2026.10.07 | TB | Healthcare | N-Power Medicine Primer

Primer • 11 Sep 2026

Expert Former Senior Executive in Sales & Marketing

Gaurika Bansal (Analyst) — Third Bridge Sector Analyst

Agenda

• Over 20 years' experience in the clinical research and diagnostics space
• Over three years' experience at N-Power Medicine (left within the last three years)

Introduction

Analyst: Could you state I agree or I disagree to the following statement. You understand the definition of material non-public information and agree not to disclose any such information, or any other information which is confidential, during the Primer Interview today.

Expert: I agree.

Questions

Could you describe what N-Power Medicine does? What products and services does the company provide?

Expert: N-Power Medicine's main business is to develop high-quality prospective external control arms to support regulatory filings. I assume it also supports HDA, so reimbursement as well, but they have not stated that. Again, that's a use case for that solution. I just don't know if they have explored that, but it's mainly for regulatory purposes.

Analyst: What products and services would you say the company provides?

Expert: It's this external control arm is the main, I mean, in some ways, it's a service. It's hybrid. But, yes, it's an external control arm to supplement or support registration of novel drugs.

Analyst: Might there be any other product for the company?

Expert: Part of the value of the solution is that you have to standardise and structure data in a high-quality manner. In doing so, that will definitely open up other business opportunities, whether it's training AI models or using it for clinical trial matching, commercial applications. There's potential and once you have high-quality clinical data, there's a lot of opportunities to use that. Again, maintaining confidentiality of what we have discussed early on with the company and what they have said publicly, it's really around this registration. Definitely, if you just think about it, there's such a gap in high-quality clinical data, standardised, structured. There are many, many applications to that, which could be considered.

What would you say is the value proposition of N-Power's products or services? What problem is the company looking to solve?

Expert: I mean, I think today, you have to almost look at the two extremes. On one hand, you have randomised clinical trials where the cost and timing to enrol patients on clinical trials is extremely expensive. I have heard numbers run anywhere from, depending and again, I'm mainly focused on oncology and N-Power's mainly focused on oncology. It's upwards of USD 200,000-300,000 per patient, depending on phase 1, phase 2 or phase 3. To standardise the data for a registrational clinical trial is extremely expensive and time-consuming. It's challenging because only 0-10% of patients participate in clinical trials. Then, at the other extreme, you have real-world data, which is extracted from claims data or EMR data. The challenge is that those data are extremely messy and incomplete. It's often collected more for reimbursement purposes than for clinical or registrational purposes. There's this big gap between real-world data and clinical trial data. I think the value proposition that N-Power is pursuing is, "Can you define a middle ground where the data is high quality enough to support health authority requirements," but not the level that's done today for randomised registrational clinical trials.

What would you say is the full range of customer segments that could use N-Power's products or services?

Expert: Realistically, I assume it's going to be the top 40 pharmaceutical companies and probably the top five or so CROs would probably be the ones that would probably be most able. Because obviously, to build this, it's so expensive. The value and the costs are going to be quite high. You would have to have companies that would have the ability to cover that. I think there are other potential opportunities, with potentially smaller biotech as well, but then there would be a different value proposition in terms of how that is financed. Whether it's done through some type of equity or other type of deal. Again, there might be other opportunities. Again, it wouldn't be paying for the services in the same way, but potentially be able to leverage this to establish a proof-of-concept or do something more efficiently.

Analyst: Might smaller biotechs be potential customer segments or are they currently present in the company?

Expert: What I have seen publicly, I think it's mainly still the top category. It's not so much in the smaller biotech. Again, there would need to be a novel business model, unless maybe like a VC or someone else. Maybe that would be the way that potentially VC would fund it or do something along those lines or there is some type of partnership with the big pharma. The funding, again, the funding likely would be beyond the reach of some small biotech unless, again, there's a pathway. Where external control arms have been used, there have been a few examples of external control arms being accepted by regulators and they have always been in the case of very niche rare diseases or like Pfizer used it once for male breast cancer. They had a drug that was for breast cancer and then they were able to do it for rare male breast cancer, which is a very orphan indication. Again, those might be opportunities depending on, again, the scale of the network and the indication.

Analyst: Could you define big pharmaceutical companies or CROs [contract research organisations] for the company using any quantitative metrics? What's the scale of such companies?

