it more because it would be in our interest, but uh not there yet. We use AI deep in our technology stack. So our philosophy first is that we don't think that AI should be the end user experience in healthcare. And that's actually true in a lot of categories. I think people are getting very excited about AI right now, because it can do all of these cool things more than we had imagined. And so we're like, oh wow, everything is gonna be automated. You're gonna show up, you're gonna shake hands with a robot, and then the injection's gonna come out of a hole in the side of the wall and eject you, and then you'll even, you know, you don't even talk to a human. And the truth is that sounds like a dystopia. And no one actually wants that. Um, personally, and I don't think I'm the strange one here, I would like to go to a medical practice, shake hands with a real human who's smiling at me, and then go to the back and have a real connection with a physician who has spent years looking at patients and knows what they're doing and is going to talk to me and understand me like a person. Uh, and there is no amount of technology that will replace that. So, where does AI play a role in healthcare? AI should serve to eliminate the toil and increase accuracy, not replace the humans. Um, do I think that some folks are going to lose jobs in healthcare just as in every other category? Probably, because as we get more efficient, it's inevitable. But we're not going to replace the human in this category. Uh, not in the same way that I think that you're going to replace the human with driverless cars. So, with that being said, how do we use AI in our stack? From the very beginning, uh, we were quite lucky the rate with how the rate of AI kind of improvement uh shook hands with our roadmap. A simple example is when we started building our ability to read insurance plans. We started with just a giant if statement tree, um, which is to say basic code that says, if copay, then price equals amount in copay. If coinsurance, then price equals rate times. You get the idea. We very quickly discovered that that doesn't work well enough because uh insurance plans are so variable across the entire category, we need the ability to read plans we'd never even seen before. Thankfully, at that exact moment in time, OLMs were just kind of starting to get going. And so we said, hey, wow, wait, that's cool. Large language model. That sounds perfect for reading an insurance plan. And so we've used large language models to read our insurance plans for the past two and a half, three years. Um, we use large language models in the claims ingestion piece to doing a lot of what we call the the ontology work, which is understanding what treatments are represented by what claims. Um, and as we build moving forwards, we're going to use AI in a lot of the deployment automation. And then some of the I like to call it user experience connective tissue. Take, for example, when the insurance company doesn't return data about the patient's insurance plan electronically. Well, status quo, that is someone at the medical practice that has to pick up the phone for 20 minutes and say, hello, um, XYZ and Choco, is this patient active? What is their copay? That should be a robot. Frankly, that should be a robot talking to a robot. And so we will build that robot.

SPEAKER_00 40:04

No, I love that. You know, I love your take on AI because I feel like a lot of the emphasis is on 6% of healthcare costs, which is the cost of physician salaries. And everybody's like, let's cut these down. I'm like, okay, you're avoiding 94% of the cost. One third of the cost in healthcare is at hospitals. Hospitals, the inflation rate of hospitals is insane. That is where the money is. And the same thing with disintermediating all of these complex businesses that I literally, before I started reading about this, didn't even know existed. Repricers and all these different people, you're like, who are these people? Why do we need them? You know, so those jobs to me, number one, that would help out and make the system more efficient. The second thing is, which people don't realize unless they're a physician or they're really in healthcare, every time you add a middle layer in healthcare, it's another weight on the back of physicians. It weighs us down. So it's not just that the system is more expensive. Every middleman, every extra administrator slows us down. We can see fewer patients. That's why we can't answer your call when you call in and say, hey, I have a question about my lab that just went to my portal and I don't know what the heck it means. Am I gonna die? All those things. And to me, there's a big physician shortage, but you go you also have physicians who spend at least probably a day and a half a week on nonsense. And if you could offload a lot of that nonsense, including bad electronic health records, which you know in the hospitals are terrible. Can you guys interact with Epic? Epic is famous for not letting any other programs interact with them. And if they do let them interact with them, like the AI scribes, they let them do it for a while, then they say, We're gonna have a competing product now.

SPEAKER_01 41:48

Right, right, right, right. Not yet. Um that that that they come soon. Right now, we focus on outpatient clinics, specialty groups, and cares. Um, hospitals are on the horizon.

SPEAKER_00 42:00

Sure. No, and that's very fair. I I like that. And how have the physician practices how have they liked your product? How have the docs liked it?

