"We are building an intelligent automation platform that seeks to transform how patients access care"

Q: Pranay, tell us a bit about your background. How did you start Notable Health? What does it do?

Pranay: At Notable Health, we’re building the intelligent automation platform for healthcare; one that fundamentally seeks to transform how patients access care and how health systems get paid for providing that care. Very early on, when we studied healthcare, what we actually found was from the time that a patient, a mere mortal like myself, actually seeks care to the time that that care is built, it takes anywhere between 22 to 35 sets of hands that are touching data in some way. It’s old, archaic, and manual.

It turns out EHRs digitize the health records but not the workflow around those at all. At Notable, we believe that healthcare workflow automation is going to be just as important over the next 5 to 10 years as ours have been over the last 30 years. That’s what Notable does.

A big part of the thesis was actually bringing our experience from our scars, skills, and priors from fintech. We spent 15 years transforming what it looks like to actually do your taxes, to use Mint — and quicken on the personal finance side — the check scanning capabilities that you may have used way back when, all the way to powering a large portion of this country’s mortgages and transforming that to a seven-minutes’ experience in the palm of your hand.

We’re bringing a lot of that experience in healthcare. It’s just been an incredible five-and-a-half years of growth for us.

Q: Automation is a hot topic. Can you share your take on where you see automation gaining the most traction? Where’s a health system looking to deploy it the most, today? Can you share your own client examples, to illustrate the point?

Pranay: The insight that we’ve had in working with health systems across the industry is kind of like with AI. Automation is another one of those buzzwords that people want to embrace but have no idea of how to start or from where. We’ve seen health systems that have created Centers of Excellence and people that are trying to figure out what use cases to actually support! But what we found is, it falls into one of three camps.

There’s a camp where it is about – “I want to build it all internally and hope that I can learn from others.” What they tend to actually avoid or ignore is the total cost of ownership. How do you actually maintain that?

There’s a second that is around – “How do I automate the mundane to reduce costs?” It usually turns out to be garbage-in-garbage-out. I’m starting on the backend and it’s low value. Perhaps it’s easy to automate but it probably shouldn’t be needed in the future with where the puck is going.

There’s a third which looks at what the market trends are, strategically, and how automation may be used for growth. Truth be told, most systems aren’t actually thinking about it that way. They’re thinking about it more as a cost-cutting endeavor and less as how to actually tackle a change in how patients are engaging with our health system. Nobody wants to call anymore. We fundamentally believe we want to eliminate the call center from the US healthcare system. That has to happen. It’s inevitable and maybe, it’ll be 1-3% of phone calls, but you don’t need the vast majority that exists today.

The second part of the growth story is, how do you actually start on the patient access side? So much of healthcare data and workflow starts with — Who is the patient? Why are they coming in? Who’s going to pay for this? With that in mind, we’ve started thinking through that strategy with our partners across different EHRs – Epic, Cerner, Athena etc. But really, we’re looking at what the patient flow looks like. So, how do you engage with the right patient at the right time? How do you collect their insurance information in a delightful way just like we’ve been used to doing in other industries?

What we’ve seen with starting on the frontend and with the access side is, if you do that right, you can engage 80-85-90% of your patients in a digital manner. That is what upstarts around are trying to do. There’s no new health tech startup that provides care that starts with a call center and yet, that’s how healthcare does it.

For us, the places where we can educate our health system partners so as to partner with them truly lies in answering, how do you start on the frontend? We tend to actually partner with operations, red cycle population, and health leaders on — how do you digitize your patient experience to delight them and your staff?

The examples that I share with our partners — be it a large health system in Utah, one in Austin, or here in California — is we’re actually seeing upwards of 80% of patients engaging with health systems digitally. We get feedback from health system staff where their workload is actually reduced by 50% on a Monday, on things that they didn’t have to do — either outbound phone calls to collect registration data, or clinical information from patients, or even have themselves scheduled or rescheduled.

Most importantly, we actually see just elimination of backend workloads — What if you had no follow-up queue? What if you didn’t have a slew of calls that needed to be made to reschedule a patient? All of that starts getting eliminated when you start out on the frontend. We see our role today, in healthcare, as actually helping educate.

One of the things that we’ve done therefore, is codify all of this into something new that we’ve actually rolled out with our partners. We call it the Notable Health Check. I’m really excited about this because what it allows us to do is understand and assess where our partners are on their digitization journey. We understand what their tools and landscape look like and make recommendations of where they would actually get value if they thought about automation and digitizing the patient experience the right way and not just for randomness, like Robotic Process Automation (RPA).

Q: RPA specifically connotes automation of tasks and workflows in ways that essentially replace a human worker with a digital worker. It’s a subcategory within the automation landscape. Are you an RPA company? How do you categorize yourself when you look at automation technologies at large?

Pranay: We actually think RPA is fairly brittle. We don’t consider ourselves an RPA company. We use a lot of different capabilities to integrate. It turns out in certain cases, using APIs are great.