Expert: I think, again, it's maybe companies that directly fund registrational phase 3 trials, that would be the requirement. Obviously, some of the companies are smaller, not able to do that. They need to partner. Any company that has the ability for the indication they're pursuing to support a phase 3 clinical trial would be the one that would warrant. Again, it's so easy to say top 20 pharma companies. You have the Lilly's, the Pfizers, the J&Js, the Roches of the world. I don't know how far that extends, but I'm sure there's companies that are next tier. They have market caps well over USD 1bn, probably even a multi, couple multi-billion market cap would probably be likely. Again, maybe the ones who are directly funding a registrational phase 3 would be a bar that would probably be there. Potentially, phase 2. Then, I think there would need to be some more creative financing to support that. But like risk share. I mean, I think that's sometimes the opportunity, potentially risk-sharing or some other opportunity, rather than just a fee-for-service-type business more of a collaboration, maybe with N-Power or another company.

Analyst: Could you quantify smaller biotech firms? What might be its scale?

Expert: I mean, I think one is that they would have to have funding to support late-stage development so that they're not just focused on some early-stage assets. The ability, they had enough funding. I don't know what that funding level is. What it's like, USD 400m-500m or something, to, again, develop a product beyond phase 1 so that they could conduct a phase 2 study or they have a broader presence to bring something to late-stage clinical development.

Could you profile N-Power's typical target customer segment within the discussed range? What might be the customer segment sweet spot based on size?

Expert: I think it's one of these larger pharmaceutical companies that probably have one or two use cases. One, it's a very competitive space. Like, increasingly, with so many molecules being developed in China these days, most of the targets, the biological targets for drugs, are often very competitive very early. Someone who's trying to get leverage in terms of speed, speed-to-market, to try to be first to market or to catch up or again, where speed and cost matter. That's probably one situation.

The other situation is more, again, rare or novel situations. Like I think the example with Pfizer in male breast cancer was a sort of perfect example. Something, maybe it's a line extension or something, where it's not realistic to be able to do a prospective randomised clinical trial, that you ideally would do a single-arm study or you would do something that would allow regulators, because the use of external control arms is best supported in more of these niche cases. I think doing something more mainstream where there are patients available is a little bit more challenging.

Analyst: Could you profile the target customers across the customer segments discussed?

Expert: It would definitely be big pharma again because they have the resources and the ability. Again, the other examples historically where pharma companies have pursued this often have not been successful. Again, you would have to have the risk appetite and the ability to mitigate, if it didn't work, to take another pathway.

Analyst: What might be the revenue split across the customer segments, such as big pharma, CROs and smaller biotech?

Expert: Today. I would say it's probably 91-100% pharma.

Analyst: Might the remaining contribution be from CROs?

Expert: I'm trying to figure out what they have said publicly and also any gaps. They had a poster at ASCO last year. The registry they have is called the Kaleido Registry. They had, in that abstract, listed that they had seven studies and they talked about the indications. I think that's a good indicator. Actually, if I had to guess, I wouldn't say any CROs are involved. I think they just, again, publicly announced in their series B that Labcorp, which is obviously a CRO as well, was one of the funders. Obviously, Labcorp has a CRO, but I'd be surprised if they have a direct CRO customer today. I think it'd probably be more likely with a biotech. It could easily be 91-100% pharma. If it's not pharma, I would assume a small biotech. Either there was an opportunity for N-Power as a proof-of-concept or there was some risk-sharing or collaboration-type model would probably be more likely than the CRO today. I think over time, CROs might see it better because it might help them catch up or recruit or anything else. Typically, I would say the pharma could be 91-100%. Then, today, it would probably be a small biotech if it wasn't a pharma company.

Can you share any customer names that are either on N-Power's website or in the public domain?

Expert: The only one that I actually have seen and trying to look and again, is that this isn't directly saying that they're working with them. Merck GHI was one of the funders and there was an announcement about Merck GHI, which is Merck's venture arm. Historically, when Merck invests with their venture arm, they also usually have a record of having it supporting Merck. Merck GHI is Merck Sharp & Dohme for Europe, but in the US, Merck. That's their venture arm. It's a global health initiative. Typically, when Merck's venture arm funds something, they also have activity with the start-up that they're working with.

Analyst: Are there any other customers that are mentioned in the public domain?

Expert: I have not seen any other in the public domain that have come up. That's the only one that I have seen. I looked. One thing in that poster from ASCO, it did look like an indication. Again, I think that gives you some clue of which companies may be involved.

What do you think are the key countries that N-Power derives its revenue from?

Expert: I think right now and again from the registry, so far, the registry has been North America, I think. Maybe even just the US. I'm not even sure that they have anything in Canada. It might just be the US. I think that's where the registry has been. There's easily the ability to apply this business and it could have even more value ex-US. Today, I believe it's very US-focused.

Would you say that the US contributes 91-100% towards N-Power's total revenue?

Expert: Yes, I'd probably say that right now.

Analyst: What would you say are the top three revenue-generating states within the US for the company?

