By Professor David Adler
The quality of a major investment decision often depends on which uncertainties the organization chose to resolve before leadership is asked to commit.
Companies invest heavily in making major investment decisions rigorous. They build financial models, governance processes, investment committees, risk assessments, and sophisticated analytics. Yet by the time those systems are activated, another set of choices has already shaped what leadership will see. Teams have decided which studies to run, which assumptions to test, which markets to investigate, and which technical questions to pursue. Those choices determine which uncertainties are reduced and which remain unresolved. The management challenge is therefore not only how to evaluate evidence well. It is how to decide early enough which evidence is worth generating.
The Harder Question Is What Evidence Matters
In complex organizations, evidence is rarely generated from a single strategic perspective. Scientists conduct experiments. Technical teams perform feasibility work. Commercial teams investigate markets. Finance refines assumptions.
But work can be valuable within a function without being especially valuable to the investment decision.
A team may spend months reducing an uncertainty because it is accessible, familiar, scientifically interesting, or part of an established process. At the same time, another unanswered question may remain unresolved even though its answer could determine whether the organization invests, stops, accelerates, or changes direction.
This is the central management problem. An organization can generate more knowledge and high quality evidence while still being poorly prepared for the decision that matters.
The distinction is not between good evidence and bad evidence. It is between evidence that improves understanding and evidence that can materially alter a choice.
A large body of information may still leave a central assumption insufficiently tested. Conversely, one focused piece of evidence may have far greater strategic value if it addresses an uncertainty on which the investment thesis depends.
Work Backward From the Choice Ahead
A more disciplined approach begins with the future decision rather than the next activity.
What will leadership ultimately need to decide? What would need to be true for the organization to proceed? What could make the opportunity materially less attractive? And what does the organization still not know that could change the answer?
Instead of beginning with the study, analysis, experiment, or market assessment that seems most obvious, leaders first identify the factors that will determine the choice. They then isolate the unresolved questions within those factors and decide which ones deserve attention now.
The value of an evidence generating activity therefore does not depend only on its scientific quality, technical sophistication, or cost. It also depends on whether different results could lead the organization to make different choices.
This does not mean every activity must directly determine an investment decision. Organizations still need exploratory research, foundational work, and cumulative learning. But when resources are limited and the decision is consequential, leaders should be explicit about which uncertainties they are trying to resolve and why those uncertainties matter.
The objective is not to generate the maximum amount of evidence. It is to resolve the uncertainties that matter most before major resources are committed.
When Scientific Progress and Investment Progress Diverge
Pharmaceutical development makes this problem unusually visible because timelines are long, capital requirements are substantial, and uncertainty persists across many stages.
Consider an oncology program with compelling biological activity. A development team may be able to generate another six months of sophisticated laboratory data. The science may be interesting, the experiments feasible, and the results may deepen confidence in the underlying biology.
But that does not mean the additional work addresses what the next investment decision will require.
Is the therapeutic hypothesis sufficiently differentiated? Is there a patient population in which meaningful clinical benefit could realistically be demonstrated? Can the relevant biology be measured reliably in patients? Is the clinical development path feasible? Does the emerging profile justify the resources required to reach the next major value creating milestone?
If those questions will determine whether the program advances, they should influence what evidence is generated now.
A program can therefore make genuine scientific progress while making much less progress against the uncertainty that governs continued investment.
Recognizing that distinction can redirect priorities toward a different experiment, analysis, biomarker strategy, clinical study, or development path. Sometimes the most valuable result will support continued investment. In other cases, it will enable an earlier decision to stop.
Turning Experience Into a Shared Discipline
Experienced leaders often make these distinctions instinctively. They recognize which assumptions are fragile, which questions are likely to become important later, and which uncertainties deserve attention now. Their judgment reflects patterns accumulated across previous programs and decisions.
When evidence priorities depend primarily on what experienced individuals happen to recognize, the quality of the process can vary across teams, programs, and leadership changes. One group may identify a consequential uncertainty early. Another may spend months producing excellent work that has little influence on the eventual choice.
Making the reasoning explicit does not mean replacing expert judgment with a rigid formula. It means making the logic behind that judgment visible enough to examine, challenge, and transfer.
What decision are we preparing to make?
What factors will determine that decision?
Which unresolved questions could materially change the outcome?
Which activities are most likely to resolve those questions before significant resources are committed?
Their purpose is not to add bureaucracy. It is to connect evidence generation more directly to strategic choice.
The aim is to make the reason for doing a piece of work as clear as the work itself.
What Senior Leaders Need to Make Explicit
The principle extends far beyond pharmaceutical research and development. Whenever organizations commit resources under uncertainty, they face two management challenges. They must decide what to do, and they must decide what they need to know before deciding what to do.
Start with the future choice. Strategic prioritization should begin with the decision the work is intended to inform. Activities gain strategic value when their purpose is connected to a consequential choice.
Separate information from decision relevant evidence. More information can create the appearance of progress. But an additional study, analysis, experiment, or market assessment has limited strategic value if its result is unlikely to affect what the organization does.
Make the logic of evidence prioritization visible. Teams should be able to explain which uncertainty an activity addresses, why that uncertainty matters, and how different results could alter the path forward.
When the decision is important, leadership should be able to see not only what evidence is being generated, but why that evidence deserves priority over the alternatives.
The Work Begins Before the Commitment
Organizations have invested heavily in improving how major decisions are evaluated. The larger opportunity may lie earlier, in becoming more deliberate about which uncertainties deserve to be resolved before those decisions arrive.
A strong governance process can assess only the evidence placed before it. If the wrong questions received attention upstream, sophistication at the decision point cannot fully compensate. The quality of a major investment decision therefore begins with a prior act of judgment: choosing what the organization needs to learn before it commits.
Acknowledgement
ChatGPT was used to assist with minor editing of the manuscript. The author reviewed and approved the final text and takes responsibility for its content.