I want to access clinical data using APIs with Fire, App Orchard, or what Cerner has with their code program etc. We want to utilize all of those and we do. We augment that with machine vision capabilities to actually integrate where there aren’t APIs.

We also use a variety of other mechanisms to actually integrate because what we found very early on is, what’s critical in healthcare — the road to purgatory in health tech — is driven by integration. It takes too long to deploy something and then know if it is of value or not. From the time that you actually can get that feedback loop closed, there’s a reorg that’s happened. There’s a change in priority. More importantly, the market has moved on.

What we found is, it’s important to marry — when we talk about automation — a beautiful patient experience that is ADA-compliant in any language that works on any device, for any human on the planet. We spend a significant amount of time on design and how to think about that. To anyone that says the elderly cannot use technology, I like to say they haven’t seen the 60-70-80% conversion rates that we actually see in the 65 + cohort at Notable.

That’s because you sweat the details on who’s engaging, when they engage, and how to engage with them. You marry that with the best workflow integration. It’s not just swivel chair RPA. It’s actually rethinking the workflow in how you collect better data from patients to power clinical and administrative workflows. That’s the industry that we are in. We’re not in just the RPA. If anything, that needs to be commoditized and useless.

Q: What does your competitive landscape look like? There’s the startup ecosystem vs. the traditional EHR companies vs. the enterprise class companies that are non-EHR but have a powerful enterprise class workflow platform.

Pranay: I’d say healthcare is a really noisy landscape of every industry that I’ve been at. There’s a point solution for absolutely everything. The reality is, it turns out nobody wants a solution. They’ve already got so many in their ecosystem that they’re trying to get to work together. The patients and staff see that at the seams they don’t interoperate and they aren’t as seamless as we would want them to be.

For us, there are two parts that I like to think about.

One, how do we set the bar with our partners in digitizing every experience? The way that we actually go about it, on the access side, entails running our health check process to understand how many patients are digital, today. What does your website look like? How are patients calling? What does the call center volume look like? How many payments are being collected? How many denials are occurring on the backend?

By collecting that information, we’re able to share a very strategic solution. That isn’t about how do I bolt on here? How do I bolt on there? It’s really one that’s focused on outcomes for our partners and often, that leads to sunsetting a lot of the points solution. You don’t need an appointment reminder vendor if you also use Notable. You don’t need a denials dashboard because we actually provide that capability. We start with the outcome and then, align strategically with our partners to actually set that up.

Now, there are certain systems of record. It could be the EHR vendor, CRM, or others that you actually want a deep integration with the power workflow or to collect data from because those are the single sources of truth. Those are the ones that we actually embed ourselves in deeply with. So, working with the EHR or the CRM vendors in many different ways resonates because if you approach a health system with this as the only way to deploy, it’s not going to work. Often, they don’t even know how to deploy something to get to success.

I’ll give you one example. We actually studied a part of our health check assessment – scheduling — and looked at this with a partner of ours in Kansas. What we found was, about 12% of their website traffic was coming from Google and then, just bouncing because they couldn’t find how to schedule. Fifty-five percent of their phone call volumes was actually tied to scheduling, of which 35 was inbound and then, 20 or so, was outbound because they had about 10 to 12% of their orders at any given point in time that were unfulfilled. They were trying to follow-up with patients to try to get them back in.

Now, once you have that data, you have strategies on how to deploy software and a platform that can actually tackle those. We gave them personalized links that they could embed onto the website for self-scheduling so that it integrates deeply with Google. We set up ways to actually engage with patients automatically based on their prior visits or when they were actually leaving the clinic. We were monitoring the orders that were placed in, checking to see when and if the patient had scheduled something in the future, texting them, and having them scheduled within a certain window as required.

It’s what the call center agent would do. But all of that is actually built on our no-code platform where you don’t need to write code for any of this. This is all point-and-click to set up that end-vision journey. What we saw, was tens of thousands of phone calls actually being avoided. You don’t actually create a faster call center, if you can read this. You completely avoid the call. We actually saw a patient satisfaction score of 98 and up to 40% of all of these health systems’ appointments actually being scheduled in a digital-first way. That was up from 5%. I share that because they could actually angle in 16 different point solutions.

In trying to pull that all together into EHR and CRM, to us, these are all the tools that bring these solutions to life. But you really have to think about how can you do this end-to-end and then, iterate really quickly. The time to actually deploying that value is key. We deploy that in six weeks. That allows us to learn and iterate. If you don’t have that kind of speed, it’s an 18-months’ journey to get your first piece of data and then, it’s not worth it.

Q: There are a number of new initiatives being launched by the big technology vendors, for instance, Epic’s Cheers CRM platform. You’re a big technology firm, too. How are you seeing your clients? Do the trade-offs matter when they have to choose between a truly innovative solution from you vs. one that may not be best-in-class yet easier to deploy, integrate, work with and have lower overheads?

Pranay: I would challenge that belief. But one fascinating point about healthcare in every sense is, every vertical has its system of record. I think healthcare is the only vertical where the system of record is actually, in certain cases, dictating what you can and can’t do.