Expert: The challenge is that until you have the success of having a registration, a lot of it is survival to make sure you have generated revenue to support the business model as you grow. Because obviously, part of it's coming from external funding, whether it's VC funding or another type of funding. Then, as much as you offset that with revenue, I think it's just trying to be more opportunistic in terms of what you could do. Again, it falls more on the real-world data and then achieving milestones on the path to these other projects.

Analyst: Are there any three top-performing US states that contribute the most revenue?

Expert: It's all everything we have been discussing already. A lot of it is getting to this prospective external control arm. That would be the majority of the revenue. Again, a lot of it is, since the data collection mechanism and machinery is so high, anything you can do to offset those costs with insight generation, real-world data projects or other things would be valuable.

What are N-Power's core revenue-generating segments? What products and services does the company offer?

Expert: Again, I think they're really focused. I mean, I think early on, there were ideas to invest in clinical trial matching or some of these other businesses and they potentially could do real-world data cohorts. I think a lot of it is working on, again, prospective external control arms that are correlated to a registrational trial. I mean, I would probably [mention] that at 71-80%. Maybe it's not unreasonable to assume that there's some other activity that might be 21-30%, which would be more traditional real-world data, which could be, maybe you're supporting some data collection or some insights about a disease area. I'd probably [mention] that at 21-30%.

What might be the relative revenue size of each segment? Might prospective external control arms contribute 71-80%?

Expert: Yes, I think that's probably right. Then, 21-30% [real-world data]. These are total guesses, because, again, once you have the registry, you have this high-quality data. Again, if someone was trying to understand something, that would be a good source. A lot of the data you have access to, you're buying something you think is going to be very positive. Then, once you get into the details, you're like, this is not usable because there's just too much missing information. You can't filter it and look at and answer the right questions. Again, with the registry, they have very systematic and high-level data, which has value in itself. I don't know exactly how to monetise it. I know the company has been careful about selling data based on how they're working with the practices. I know the practices they work with. There are clinical practices today that don't want to sell their data. I think you almost have to look at cohorts as opposed to selling data. A lot of the competitive companies actually sell data and a lot of the clinicians who are working in practice don't want their data necessarily to be sold.

What is the revenue model for each of N-Power's segments – transactional, recurring or reoccurring? How is the company earning money from each segment, starting with prospective external control arms? What is its pricing basis and payment frequency?

Expert: I assume it's milestone-driven, to hit certain milestones based on recruitment or getting to certain regulatory or other milestones. I think that's likely. It could be per patient, but I think it's going to be some combination of per patient and achievement of milestones. Because again, for an external control arm, it either works or it doesn't work. I think companies would be less inclined to pay for something that just doesn't work just for effort. There would have to be some funding to support the effort because it's obviously very expensive to collect these data and to support the work. On the other hand, I think they'd be very careful about supporting something that, if it's not accepted, it's not really worth anything to them.

Analyst: What would you say is the scenario for real-world data? How is the company earning money through the segment?

Expert: That could either be, again, I think it'll depend a lot on the customer, but it'd either be project-related. Again, more milestone-driven to get to a certain deliverable or potentially per patient type data. If they were looking for evidence on a certain thing, maybe per-patient evidence.

Analyst: What would you classify the revenue model as for prospective external control arm – transactional, reoccurring or recurring?

Expert: That's a really interesting question. I mean, just like most pharma companies today, at least, again, the customer segment that we're looking at is more franchise-based. They focus on disease-area franchises. I think the preference, again, the investment for them to build such a strong data asset would be that you could do either reoccurring or recurring model where you could use that asset over and over again. You can use it for more than one use case. I think that's probably more likely, but I think for some companies, again, it's because the model is a binary, it either works or doesn't work. I think a lot of them would start with a more transactional arrangement and would only commit after they saw that it was successful.

Analyst: What is the exact revenue model that N-Power follows to earn money from the segment?

Expert: I think it's one of the latter two. I think they're trying to, again, as you get yourself established, you accept what you can get to try to be more transactional. I think what you'd want is something more recurring that, again, you're building this asset that the company can continually use for different use cases. For example, if you used it, again, I don't even think we have had these conversations [in the past], to get registered, let's say you had a single-arm study that you registered. It got approved and used as an external control arm. That would be valuable and again, you achieve that. Then, there are obviously health authorities around the world. Would you use it within other health authorities? Then, after you got the drug approved in different geographies, then could you work that with payers as well? Once you build the asset, there are a lot of opportunities to leverage it for other applications, both in terms of registrational and even commercial applications because then, again, you have a good understanding of characteristics of patients, which could give you insights more than traditional real-world data. I do think it's recurring once you build a strong asset. Because such a high lift, the cost structure is so high per patient to build it, a company that invested to build that asset, they would want to use it over and over again.

Analyst: What would you say is the revenue model for real-world data?

