"AI in Market Access: Do We Have What It Takes?"
An expert interview with Menasheh Fogel, Fractional CIO and former IT executive in the multinational pharmaceutical industry
Barbara Walter (Moderator): Market access sits at the heart of every drug's commercial success - yet the data exists in silos, and the automation hasn't followed. Menasheh Fogel has spent over two decades building IT infrastructure in life sciences, most recently leading IT teams dedicated to market access operations in multinational pharma. He has seen this friction from the inside. We sat down with him to understand what is actually getting in the way - and what it takes to change it.
Menasheh, you've driven digital transformation in market access firsthand. What's actually stopping AI adoption in Market Access - is it the technology, the data, or something else entirely?
Menasheh Fogel (MF): Honestly, it's neither the technology nor the data - at least not primarily. What I typically see is that market access tends to rely instead on experience, negotiation, and network when it comes to getting products on payer registries, rather than on automation.
Moderator: So less a technology problem - more a cultural and organisational one?
MF: Organisational structure plays into this too. Market access needs to be seen as a critical enabler from the beginning - partnering with R&D colleagues from day one. But in reality, R&D teams are orders of magnitude larger in resources, budgets, and organisational positioning. Getting the market access voice heard is a real challenge, especially when development teams are focused on compressing clinical timelines. Yet even if we meet our clinical targets, we need to ask early on: will this compound actually have a commercial market, at the price we need?
As a side note on AI adoption - dossier preparation is very text heavy, so I suspect that LLMs are already being used for this, though possibly not in a streamlined way with rich data context from other functions.
Moderator: That observation resonates with what we see in our own work - AI tools deployed in isolation consistently underdeliver. The moment you connect them horizontally to data from adjacent functions, the picture changes entirely. Which raises the question of how deep that dependency actually runs in market access.
MF: Market access is highly dependent on adjacent functions - especially drug development. The clinical study data and endpoint attainment feed directly into the pricing case. Outcome-based pricing requires a great deal of Real-World Evidence to demonstrate results. But here you run into a different set of challenges: how do you manage data collection, whether the payer will 'trust' the manufacturer to calculate outcomes, and whether the manufacturer should even be allowed to see individual patient results?
Moderator: So what actually works in practice?
MF: The most successful outcome-based schemes are those tied to concrete, easy-to-prove results - patient re-hospitalisation within a specific time window, for example. In those cases, the payer contacts the manufacturer for a rebate. Another critical data collaboration point is with Medical Affairs, who define the patient population and intended use. That information in turn builds the value case.
Moderator: So the data is there - spread across R&D, Medical Affairs, commercial - but rarely connected in a way that AI can actually work with. Given that reality, where does a market access leader even begin?
MF: Dossier drafting is the most straightforward entry point, and I wouldn't be surprised if many market access professionals are already doing this informally. From there, I'd look at three areas. First, an Health Technology Assessment (HTA) listening agent that feeds into launch sequencing - that was actually one of the first use cases I thought of when preparing for today’s conversation. Second, I would consider setting up an agent for monitoring tenders. By increasing tender participation you will very likely increase the award rate. Third, quantifying health outcomes - a key input for pricing discussions. This is a heavy mathematical discipline, with enough human and experience factors that AI ought to be able to capture.
Moderator: Could you give an example for that?
MF: UK NICE has a well-defined approach for calculating QALYs - Quality Adjusted Life Years - a mathematical formula for determining how a new compound affects a patient's overall life improvement and prognosis. So all of these should be a fairly good place to start.
Moderator: How do you assess the business impact of these use cases - where is the real RoAI?
MF: Any incremental gain in market access quality has a direct and significant impact on the top line. What we are talking about is no less than the price a product can command, its marketability, and whether it gains access to payer registries at all.
Dossier writing decreases turnaround time - shorter time to market. Launch sequencing with reference pricing optimises market entry and avoids premature price erosion. Tender monitoring increases participation and drives sales. The QALY calculator accelerates outcome quantification, improving positioning and pricing. Outcome-based pricing and Pay-for-Performance may be required especially in high-price therapies such as cell and gene therapies - getting this right means avoiding price erosion.
Moderator: And the use case with the highest ROI potential?
MF: The use case with by far the highest ROI potential is endpoint and target valuation during clinical studies. The impact won't be felt for potentially years to come - but imagine an AI that looks at the compound candidate, assesses unmet need, and models its ultimate impact on the whole health system. Payers will agree to higher prices if you can demonstrate not only that the compound improves an individual patient's life, but also that it avoids other, costlier interventions downstream. Getting that right means entering the market with the right price from day one.
Moderator: Stepping back from the tactical point of view - what needs to change at leadership level for this to move from isolated pilots to an integrated platform?
MF: Like most changes, the biggest hurdle is culture. We need to ask ourselves whether we are ready to embed commercial valuation early in the drug development phase - or even at discovery, as an aspiration. In my experience, market access colleagues already understand the importance of this. The rest of the organisation needs to come along.
We also need to be open to new ways of working - and as technologists, we need to be humble in our approach to these problems. Finally, concrete success stories from early adopters, with clear pathways showing how these tools create value, will go a long way toward driving change.
Moderator: An approach we very much believe in and actively support. If Pharma gets this right, could AI actually enable value-based reimbursement at scale - something the industry has talked about for years but rarely delivered on?
MF: I do believe AI can better enable this, especially where outcomes are complex to measure. But beyond measurement complexity, there are harder questions: how is data obtained, how is patient and HCP privacy ensured, who is responsible for outcome determination, and what consent is required? These are complicated questions in a multi-stakeholder environment that I don't think can be solved just by throwing AI at them.
Moderator: And what could a realistic 12-month roadmap look like, and where is the hidden complexity most people don't see?
MF: The use cases I identified are mostly arranged in order of how you could construct a roadmap for AI in market access. I would tend to start with low data complexity cases like the dossier assistant, tender monitoring, and potentially the QALY calculator.
The launch sequencing use case has a hidden mathematical optimisation complexity beyond just the HTA monitoring. Imagine a large matrix of countries describing who refers to whom, and under what conditions and indications - which would continuously need to be updated based on new information. That is non-trivial.
Pay-for-Performance is by far the most complex use case, due to the stakeholder and data landscape. It's also not clear which stakeholders really have the stomach to dive into this, especially in complex cases.
The early candidate valuation use case may only be technically complex in terms of integrating internal and public historical data. With the right organisational buy-in, I suspect you might be able to automate and accelerate the valuation of endpoints and candidates with AI tools easier than might be expected - providing high-value strategic and long-term benefit.
Moderator: The complexity is real, but the path is clearer than many assume. Those who move first will not just close gaps. They will redefine what market access can achieve, it is time to roll up our sleeves. Menasheh, thank you for your time and your refreshingly honest take on where the industry stands.
If you'd like to learn more about customized AI strategies for market access, we look forward to discussing this with you.
If you have any further questions, please feel free to contact Bertram Weiss.
LinkedIn: https://www.linkedin.com/in/drbertramweiss/
Email: bertram.weiss@merantix-momentum.com
About Menasheh Fogel
.jpg)
Menasheh Fogel is a Fractional CIO for high-growth biotechs and mid-size companies, as well as a documentary filmmaker. With over 20 years of IT leadership experience in the Life Sciences he has breadth across verticals and functions, coupled with depth along the entire digital value chain from strategy, architecture to operations and infrastructure. In his most recent executive roles, he led IT teams dedicated to market access and CGTs in multinational pharma.
Subscribe to the Merantix Momentum Newsletter now.
More articles
From bottleneck to breakthrough: Why AI is crucial for progress in healthcare and pharma
.png)
.png)
.png)
"AI in Market Access: Do We Have What It Takes?"
An expert interview with Menasheh Fogel, Fractional CIO and former IT executive in the multinational pharmaceutical industry
Barbara Walter (Moderator): Market access sits at the heart of every drug's commercial success - yet the data exists in silos, and the automation hasn't followed. Menasheh Fogel has spent over two decades building IT infrastructure in life sciences, most recently leading IT teams dedicated to market access operations in multinational pharma. He has seen this friction from the inside. We sat down with him to understand what is actually getting in the way - and what it takes to change it.
Menasheh, you've driven digital transformation in market access firsthand. What's actually stopping AI adoption in Market Access - is it the technology, the data, or something else entirely?
Menasheh Fogel (MF): Honestly, it's neither the technology nor the data - at least not primarily. What I typically see is that market access tends to rely instead on experience, negotiation, and network when it comes to getting products on payer registries, rather than on automation.
Moderator: So less a technology problem - more a cultural and organisational one?
MF: Organisational structure plays into this too. Market access needs to be seen as a critical enabler from the beginning - partnering with R&D colleagues from day one. But in reality, R&D teams are orders of magnitude larger in resources, budgets, and organisational positioning. Getting the market access voice heard is a real challenge, especially when development teams are focused on compressing clinical timelines. Yet even if we meet our clinical targets, we need to ask early on: will this compound actually have a commercial market, at the price we need?
As a side note on AI adoption - dossier preparation is very text heavy, so I suspect that LLMs are already being used for this, though possibly not in a streamlined way with rich data context from other functions.
Moderator: That observation resonates with what we see in our own work - AI tools deployed in isolation consistently underdeliver. The moment you connect them horizontally to data from adjacent functions, the picture changes entirely. Which raises the question of how deep that dependency actually runs in market access.
MF: Market access is highly dependent on adjacent functions - especially drug development. The clinical study data and endpoint attainment feed directly into the pricing case. Outcome-based pricing requires a great deal of Real-World Evidence to demonstrate results. But here you run into a different set of challenges: how do you manage data collection, whether the payer will 'trust' the manufacturer to calculate outcomes, and whether the manufacturer should even be allowed to see individual patient results?
Moderator: So what actually works in practice?
MF: The most successful outcome-based schemes are those tied to concrete, easy-to-prove results - patient re-hospitalisation within a specific time window, for example. In those cases, the payer contacts the manufacturer for a rebate. Another critical data collaboration point is with Medical Affairs, who define the patient population and intended use. That information in turn builds the value case.
Moderator: So the data is there - spread across R&D, Medical Affairs, commercial - but rarely connected in a way that AI can actually work with. Given that reality, where does a market access leader even begin?
MF: Dossier drafting is the most straightforward entry point, and I wouldn't be surprised if many market access professionals are already doing this informally. From there, I'd look at three areas. First, an Health Technology Assessment (HTA) listening agent that feeds into launch sequencing - that was actually one of the first use cases I thought of when preparing for today’s conversation. Second, I would consider setting up an agent for monitoring tenders. By increasing tender participation you will very likely increase the award rate. Third, quantifying health outcomes - a key input for pricing discussions. This is a heavy mathematical discipline, with enough human and experience factors that AI ought to be able to capture.
Moderator: Could you give an example for that?
MF: UK NICE has a well-defined approach for calculating QALYs - Quality Adjusted Life Years - a mathematical formula for determining how a new compound affects a patient's overall life improvement and prognosis. So all of these should be a fairly good place to start.
Moderator: How do you assess the business impact of these use cases - where is the real RoAI?
MF: Any incremental gain in market access quality has a direct and significant impact on the top line. What we are talking about is no less than the price a product can command, its marketability, and whether it gains access to payer registries at all.
Dossier writing decreases turnaround time - shorter time to market. Launch sequencing with reference pricing optimises market entry and avoids premature price erosion. Tender monitoring increases participation and drives sales. The QALY calculator accelerates outcome quantification, improving positioning and pricing. Outcome-based pricing and Pay-for-Performance may be required especially in high-price therapies such as cell and gene therapies - getting this right means avoiding price erosion.
Moderator: And the use case with the highest ROI potential?
MF: The use case with by far the highest ROI potential is endpoint and target valuation during clinical studies. The impact won't be felt for potentially years to come - but imagine an AI that looks at the compound candidate, assesses unmet need, and models its ultimate impact on the whole health system. Payers will agree to higher prices if you can demonstrate not only that the compound improves an individual patient's life, but also that it avoids other, costlier interventions downstream. Getting that right means entering the market with the right price from day one.
Moderator: Stepping back from the tactical point of view - what needs to change at leadership level for this to move from isolated pilots to an integrated platform?
MF: Like most changes, the biggest hurdle is culture. We need to ask ourselves whether we are ready to embed commercial valuation early in the drug development phase - or even at discovery, as an aspiration. In my experience, market access colleagues already understand the importance of this. The rest of the organisation needs to come along.
We also need to be open to new ways of working - and as technologists, we need to be humble in our approach to these problems. Finally, concrete success stories from early adopters, with clear pathways showing how these tools create value, will go a long way toward driving change.
Moderator: An approach we very much believe in and actively support. If Pharma gets this right, could AI actually enable value-based reimbursement at scale - something the industry has talked about for years but rarely delivered on?
MF: I do believe AI can better enable this, especially where outcomes are complex to measure. But beyond measurement complexity, there are harder questions: how is data obtained, how is patient and HCP privacy ensured, who is responsible for outcome determination, and what consent is required? These are complicated questions in a multi-stakeholder environment that I don't think can be solved just by throwing AI at them.
Moderator: And what could a realistic 12-month roadmap look like, and where is the hidden complexity most people don't see?
MF: The use cases I identified are mostly arranged in order of how you could construct a roadmap for AI in market access. I would tend to start with low data complexity cases like the dossier assistant, tender monitoring, and potentially the QALY calculator.
The launch sequencing use case has a hidden mathematical optimisation complexity beyond just the HTA monitoring. Imagine a large matrix of countries describing who refers to whom, and under what conditions and indications - which would continuously need to be updated based on new information. That is non-trivial.
Pay-for-Performance is by far the most complex use case, due to the stakeholder and data landscape. It's also not clear which stakeholders really have the stomach to dive into this, especially in complex cases.
The early candidate valuation use case may only be technically complex in terms of integrating internal and public historical data. With the right organisational buy-in, I suspect you might be able to automate and accelerate the valuation of endpoints and candidates with AI tools easier than might be expected - providing high-value strategic and long-term benefit.
Moderator: The complexity is real, but the path is clearer than many assume. Those who move first will not just close gaps. They will redefine what market access can achieve, it is time to roll up our sleeves. Menasheh, thank you for your time and your refreshingly honest take on where the industry stands.
If you'd like to learn more about customized AI strategies for market access, we look forward to discussing this with you.
If you have any further questions, please feel free to contact Bertram Weiss.
LinkedIn: https://www.linkedin.com/in/drbertramweiss/
Email: bertram.weiss@merantix-momentum.com
About Menasheh Fogel
Menasheh Fogel is a Fractional CIO for high-growth biotechs and mid-size companies, as well as a documentary filmmaker. With over 20 years of IT leadership experience in the Life Sciences he has breadth across verticals and functions, coupled with depth along the entire digital value chain from strategy, architecture to operations and infrastructure. In his most recent executive roles, he led IT teams dedicated to market access and CGTs in multinational pharma.
.png)
.png)