That’s flawed in how people actually think about running their health system. I’d challenge the assumption that it might be easier or cheaper to actually deploy because what we’ve seen is, the only path to success is iteration. When we partner a part of that health check process that we go through, we actually assess how well an organization adopts Agile. We ask — Do you know what a product owner is? Are we going to get stuck in committees before we can actually make decisions? If that’s the case, you might have all the willpower in the world but you don’t have the ability yet. We need to actually help you understand what iteration look looks like. What does a two week deploy look like?

There was a CIO recently that asked me, “What does your release cadence look like?” It was 11 a.m. in the morning and I just went on Slack. We have a channel called the Dev Deploys and I was counting quietly. The response was, “Well, it’s 11 a.m. We’ve done 56. It deploys this morning, and that’s just how we work.” And he asked, “Oh, you do? No downtime deploys.” I guess that’s how you build a true technology iterative cycle such that you can actually learn through this.

The reason that I share those stories is, when we sit down with CIOs and technology leaders, we try to decipher, are you a technology leader that wants to play it safe and do the least amount of change possible because that’s actually how you see yourself succeeding? If yes, that’s okay. I would rather know that up-front and not waste each other’s time. Or, are you the technology leader that aligns with the business to drive change? We tend to work best with the latter, not the former.

With the latter, we tend to actually set the bar on — How can you actually reduce the total cost of ownership? How can you actually set up? We always think about it as the EHR should exist; it’s the system of record, so, it will always be there. The system of automation that actually sits on top of that and transforms patient to payer or booking the bill, that’s who we are. That’s the role that we play.

Then, there’s a number of different workflows — everything from outbound scheduling to prior off, checking your prior status and texting the patient on their status or having them reschedule – all of those are intertwined flows. They should not be point solutions across the board. When you start to think about it that way, you get a lower total cost of ownership and higher value much faster than when the legacy systems might actually try to try to mimic what it is that we’re doing.

Fun fact — there may or may not be a legacy system that is a monthly meeting on trying to copy Notable’s features. We love that because we’re actually helping move the industry forward, and that’s the role that we have to play.

Q: There is an emergence of a mindset around treating digital health as a product management function. How do you create value for the end customer? How do you back into what you need to do to build it out?

Pranay: Can I just add to that with an anecdotal example? We’ve actually seen partners start out wanting to build themselves and then, very quickly realize that they’re not an R&D shop and they won’t have the resources. There’s only one master that you can have. Still, we have to share and educate. Sometimes we tend to think that everything is a form and it’s just a text message and a dropdown. Patients can do X, Y, Z, and I have the tools to do that.

But where we actually lack is in that product function of thinking through deeply. How do you productize intelligence at scale? With the world that we’re in, the example that I share is, we started early on with just scanning insurance cards. What we found was, we could actually do it at 95, 97% accuracy versus a human that would transpose zeros and ones.

But the bigger challenge actually came in matching it to a payer plan. It’s that that denial on the backend doesn’t happen. Two years ago, when we first started to look at this, health systems would come to us and say, “Hey, here’s the rule.” We’d look for this PO box number and contort our bodies, do the rain dance, and hopefully, take the right payer plan up.

To productize that in scale, we determined that it was not something that we should just ask every customer. We actually needed to think about how to instrument the platform to learn from what their staff was doing and automatically create a feedback loop that could actually scale such that we could reduce and eliminate their eligibility-based denials on the backend. Those are some of the places where productizing AI and automation is hard. It isn’t just like the swivel chair RPA or, you know, that my portal can do this.

Q: You’re in the middle of the entire startup ecosystem and have raised a lot of funding for your company. But, it’s uncertain times for digital health funding at large. What does that mean for a digital health startup and what’s your advice to them? How are you preparing your own company to work through the next four months or longer?

Pranay: We talk about this as a team. Our focus is on building a company that outlives us. There’s so much to be done in an industry that has been severely underserved and in really building something that we are proud of.

On the external side, the cost of capital is changing and that just pushes for sustainable growth in a reliable fashion. Internally, our focus is always the same. We’re fortunate enough where people ask me about runway and things of that sort. We are not cash strapped in any way in terms of the revenue that we’re making and the growth that we see in the tailwinds that we have with the industry needing what we provide is great.

However, our focus is just maniacally on what that patient feedback looks like. Every patient that actually uses the Notable interface has their feedback streaming into Slack for the entire company to see. That’s what drives us. When you see somebody that gives you positive feedback and says, “I’m a blind man that did this using voiceover”, that’s compelling. That’s reason to exist.

On the flip side, when you actually see our class report with 100% of our customers recommending us, that’s not to say every deploy is swimmingly smooth, but you can go in with your scars and what you’ve learned. Talk to any one of these customers and they would still recommend working with us. At the end of the day, that’s where value is created. For us, that is our focus.

For anyone listening to this and building a business in healthcare, I’ll say, you got to focus on your customers. At the end of the day, I think that drives you. That is our maniacal focus across the board.

We hope you enjoyed this podcast. Subscribe to our podcast series at  www.thebigunlock.com  and write to us at  info@thebigunlock.com

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity