In drug discovery, great science alone is not enough because commercial viability ultimately decides which programmes survive and attract partners. This Q&A explores how integrating Business Development and Licensing (BD&L) from the earliest stages can guide R&D strategy, sharpen decisions and de-risk the path to market.
While early R&D scientists focus on the core science, such as efficacy, mechanism and toxicity, they operate within a business ecosystem where business development and licensing (BD&L) decisions fundamentally dictate whether their project receives continued funding (go/no-go), finds a development partner (licensing out), or is replaced by a better acquired asset (licensing in). Crucially, scientists must consider not only the ‘tractability’ (whether a drug can hit the target) but also ‘commercial viability’ (whether the target addresses a significant unmet need or a large patient population). BD&L heavily influences this second aspect by providing essential market intelligence.
In this Q&A, I ask Carlos Velez, a BD&L veteran, about the common themes in this strategic game. Carlos distilled 25 years of BD&L experience into a single book covering the full industry spectrum from licensing strategies and deal structure to negotiation tactics, post-deal execution and the emerging role of AI. He recently added a free chapter on out-licensing strategy, which includes how to assess whether a candidate is truly licensable and a practical rubric for out-licensing decisions. Furthermore, Carlos also develops and presents customised training programmes for companies seeking to improve their in- and out-licensing processes.
Remco: Given that BD&L is often viewed as a transactional function relevant only at key drug development inflection points, why should resource-constrained early-stage drug discovery companies integrate the assessment of a candidate’s licensing viability into their initial R&D strategy?
Carlos: BD&L is more than just ‘doing deals.’ The out-licensing process should start years in advance, when the company’s initial drug development strategic choices are being made. As part of this decision making, understanding what is likely to transact via a license at some point in the future should, in part, determine which programmes are funded and which are not. Thus, early understanding of licensing trends, as well as engagement with prospective licensees, may help early-stage companies avoid spending resources on candidates that prospective licensees will not value.
BD&L is more than just ‘doing deals.’ The out-licensing process should start years in advance, when the company’s initial drug development strategic choices are being made.Â
It is also an important exercise to help assess competition, monitor treatment evolution and identify true therapeutic ‘white spaces’ that represent actual unmet medical needs. Even at target-ID or hit-discovery stages, credible BD&L conversations can guide decisions on candidate differentiation, assay design, or model selection and utilisation. This alignment between scientific direction and commercial reality can meaningfully increase the probability that a future licensing transaction will occur. Thus, the most successful licensing outcomes originate from decisions made long before a data package is assembled.
Remco: In the event that a discovery company has no animal data or lead candidates, existing only as a technology platform, how is BD&L still relevant? Specifically, how can BD&L secure external validation and risk-sharing mechanisms through early collaborations to support the platform’s valuation?
Carlos: Even at the platform stage, talking to prospective licensing partners could lead to various collaborations, such as research agreements. This is especially the case with higher risk programmes, novel modalities or undruggable targets. While there may not be a specific candidate to license at this stage, low-risk collaborations could lead to target-selection partnerships, equity investments, option-based deals or data-generation collaborations.
A discovery platform becomes more valuable when external partners validate it, making it more favourable when securing that next round of venture financing.
In addition, a discovery platform becomes more valuable when external partners validate it, making it more favourable when securing that next round of venture financing. For investors, seeing evidence that industry players view the platform as credible and differentiated can materially influence valuation and financing terms.
Thus, early BD&L is a risk-sharing mechanism, allowing discovery-stage companies to potentially extend runways without surrendering full ownership of a platform or a candidate. In today’s capital-constrained environment, this external validation can make the difference between advancing a promising technology and shutting it down prematurely.
Remco: For a licensing discussion to progress to a deeper level of due diligence, what role would BD&L play in shaping IP and devising strategies to extract meaningful decision-making information?
Carlos: Early BD&L discussions may reveal what types of data, assays or IP positions prospective licensees consider essential. For example, a company may learn that additional in vivo validation, alternative animal models, biomarker development or specific formulation work will materially increase partnering interest. Similarly, early partner feedback may influence patent strategy, such as prioritising composition-of-matter claims, developing broader platform claims or securing freedom-to-operate. Thus, aligning data generation and IP strategy with anticipated partner expectations can accelerate future deals and reduce the risk of costly rework.
Data should be organised and presented in a logical, easy-to-find structure to help the process move quickly from initial diligence to deeper diligence.Â
However, companies should carefully consider whether executing on a suggestion from a prospective licensee is the right path forwards, as others within that licensee’s organisation may have different opinions on what data are needed and which are not. A balanced approach is required. Input from prospective partners can be extremely valuable, but companies must ensure that they remain focused on building assets and IP positions that support their own strategic interests, rather than simply responding to external requests.
Simply dumping research reports into a data room is not the answer. Data should be organised and presented in a logical, easy-to-find structure to help the process move quickly from initial diligence to deeper diligence. As AI-based diligence platforms evolve, data organisation and presentation will become increasingly important.
Remco: Considering the risks of generating legal or contractual issues, how can early-stage companies structure research collaborations, which may involve joint IP creation or shared data access, to avoid complicating future licensing discussions?
Carlos: Early collaborative studies, joint IP creation, or access to proprietary technologies can definitely complicate downstream negotiations with other companies if ownership, rights of use and territorial rights are not clearly defined. Licensees performing diligence will often scrutinise prior collaborations for encumbrances (such as option rights), background IP obligations or hidden financial commitments.
An early partnership may also create expectations around co-development rights, geographic exclusivity or even sublicensing obligations. Therefore, structuring early collaborations with an understanding of how they may influence later licensing efforts is critical. The key is to retain flexibility and control while enabling sufficient collaboration to advance the science. Legal counsel is vital in these situations.
In other words, early decisions can have significant later-stage implications that are difficult to unwind. Discovery companies should therefore resist the temptation to accept unfavourable terms simply to initiate a collaboration, especially if the long-term strategic value of the programme could be compromised.
Remco: Where does AI fit into BD&L for early-stage drug discovery companies? Specifically, how can the AI-generated computational evidence and insights be leveraged during due diligence and valuation discussions to demonstrate a candidate’s increased probability of success (PoS) and accelerate licensing discussions?
Carlos: We are in the earliest days of AI implementation. Nevertheless, we can already see some of the potential benefits. From the discovery company perspective, AI can help with any number of critical drug discovery and development activities, from target and lead identification through clinical trial design. Indeed, we are already seeing some licensors ask to review AI-supporting evidence as part of their diligence.
But large language models can also be used to identify potential partners, conduct competitive intelligence, and inform candidate valuation and negotiations. AI can also be used for supportive work, such as Freedom to Operate analyses and competitive landscaping.
As discussed in our prior articles,1-5 true acceptance of AI as a drug discovery enabler will likely accelerate once more AI-derived candidates progress into Phase II studies and beyond. At that point, AI-generated insights will carry greater weight during diligence and BD&L discussions may proceed more rapidly based on computational evidence.
Carlos: Speaking of AI, I have a question for you. What are the quantifiable, tangible outcomes that both AI and Explainable AI (xAI) deliver in terms of predictive accuracy, resource reallocation and commercial forecasting for drug candidates?
Remco: xAI tackles the ‘black box’ nature of complex models applied to numerical and text-based data by providing clear, trustworthy justifications for portfolio decisions – essential for stakeholders like managers and regulators. This combined AI/xAI approach delivers significant, quantifiable outcomes: it elevates drug trial prediction accuracy, enables the strategic reallocation of resources by identifying weak candidates early, and dynamically updates asset valuations using real-time market data from different sources to ensure a probabilistic, optimised management of the drug portfolio.
Remco: To make themselves commercially compelling, what are the most critical, actionable steps early-stage R&D teams must take to build a compelling data package and strategic narrative that attracts potential BD&L partners? Specifically, how do R&D activities translate scientific findings into the valuation metrics (differentiation, innovation, optionality) that licensing partners require?
Carlos: Licensees are increasingly looking for highly differentiated, innovative candidates with the optionality to be developed across multiple indications and/or multiple formulations. Thus, building a data package and a coherent story around these three aspects (differentiation, innovation, optionality) is critical. But this takes time, planning and the resources to do this.
Plans to advance the candidate(s) should be sketched out, recognising that a potential licensee may have different opinions on the next steps in the development process. Companies that demonstrate flexibility, a clear understanding of disease biology and a realistic appreciation of development risk tend to attract more interest. Even relatively modest steps, such as generating pilot in vivo data, establishing a biomarker plan or securing a strong IP position, can materially improve partnering prospects.
Remco: What specific advice do you offer R&D folk who enroll in your course on how to leverage an understanding of BD&L to transition into strategic roles, secure internal project funding or enhance long-term career marketability?
Carlos: The participants in my course come from many different backgrounds ranging from preclinical and clinical scientists through regulatory affairs and legal, some of whom may not have a scientific background. My advice from a career perspective is to always define and assess the clinical problems they are trying to solve, and to objectively assess and value them. It is easy to think that great science is valuable. But sometimes great science does not actually address a clinical problem. Other times, it may address a problem today, but there may be other candidates in development that are simply better, more innovative and more likely to become standard of care in the future. So understanding how treatment paradigms may evolve over time is critical for BD&L and for securing funding and/or career development.
Conclusion
The probability of success (PoS) serves as the crucial quantitative lever connecting R&D and BD&L within the pharmaceutical lifecycle. While R&D focuses on scientific execution, such as generating data on a drug’s efficacy, mechanism and toxicity, BD&L uses the PoS calculation to make strategic and financial decisions. PoS assessments drive portfolio prioritisation, resource allocation and crucial Go/No-Go decisions, essentially determining which R&D projects receive continued funding or are presented for licensing. This early consideration of commercial viability, informed by BD&L’s market intelligence, ensures R&D focuses on targets that address significant unmet medical needs, validating the scientific work and attracting external interest to maximise value.
By leveraging AI to analyse massive, often siloed, datasets and clinical genomics, companies can gain deeper insights and generate AI-supporting evidence that enhances the perceived quality of a candidate.Â
The increasing integration of AI is directly impacting the PoS calculation and its role in BD&L, enhancing the predictive power of early-stage assessments. AI tools, including multimodal language models, enhance drug development across multiple activities, from target identification through clinical trial design. By leveraging AI to analyse massive, often siloed, datasets and clinical genomics, companies can gain deeper insights and generate AI-supporting evidence that enhances the perceived quality of a candidate. As more AI-derived candidates successfully progress into later stages, these computational insights will carry greater weight during due diligence, potentially accelerating the BD&L valuation process.
In conclusion, BD&L is not a mere transactional function, but a strategic discipline that provides the map and compass for early R&D. For R&D scientists, this means transitioning into strategic thinkers who ensure that a great scientific breakthrough is also a great business opportunity. The key advice for professionals is to consistently define, objectively assess and value the clinical problems their programmes solve. Scientists must look beyond the immediate novelty of the science to determine whether their asset truly addresses an unmet need or if better, more innovative candidates are already in development. Understanding how treatment paradigms evolve over time is critical for securing funding and career development. As the industry continues to integrate AI-driven discovery approaches, these early strategic interactions are becoming even more crucial for informing scientific direction and increasing the likelihood of long-term success.
About the experts
Remco Jan Geukes Foppen, PhD, is an AI and life sciences expert specialising in the pharmaceutical sector and founder of Explainambiguity. With a global perspective, he integrates and implements AI-driven strategies that impact business decisions; always considering the human element. His leadership has driven international commercial success in areas including image analysis, data management, bioinformatics, advanced clinical trial data analysis leveraging machine learning and federated learning. Remco Jan Geukes Foppen’s academic background includes a PhD in biology and a master’s degree in chemistry, both from the University of Amsterdam.
Carlos N Velez, PhD, MBA, is a pharmaceutical and biotechnology strategic advisor, with 25 years’ experience in consulting, venture capital, corporate strategy and entrepreneurship. Carlos specialises in helping pharmaceutical and biotechnology companies develop their in- and out-licensing strategies, with additional expertise and experience in portfolio assessment and prioritisation, drug candidate valuation, valuation and related services. He also develops and presents customised training programmes (both live and virtual) for companies seeking to improve their in- and out-licensing processes. He holds a PhD in pharmacy from the University of North Carolina at Chapel Hill, and an MBA from the Rochester Institute of Technology.
References
Geukes Foppen RJ, Gioia V, Zoccoli A, Velez CN. Early evidence and emerging trends: How AI is shaping drug discovery and clinical development [Internet]. Drug Target Review. 2025. Available from: https://www.drugtargetreview.com/article/158593/early-evidence-and-emerging-trends-how-ai-is-shaping-drug-discovery-and-clinical-development/
Geukes Foppen RJ, Gioia V, Zoccoli A, Velez CN. The rise of multimodal language models in drug development [Internet]. European Pharmaceutical Review. 2025 [cited 2026 Jan 7]. Available from: https://www.europeanpharmaceuticalreview.com/article/256440/the-rise-of-multimodal-language-models-in-drug-development
Geukes Foppen RJ, Gioia V, Zoccoli A, Velez CN. Navigating the AI revolution: a roadmap for pharma’s future [Internet]. Drug Target Review. 2025. Available from: https://www.drugtargetreview.com/article/157270/navigating-the-ai-revolution-a-roadmap-for-pharmas-future/
Geukes Foppen RJ, Gioia V, Zoccoli A, Velez CN. Using clinical genomics and AI in drug development [Internet]. Drug Target Review. 2025. Available from: https://www.drugtargetreview.com/article/155906/clinical-genomics-ai-drug-success/
Geukes Foppen RJ, Gioia V, Zoccoli A, Velez CN. From siloed data to breakthroughs: multimodal AI in drug discovery [Internet]. Drug Target Review. 2025. Available from: https://www.drugtargetreview.com/article/160597/from-siloed-data-to-breakthroughs-multimodal-ai-in-drug-discovery/
Literature