Expert: I mean, it's a little bit the same thinking, but I'd say that's probably more transactional because I think the company has tried to really focus on the prospective external control arms. I think the real-world data would be just more to support cash flow and support expenses as they build the broader model.

What are the use cases for each of N-Power's segments, starting with prospective external control arms? What are its features and capabilities?

Expert: It's supporting registration of a drug with an NDA, for approval and acceleration on those programmes. Yes, it's acceleration and, essentially, even cost reduction because, if you had, let's say, you, regulator, agree that 41-50% of your patients for a control arm could come from an external control arm. Let's say your base case was that you had to have 200 patients in the trial, 100 in the experimental arm and 100 in the control arm. If a regulator accepted this model, potentially, you'd only need 150 patients. Recruiting 150 patients is a lot faster and less expensive than recruiting 200 patients. That would drive speed and cost reduction. Because as expensive as an external controlled arm is, it's still a lot cheaper than enrolling patients in a randomised clinical trial.

Analyst: What might be a use case for real-world data? What are its features and capabilities?

Expert: I think that real-world data is a well-established business model. I think that's just more to drive insights, both supporting research and/or evidence generation. I mean, actually, one other customer segment that I didn't mention above, which could be used, would be large diagnostic companies as well. Large diagnostic or digital companies would be other opportunities. Again, I don't think they exist today. I think those are definitely opportunities. Because again, if you had a test that you wanted to validate or show, like, a lot of times, it's driving clinical utility. Those could be opportunities that could be used.

Analyst: Might large diagnostic companies be the same size as the big pharma companies that you mentioned?

Expert: I mean, diagnostic companies are typically smaller. I think they're probably tenfold less big than the large pharma. They're probably more in size than some of these smaller biotech, maybe a little bit bigger.

Analyst: Are large diagnostic companies also potential customers, in your opinion? Do they currently not contribute to N-Power's revenue?

Expert: Not that I have seen. Not that there's been any indication, but it's definitely a model because gain, if you're collecting data, if you could collect data on a certain test, in practice, you could use that for validation and registration as well.

Analyst: Are there any use cases for the real-world data segment, apart from supporting research and evidence generation?

Expert: I mean, again, there are a lot of categories. It's the insights generation, again, which supports internal decision-making, potentially. Yes, maybe decision-making tools could be used. Again, if you have a new drug in a new area and you want to understand the unmet need, then you might say, "What are the outcomes of the patients today or what are some of the key characteristics of these patients?" Again, that could support your development because then you might build the design and protocol based on that. There's commercial insights and there's R&D insights and then there's decision-making that could be supported by that.

Analyst: Would you say that N-Power is quite aligned to be a CRO? Is it a CRO itself?

Expert: That's a really good and interesting question. Potentially, yes. If you had a segment of the market that could, again, enrol patients because, if you have a network of clinics and hospitals that are working there, the prospective external control is more passive, but you could also enrol patients because you're identifying patients regardless. You could actively enrol them too. That's more of a future thing. They would take more effort and there's obviously a lot of competition for CROs. Or it could be a hybrid as well because sometimes some pharma companies heavily outsource their work and some do it more internally. It might, for companies that focus more internally, like running their studies more internally, that they could supplement some of the services that are missing from a CRO because that's a big decision for companies is whether they work with the CRO or not. Again, maybe a company that's using CROs less may be more inclined to be able to leverage N-Power's model to supplement what they're doing vs some that totally outsource to CROs.

Analyst: Would you say that N-Power has some therapeutic focus, service areas or phase focus where the company operates?

Expert: Phase 1 would be really hard for them because it's really around safety and again, you're not necessarily those patients are enrolled. I think the near-term opportunity is definitely phase 2 because it's smaller. Most phase 2s tend to be single-arm, whereas phase 3s are typically randomised. I think phase 2 would definitely be much more of the sweet spot for this model.

Analyst: Could you elaborate on investigator payments, data management, outcomes recording and just the interoperability across systems, given that N-Power is involved in the EHR and real-world data aspect? Is it internal technological infrastructure? Does the company outsource it?

Expert: My guess is it's probably more internal. But again, I think they would if they're smart. This is something that evolved a lot in the early days. But yes, I think if they're smart, leveraging what you could use off the shelf, but the proprietary technology would more be internal. Then, one of the big challenges, again, in clinical research, in general, is that physician notes are horrible. They're messy, hard to read. It's hard and that's one of the challenges of real-world data. You have to standardise and structure their notes. Yes, even just some of the emerging clinical notes and/or templates or anything you can do to make the clinical data structured in routine care, is really essential for this to work. Initially, it's a very manual process. Over time, the more you can automate that, it's going to reduce the cost because collecting the data on every single patient is extremely expensive and you need a lot of people to do that. Over time, again, the more you can automate, streamline and everything else, it's a little bit of a differentiator.