SPEAKER_01 42:11

Um thankfully, for well, us and the technology we're trying to spread, they like it a lot. Um, so I will say that it is entirely coincidental that we provide the amount of value that we do to practices. Because when we started, we were just consumer fundamentalists. We wanted to build that consumer experience, we had our eyes set on that. And so we said we will stop at nothing to make these prices so that patients can see prices and click by buttons. And then after we spent three years building that and we had to actually get the technology out there, we asked the question, okay, now what does this do for the practice economics? Um, because we we need them to use it. And it turns out that it is incredibly useful for practice economics. Um at the start, you know, we we kind of had a sense that it might be because you can't have a buyer-seller market where the seller doesn't know their own prices that's going to break things fundamentally without even knowing how it breaks things fundamentally. And well, when we deployed, what we discovered is that well, most practices have 20 to 50 percent patient default sitting on their balance sheet. Because when the bill is sent after the fact, um, A, the patient has moved on with their lives, B, the number feels unfair. The patient didn't agree to pay that amount, the patient had no idea they're going to be paying. Um and so 20 to 50% of the time, the patient just doesn't pay. And what do most practices do? They either write it off immediately or they make one or two attempts to make a phone call and then they write it off or sell it to collections. And so practices are just losing those dollars, and that can be anything from hundreds of thousands to multiple millions of dollars every single year. Now, what we do is just really straightforward. We make the price before the service, and so we allow the practice to ask. And I want to be clear about this. We're not forcing anyone to do anything. We're not saying patient has to pay or no service is delivered. That is always the practice's prerogative. The practice can always say, look, you can't afford it, you've got your wallet, we'll see you anyways. And we encourage that behavior. We shift is just the ability to ask. Before the service takes place, the patient receives a text message that shows them their price. And if the patient ignores that text message, then when they show up in the office, the front desk admin can say, This is your price today, can you pay it? And when the practice just asks, it turns out the patients pay, depending on the specialty and the location, either north of 80% of the time, sometimes as high as 95% of the time. And so accordingly, practices make anything from 20% to 100% more inpatient revenue. And they're getting their money six to eight months sooner. Because, and I think most practice owners would know this all too well, for the services delivered over the course of January with conventional patient billing, those dollars will be flowing in until August.

SPEAKER_00 44:44

Yeah.

SPEAKER_01 44:45

And when the money is paid in advance, that's not the case.

SPEAKER_00 44:56

Now this sounds really bizarre. It's only in healthcare. If I as a doctor said, you know what, I feel bad for these Medicare patients or Medicaid patients, I don't think they can afford these co-pays, I'm just going to waive them all. If I do that, I could be charged with the ant under the anti-kickback statute and false claims act under Medicare fraud. I could be sent to prison, fined, and kicked out of those programs. On the commercial side, if I don't make a good effort, and they, you know, they'll audit you periodically to collect co-pays, they can go back and call back all the money they paid me. So there's a huge issue with these co-pays. It's you know really a choke point for practices. So I really have to give you a lot of credit for being a non-American. You come over from Cape Town, and then for being a non-healthcare person to find this real pain point for practices. That's great that you found that out and have offered such a great solution. I think that's wonderful. And I I put a lot of my waking hours into really pushing people to say, to think differently about insurance. I think we're on the same line with that, of that should really be for catastrophic events only. And you should pay out of pocket. You know, in the 1960s, Americans paid 50% of the healthcare out of pocket. And I was a child, and you go to the doctor's office, you just write a check, pay them right at time of service. And it was so simple and it was so cheap. And even now, if you cash paid for physician services, it's going to be less than it would cost you to take your special partner out to a nice dinner in New York City. I mean, so it's like, why am I insuring this? And and most prescriptions are $6 prescriptions, and you know, we go through all these things, and then we're getting prior authorizations for scans that you can get for $200 if you're a good shopper. So that's the part that's crazy to me.

SPEAKER_01 46:51

I I'm so on the same page as you, uh philosophically completely aligned. When I think about the bright future for US healthcare, it's a picture where everyone is on a catastrophic only insurance plan and the insurance companies have nothing to say about the proactive care that patients consume. Um, because exactly to your point, cash rates are sane, they don't need to be covered. And if you fix the healthcare pricing mechanism, those rates for proactive services will be sane across the board, not just on the cash side. Um, and so you know, when when I think about the kind of things we can do once we hit our network state, a simple one is I'm sure you've seen the Spotify rewind before by the end of the year. Spotify says, you know, these are your top artists, etc. People love them. You can send people an insurance rewind. You spent $14,000 on your healthcare this year. Most people say, What? $14,000? I spent $1,500 at the most. And then you say, Yeah, that's true, but $12,500 came out of your paycheck and insurance premiums. And if you were on a catastrophic plan, you would have spent $3K on your healthcare this year. And that's how you start to create the behavior and get people to swap over. And then we can fix the broken incentives from the ground up. Because for what it's worth, there should not be a baked-in disincentive for people both consuming proactive care in the market. That makes no sense. That's why people keep getting sick.

SPEAKER_00 48:08

Um, I love that. I also think it'd be really cool if an employer is a self-funded employer would just say to patients, you know, we're gonna give you a couple of options here. You know, where you have this plan where it's, you know, because the average family plan now is about $30,000 in premiums and deductibles. You know, we say we can offer you that, or you know, you can have the option of doing other things and we'll give you the money. Once people start realizing, wait a minute, this is my money? Because right now they're thinking, well, the company's paying it, I don't care. You're like, okay, and then the next day they're like, but I'm not getting a raise. I don't know what's going on with this company. And you know, and that's what happened. And I know a lot of I'm friends with a lot of employers and they struggle with this. They in the the town I'm in, a lot of these employers cover 100% of it, but then they can't give raises out and they struggle with it. And I feel terrible for and the employees don't see that, they don't see that struggle, and they feel frustrated because they're caught in the middle. But you're right. I mean, the average person spends a fraction of the premiums. I mean, if you're young and healthy and you're looking at, you know, $24,000 a year in cost, you're like, I'm gonna spend a thousand at most. I can go a couple urgent care visits, you know, I can get you know very inexpensive things. And I think that's why plans like crowd health are taking off. If people are young and healthy, they can really use that to get much better prices than you could for either in traditional insurance or but I agree with you. I think the whole game has to be up. We and a lot of it, and this kind of what drives me is just educating the public, just saying you don't have to accept this. And educating physicians, you don't have to be employed, you don't have to go work at a hospital. And I love and I love your whole attitude. It's like you you know you determine your fate, it's not determined for you. You don't have to go to a residency and go work for a hospital and just be this drone your whole life. You know, you you can do direct patient care, you can do cash pay, and you can do whatever you want then. Then you've opted out of the system, you can treat people the way they should be treated.

SPEAKER_01 50:13

Completely. You know, you you you mentioned um the contractual obligations physicians have to charge those co-pays and coinsurances. And we always find that one so tricky because we're very aware, you know, we we we we've heard this so many times speaking to to physicians, you know, we have to charge the code basis. And why? Because the the insurance company wants there to be a disincentive um for the patient to consume care. Yes, and trying to force that disincentive. And yeah, we speak to so many physicians who say, you know, the patient comes in, they clearly can't afford it. There, they say, please, I I can't pay for this, and the physician waves it. And that's a good thing. That that is how the system should work. Um, but today folks just have their hands tied uh contractually, and frankly, they have to break rules to do the right thing because the system is well, frankly, um it's prohibitive. It's it's it's it's overly regulated, and what we really need is a dose of the uh free market capitalism that makes this country work so well. Um, we need it in this category too.

SPEAKER_00 51:13

I I agree a hundred percent with that. Is there anything else as we're winding down that you would like to add? Something that I didn't ask you about.

SPEAKER_01 51:21

You know, one final piece. Um I think we were naturally building to it. You were saying master of your own destiny. Uh this is very much at the core of the way we think about things. We like to talk about how once upon a time being a physician was fundamentally an entrepreneurial pursuit. Um, you were always a small or large business owner if you open to medical practice. And today, most physicians are salaried, as you point out. And it's because, frankly, the system's become too complicated. Not because they don't want to be business owners, because the system is too complicated, they don't want to become full-time administrators, um, they want to be physicians, and that seems like the only pathway. When we set out to build, we spent a lot of time thinking about you know what actually makes a company a company. And for what is worth my failure mode and the failure mode of a number of folks on the team is that I tend to sometimes spend too much time in the theory and I'm not applied enough. And for me, the company journey was learning how to become progressive or applied to just take the things that computer around them. But in that period of theory, um, one idea that I came to that I hold quite closely is what is a company actually? Um, is it the product, is it the people, is it the technology, is it the combination of those things? Um, I think it is neither, though all of those things are reflections of what the company is. The company is at its core, a deeply wound bundle of philosophy and ideology. And that deeply wound bundle is actually ineffable. You can't take it and put it on a page and say, oh, that's the company. It's this core, um, you can see the distortions it creates in the world, almost like a singularity, um, but you can't poke into it exactly. And yet when it's real and the people feel it, they all know when it's when they're in its presence. And then um when it changes things in the world, when it creates those distortions, those distortions always map or mirror what that ideology, what that bundle is. I hope it has come across in this conversation. Superscript, you know, if you take away the pricing technology, you take away what I told you about, the plans, etc. What we're really here to do is to bring freedom to this market. Um, that's what we believe in. We believe in sophistication of people, physicians and patients. We believe in freedom. We believe that people should have the ability to solve determined. We believe that people should have the right to access this information. And so, yes, we can talk about the tactics, and the tactics are really important. Zooming out in 10 years' time, when we have a conversation and say, well, what did we actually end up doing here? It will be that. It will be directionally that, and that I can assure you that I can promise you. Um, because that's what we do here, and that's what everyone at the company set out to do.

SPEAKER_00 53:59

So that was a great close. I love that. And I couldn't answer a better close. And I wasn't sure what you're gonna come back at me with, but that that was phenomenal. So I really want to thank you for being on here, and I've learned a lot, and that that's the most humbling thing about healthcare policy. And I've been in healthcare for 40 years and I'm learning a lot. I feel like a newbie someday. So that's phenomenal. And I want to thank all the listeners if you made it this far. And please like and follow for more. Let us know in the comments any thoughts that you have and let us know what struggles you've had with hospital pricing. And I would add, and I agree 100%, I was thinking about that yesterday, is the whole game is to make it so painful for people to go to the emergency room or go to the hospital. Because you go to the emergency room, you just get this cascading of bills that keep coming. You know, sometimes it's months later, and you're like, you think you're finally done with an episode, and you've got very little actually care for that episode, which is the frustrating part. You're like, they hardly did anything. And it certainly is that disincentive. So we definitely have to turn the system on its head. I have complete faith. I think you've started your first of your 12 Herculane tasks in healthcare, and I'm excited to see what the other 11 are going to be. So thanks everybody for listening, and we'll see you next episode.