Which of N-Power's two segments would you say is going to be fastest growing and could drive the largest share of revenue in future?

Expert: Definitely [prospective external control arm].

Could you name the industry that best describes N-Power's core operations?

Expert: That's a really interesting question that I know we struggled with a lot, because you're not a CRO. You're not a real-world data company and so you almost want to create your own category, which is the ideal type of approach. Other companies have used this wording, this regulatory-grade data or support or prospective registries or kind of anything along those spaces, but it is really trying to establish its own category. Because when we talk to competitors, a lot of them are more real-world data companies that are trying to elevate the quality of their data through other mechanisms to get to this regulatory-grade data. It's real-world regulatory-grade data or something around prospective registry. It's that spot between a CRO and a real-world data company.

Analyst: Might the real-world regulatory-grade data service be an appropriate categorisation?

Expert: Yes. I might evolve it as we discussed, but yes.

What has the AAGR been for the real-world regulatory-grade data service market over the past three years?

Expert: I think because it started from such a low base, I would say it's over 20%. Because again, if you're starting from such a low base, it's very easy to get over 20%.

What do you think have been the 1-2 key drivers of historical market growth?

Expert: I think a lot of it is companies and individuals like companies having the vision to try something. The ability, one is the customer need and then also the willingness to be innovative and try something new. I have a good analogy that's come to fruition that might be helpful. One market that's developed, it's going very well right now, is MRD. Have you done research with MRD? It's minimal residual disease. Basically, it's a biomarker to measure minimal residual disease. Why that's important is it's a surrogate endpoint for clinical trials. Most clinical trials, especially in oncology, traditionally, what regulators want is to get overall survival. You randomise your experimental and control arms and the endpoint would be overall survival. In many diseases, patients are living longer. Getting to overall survival takes many years. In some cases, to do a study might take 10 years, which is obviously very expensive. There's a big focus on using surrogate endpoints, where you can have an earlier endpoint that is correlated to survival and allows you to do a clinical trial faster because the readout is based on a different endpoint. There's a technology called MRD that's used and right now, there have been the first cases where studies are approved based on MRD rather than overall survival or progression-free survival and you can report out the study much earlier. This is the surrogate. I would maybe, instead of MRD, maybe say it's about the development of surrogate endpoints and then you could use example is MRD.

What do you think will be the market's AAGR over the next three years?

Expert: I mean, a lot of it, it's like, again, what I have said, it's like bimodal. It's like once I show, then it's like, it'll pick up because, again, it's such a driver of speed and cost. Again, I think it's going to be, until they have their first proof-of-concept that's accepted by a regulator, ideally, the FDA, it's going to be low-to-medium. It'd probably be in 6-10% or the more things get closer to where it's accepted, I think you're going to get, once it's shown, there's going to be a big growth. I think in the next three years, I'd say it's probably 11-20%, just to support the business model. But once there's evidence to show that it works, then I think that would open it up a lot after that.

What do you think will be the 1-2 key drivers of future market growth?

Expert: 100% of regulatory validation of the model.

Analyst: Is there any other key driver of future growth?

Expert: That one's so important that I'm not sure. It's not regulatory. It's really not of the, I said regulatory, not data validation. It's really regulatory. It's more a concept or regulatory acceptance, validation and acceptance because it's going to be expensive to build this. It's a lot cheaper and faster than doing a traditional study. If you can establish that, that's going to actually really drive growth and willingness. Even, I mean, one sign would be if an existing customer, if they were working in, let's say, colon cancer and then they did a lung cancer. Once they saw proof-of-concept, because typically, their current customers will probably get feedback first before it's visible by the market. If your company is doing that, you'd have to go to the FDA. If you showed them the study design and everything else and they were onboard and accepting it, then you would know that this is a viable market and then you'd be likely to add or initiate other programmes. Current customers would have that advantage because they would see the driver first. Then, I think the real growth one is there's a publicly disclosed model that supports that there's a public disclosure of the regulatory validation. First would be internal validation for a company for existing customers and then the other one would be a public disclosure.

To what degree do you think the real-world regulatory-grade data service industry is correlated to economic cyclicality? How might the industry perform in an economic downturn? Might it be counter-cyclical, very cyclical, somewhat cyclical, somewhat resilient or resilient?

Expert: Again, once you have hit the threshold of showing that it worked, I think it's very resilient because, again, there's no shortage of patients. There's lots of competition. Again, once you demonstrate value, I think it's very resilient because it's increasing speed and decreasing cost.

Based on the public information, what impactful technological evolutions or innovations should we anticipate in the industry over the next 3-5 years?

Expert: It's definitely AI-based. I mean, the data collection, AI-based data collection that can support high-quality, standardised clinical data is going to be the key.

Do you see any downside risks that might materially impact market growth over the next three years?

Expert: I think it's almost economically driven. You're building something so big and robust. Do you have enough customer support for the business model? I think the economics is the biggest downside risk that the cost of development exceeds the value. Because if you're a pharma company and you're like, "Okay, we can do a randomised control trial. It's going to cost us X dollars." Then, if you go to the N-Power model and you could do X minus 20-40%. There's a balance of risk and benefit. You can reduce, if you're doing a randomised control study, there's a lot less risk. But if you pursue the N-Power model, you have risk and cost. Wherever that breakeven is, is really key. Really understand the breakeven between cost and risk. I think the biggest risk for the model is that it becomes too expensive to administer. Yes, the data, the ability to make the model more efficient, cost-effective, is going to be really key for the ability to market the company to be sustainable.

Analyst: Would you say it's the same case with the whole market?

Expert: Well, in some ways, they're almost doing a more of the ground up and then N-Power's doing from top down in a way. N-Power's taking the approach of doing this extremely expensive, high-quality data collection model where a lot of real-world data companies were started with whatever they can and they were increasingly curating data to get it to be high quality. Again, the more they can maybe improve the quality of what they're doing, there's a convergence on that.

Analyst: Might high costs impact the whole market?

Expert: Right. Like again, how much of the AI data collection, structure, standardised data can be done using technology as opposed to people.

How fragmented would you consider the real-world regulatory-grade data service industry to be? Is it a monopoly? Does it have 2-5 strong players? Is it fragmented or highly fragmented?

Expert: I would say probably 2-5 strong players, again because it's, I mean, if you put it in the context of real-world data, there's hundreds, but I think the real ones, it's like cars. It's like there are a lot of cars out there, but then the ones that are very high-end, there are only a few. It's almost that's a good analogy. It's only so, again, if you talk about it in the car industry, there's tonnes of players. In this high-end data, there are only a few players.

How protected would you consider the real-world regulatory-grade data service market to be? What might be the barriers to entry in it?

Expert: In oncology, so much of it is longitudinal data. Again, if you have been doing this for three years and a new company starts, they only have six months of data where you have already three years of data. The longitudinal data capture is a big barrier to entry because the sooner you start, the more data you have. As long as you're collecting data, the use cases match. The more mature your data is in this standardised, structured way, that's a definitely barrier to entry because let's say you're doing something in lung cancer and you have three years and a new company comes on board and they're starting fresh, they would only have six months of data. The longitudinal aspect of the data is a big barrier to entry.

Analyst: Would you say that the barriers to entry are high or low in the industry?

Expert: I think it's high just based on cost.

What might be the barriers to scale in the market?

Expert: I think that scale is the number one challenge because again, I think with N-Power, the scale of the network is really key and then even barriers to scale because again, if you use an external control arm and this might be another, I mean, it's barrier to scale and entry is the globalised aspect of it. For a phase 2, you might be able to do an external control arm just on US patients. If you're doing a global study, a global approval, you're going to need to have representative data from around the world. You're going to need European patients. You're going to need Asian patients. You're going to need everything else. Again, I think having this global footprint is going to be really key for the eventual successful model because regulators are going to expect that diversity of patients and race, ethnicity, different aspects.

Who would you say are N-Power's top three competitors?

Expert: I think Flatiron is definitely one.

Analyst: Who might be the second competitor?

Expert: Probably ConcertAI.

Analyst: Who might be the third competitor?

Expert: You could go in a number of directions. Because there are certain companies that already have a network. If you have an existing network, you might be able to do that more easily. Then, you just have to collect the data that way. I think ConcertAI is number two. Probably McKesson could be one. There are three strands I'm thinking about, so it's hard. Why don't we do McKesson as the third one? McKesson's obviously is a huge multi-global company, so it's a division of McKesson. The fourth one, why don't we go with a company called Aetion?

What are N-Power, Flatiron, ConcertAI, McKesson and Aetion's market shares – less than 5%, 5-10% or higher?

Expert: I would say for today like, of the actual business for this application, I'd say they have a pretty low market share because I think the other players are more proven. Then, why don't we say 11-20%? [Flatiron], I think they're probably considered the leader. Probably 51-60%. [ConcertAI], I'd say they're broad. Maybe you could [say] 11-20%. I think McKesson is probably very low because, I think for them, it's more about the potential to do this as opposed to actually doing it. I'd [say] less than 5% today. [Aetion], I think they might be higher. I could see them being 21-30%. Because what they do is they use existing clinical trial data, which is, again, very structured and organised and they use it for this purpose. There's been some success with that model as well.

What were N-Power, ConcertAI and Aetion's revenues for the last financial year in USD 100m ranges? Flatiron is a subsidiary of a public company and McKesson is a public company. Could you start with N-Power?

Expert: They're way below. I think it's USD 0-50m. I don't think the revenue is very high. I mean, it's not USD 0, but yes, we can [say] USD 0-50m is probably appropriate. Flatiron is owned by Roche. I don't know if you can find it even in the public records. I have no idea. The thing with Flatiron is it's a lot of these models are multifaceted. It's very hard to quantify. Flatiron, they sell data. They're providing EMRs to health systems. They're doing a lot. I mean, this is high level of big error bars on this.

ConcertAI, I was trying to actually research that a little bit because they're not public. I think they're probably USD 50m-100m. I think it's probably in that range. Again, there was something that I found that they were USD 200m-300m in revenue at one point. I would say, we could [mention] them higher, though. Let's say USD 200m-500m.

Analyst: What might be Aetion's revenue?

Expert: I don't know how much scale they have. That's the only thing that I worry about there. Why don't we say USD 100m-200m?

What might be N-Power, ConcertAI and Aetion's revenue growth rates in percentage ranges?

Expert: I think they have to get to over 20% [for N-Power]. [ConcertAI], I think they might be more like 11-20%. I think they have a higher level of growth. They could be over 20% as well. [For Aetion], I have no idea, but I will say it's 11-20%.

What has been each player's growth driver for the last financial year? Has N-Power's growth been organic, acquisition-based or mixed?

Expert: I would say organic. [Flatiron has been] acquisition-based. [ConcertAI is] probably mixed because they have acquired a bunch, but I think now they need to. McKesson would probably be mixed. Aetion is probably more organic.

What would you say has the industry average EBITDA margin been in a percentage range?

Expert: I think it's 21-30%.

Would you say each player has been performing above, on or below industry, starting with N-Power?

Expert: Below. [Flatiron], on industry. [ConcertAI], maybe above industry. [McKesson], probably below industry. [Aetion], I think probably on industry.

How do you think each player differentiates itself from others, starting with N-Power?

Expert: I think it's quality. They're trying to go premium. They're going more premium. They're definitely positioning for quality. Maybe a secondary. I don't know if the tech is better. I would say it's quality.

Analyst: How are others performing?

Expert: I think Flatiron technology. ConcertAI, I would say maybe breadth because they have an existing registry. I think with McKesson, it's more the integration because of its synergies with their existing business.

Analyst: How might Aetion perform?

Expert: I think probably quality for them because they're using clinical trial data.

Analyst: How might the quality of other competitors in the market compare to Aetion and N-Power's?

Expert: I think that's a shortcoming for the others because, again, they're more, the other three are more real-world data companies and the quality of the data is questionable to beat this regulatory grade that's going to be required. I think that's been the issue. I mean, Flatiron really has improved the quality of real data significantly. Again, it's hit a threshold based on their model. They are trying to evolve their model to get to a higher quality option.

Analyst: Why would you say it is different for Flatiron in the first place?

Expert: It's just that they can't get around having missing data. It's missing data, non-perspective. It's always going to be what's available. For example, having time points for certain interventions. You might need imaging every six months. If you're just getting the data from routine care, patients come at different times. You might have one patient at three months, another patient at nine months, another patient at two years. There's no consistency in the data because it's taking what's already there. Where N-Power has a registry where they're systematically collecting the data. Aetion has clinical trial data. Most clinical trials, there are certain standards that they can use from existing clinical trial data. The challenge is that they only have a limited resource. Aetion only has the data that they have from the existing trials. They don't have it from other sources.

What might be customers' top three KPC [key purchasing criteria] when purchasing from N-Power? How does each rank by importance?

Expert: I think something around dependability or the ability to deliver, like certainty or dependability, because, again, if you have a clinical trial, you have a launch timeframe and everything else. I think that's going to be really important. I think viability as well. N-Power is still very much of a start-up, so are they going to be around in five years? If you're investing in one of your main drugs, you're going to want to make sure that the company is going to be there. You're not going to want to invest if they're going to go out of business. Those two are so important. Again, it just goes back to the proof-of-concept that you have demonstrated proof-of-concept.

Analyst: Might proof-of-concept be based on the quality of data?

Expert: Going back to the regulatory approval.

How important would you think price is to customers in the real-world regulatory-grade data service market? Is it commoditised, somewhat commoditised or not commoditised, and why?

Expert: It's not commoditised, I think, because, one, there's not that many players who are delivering data at this level. But also, the value is so high. If you could accelerate one of your drugs by six months, it's incredible value in the market in terms of having your drug on market earlier, first to market or beating a competitor. Again, the value side is so high that that's why it's not commoditised, if that makes sense.

What do you think has been the key driver of N-Power's historical growth? How fast do you think the company could grow over the next three years?

Expert: Scale of the network. Because again, if you only have a couple centres, it's very hard to deliver the data. I think once you have more scale, that's going to be the biggest driver. I think the market could grow very fast because I think they have been at it for almost five years now. Again, I think they're probably getting, I assume they're getting close to having a really clear proof-of-concept. In the next three years, I think it could grow by over 20%.

In your opinion, what could be the most interesting ways to grow N-Power organically?

Expert: I think new product innovation is really key because again, it's good that they're focused on this perspective. Again, there's so much need for quality clinical data. I think there's a lot of innovative products to do that to supplement the current things. Even they work with health systems today and they provide a lot of value to health systems. To the best of my knowledge, I don't think they're charging those health systems. Actually, they're providing a lot of value to the health systems as well. You could almost have a two-sided platform and so product innovation is really the key.

Analyst: Could you elaborate on the product?

Expert: Again, services, if they make, if they can support health systems to be more efficient, operate better decision-making value and then other uses of the data, like training AI models coming up. You could also develop synthetic control arms. External [control arm]. That's another option because if you have this high-quality data, you could probably develop synthetic arms as well.

What could be some interesting ways to improve N-Power's margins?

Expert: I think using the data for multiple use cases. Again, collecting this data for one patient, let's say you're using it for an external control arm. If you could use that same data for other customers, for other use cases. That's going to drive the margin significantly.

Can N-Power make any acquisitions?

Expert: To the point I mentioned about global, like some type of partnerships globally. Maybe we talked about CRO, like buying a CRO. I think the likelihood is probably that they get acquired, like a proof-of-concept that they get acquired by a CRO. I'd say buying a small CRO, so it's not one of the bigger player. I would say a small regional or ex-US CRO.

Based on N-Power's publicly announced acquisitions, how would you describe the company's M&A strategy? Has it been more opportunistic or growth-oriented?

Expert: Probably growth. Again, it's all related to funding and available resources. I think it is growth-related. Because again, the more you expand your network, the more value is of the overall opportunity. Again, I think they're more likely to get acquired than to acquire someone.

In your opinion, could it make sense for N-Power to acquire, merge with or be acquired by any players?

Expert: I think a CRO because then it would add a whole another business to their portfolio because then you could run the clinical trial and the external arm at the same time would be highly valuable to a pharma company. Some of the competitors we did, like, Flatiron potentially could do it. I don't think pharma. I mean, I know Roche is, but I don't think a pharma would because, again, the value is, again, when it could be used across different companies. One company wouldn't have a portfolio big enough to support it. I think it's a data company or a real-world data company, maybe CRO or a real-world data company, maybe.

Analyst: What might be the size of a CRO or real-world data company that could acquire?

Expert: I think it's going to have to be a large, again, multi, probably over USD 1bn market cap company who has access capital.

Analyst: What might be the particular reasoning behind acquiring? Is it just the synergy that N-Power brings? What might be the value of a CRO or a real-world data company?

Expert: I think it's a new product. It's a new product line. It's a new opportunity for them to develop something. Again, N-Power's building this asset, which could be leveraged much stronger by a CRO or another company. Again, it gives them a whole product line that they could scale. They could scale much easier than an N-Power can scale. If you had one of the top CROs, they could easily turn into a global business and they could really leverage the value from it.

What would you say are 1-2 biggest risks faced by N-Power?

Expert: Again, I think the economics of the model and scale.

Analyst: Might economics mean that the cost will be way higher than the potential revenue?

Expert: The potential revenue. Yes.

Analyst: Could you elaborate on scale?

Expert: Not scaling and diversifying the network again. Because again, if you're just US, you can only do support those use cases. You need to be global, the ability to do other diseases. Scale and optionality diversity.

In your opinion, is there anything that the market does not properly understand about N-Power?

Expert: Yes. I don't think a lot of potential customers really differentiate the quality of the data vs what's existing. The gap in quality of data between what they could get, let's say, from a Flatiron or what they can get directly. I think the quality gap is really the biggest. I think this two-sided platform model is the other thing. The value that the health systems get from the model is very high. I think that's the stickiness of it. Because there's been a lot of these speciality community oncology centres and health systems, they have all these companies coming at them to take their data or buy their data or use them or leverage them. There are a lot of risks with that. N-Power is actually providing true value to these centres by helping them organise their data, be more efficient. Many physicians are getting burnt out these days because of the administrative burden of everyday practice. Again, I think they improve the quality for the centres, which improves it. There is a spin-wheel that's really driving value. I think that, having that customer buy-in from the physician side, again, I think that a lot of the value is that the hospitals are getting value from this. It's something that's really unique because, again, you might be able to tap into something very unique because you're helping the hospitals be more efficient.