When Does the Cost of Answering a Question Exceed the Value of Knowing the Answer?

Can a Worthwhile Question Still Be Too Expensive to Answer?

Some research questions are clearly valuable but extraordinarily demanding. Answering them well might require thousands of participants, years of follow-up, expensive equipment, access to difficult-to-reach populations, extensive staff time, or substantial burdens on participants and institutions.

At some point, an uncomfortable question arises: is knowing the answer worth what it would take to obtain it?

This is not merely a budgeting question. Research resources are finite. Money spent on one study cannot simultaneously fund another. Participants' time, researchers' effort, institutional capacity, access to rare populations, and years of scientific attention also have alternative uses.

An important question can therefore be scientifically worthwhile in principle while a particular attempt to answer it is not worthwhile at its current cost.

Research Becomes Difficult to Justify When Its Expected Informational Benefit No Longer Warrants Its Full Cost

Research Has an Opportunity Cost Even When the Question Is Important

The Relevant Cost Is Larger Than the Study Budget

When researchers hear “cost,” they may think first about grants, equipment, software, laboratory procedures, travel, or participant incentives. Those expenses matter, but they do not capture the entire cost of producing evidence.

A research project may also consume researcher time, participant effort, institutional access, scarce samples or datasets, administrative capacity, specialist expertise, and years during which alternative research could have been undertaken.

Some studies also expose participants to inconvenience, privacy risks, invasive procedures, or other burdens. Ethical review addresses whether such burdens are acceptable, but their existence remains relevant when considering whether the information sought is sufficiently valuable to warrant generating it.

Opportunity Cost Is Central to Research Prioritization

Opportunity cost is what is forgone when resources are committed to one option rather than another. In research, the comparison is therefore not simply:

“Can we afford this study?”

It is also:

“Is this the best use of these research resources compared with the alternatives?”

A study costing a modest amount can have a high opportunity cost if the same resources could resolve a much more consequential uncertainty. Conversely, an expensive study may be defensible if it addresses an uncertainty with very large consequences and has a strong prospect of producing decision-relevant evidence.

This is why research prioritization cannot be reduced to selecting the cheapest projects.

Value of Information Provides a Formal Version of This Reasoning

Decision theory offers a more formal framework for comparing the benefits of additional evidence with the costs of acquiring it. Value of information analysis estimates the expected benefit of reducing uncertainty so that decisions can be improved. It has been developed particularly extensively in health economics and health-policy decision-making, although the underlying principles are relevant more broadly.

Different value-of-information measures answer different questions. The expected value of perfect information considers the maximum expected benefit if relevant uncertainty could be eliminated. The expected value of sample information concerns the expected benefit from the information that a particular proposed study could realistically provide. The expected net benefit of sampling then compares that expected informational benefit with the cost of conducting the research.

The formal calculations require a defined decision model and assumptions that will not suit every research field. Still, the conceptual lesson is widely useful: the value of knowing something should be compared with what it costs to know it.

An Important Uncertainty Can Have Low Research Value if Better Information Would Not Change Much

Suppose researchers remain uncertain about the precise magnitude of an effect, but every value within the plausible range would lead to the same decision. Additional precision may be scientifically interesting, yet its decision value could be limited.

By contrast, a smaller uncertainty can be highly consequential when plausible values fall on opposite sides of an important decision threshold.

Research value therefore depends not simply on how uncertain we are but on the consequences of that uncertainty. This is why asking how much difference knowing the answer would make should come before asking how much you are willing to spend obtaining it.

The Proposed Study Will Never Deliver Perfect Information

Researchers should also distinguish the value of knowing the true answer from the value of conducting a particular study.

A question can have enormous potential value while a weak study of it has little expected informational value. Spending heavily on a design that is unlikely to shift understanding or decisions is difficult to justify merely because the underlying topic is important.

Consider How Many People or Decisions Could Benefit

The value of information can increase when the resulting knowledge applies repeatedly or affects many people. In formal health-economic applications, population-level value may account for the number of current and future people affected by a decision.

The broader principle applies outside health care too. Evidence influencing a recurring national decision may justify greater research investment than equally costly evidence affecting a one-time, low-consequence choice. Yet population size should not become the only measure of value. Research concerning a small population may be highly consequential when the stakes for those affected are substantial.

Information Can Lose Value Over Time

The useful lifetime of the evidence also matters.

A lengthy study of a rapidly changing technology may produce a rigorous answer to a question that has become obsolete by publication. A long-term investigation of a stable biological, historical, theoretical, or social phenomenon may remain useful for decades.

Research duration is therefore not merely a scheduling inconvenience. Delay can reduce the expected value of information when decisions must be made before results become available or when the object of study changes quickly.

Some Costs Cannot Be Reduced to Money

Formal economic approaches often express costs and benefits in common quantitative units. Many research decisions do not permit such neat conversion.

How should a researcher price participant fatigue, community trust, privacy exposure, the use of a rare archival collection, or the burden imposed on a vulnerable population? Often, one should not pretend that every consideration can be converted cleanly into currency.

These factors can still be incorporated into a structured judgment. The absence of a single numerical unit does not eliminate the trade-off.

Research Can Become Worthwhile Again if the Design Changes

Concluding that a proposed study costs more than its expected information is worth does not necessarily mean abandoning the question.

The study might be redesigned. Researchers may reduce unnecessary measurements, use existing data, focus on the parameters that contribute most to consequential uncertainty, collaborate across institutions, adopt an efficient sampling strategy, or conduct an intermediate study before committing to the full project.

Value-of-information methods can formally help identify which uncertainties deserve additional data collection in appropriate decision models. More generally, the same principle encourages researchers to ask whether every component of an expensive design is actually necessary to produce the information that matters.

Sometimes a smaller answerable question can preserve much of the informational value while requiring substantially fewer resources.

An Expensive Study Is Not Automatically Worthwhile Because the Outcome Matters

Common Mistakes When Comparing Research Cost With Research Value

Compare the Study You Could Conduct With the Information You Actually Expect to Gain

You do not need a formal economic model every time you choose a research question. A structured comparison can still expose weak assumptions about research value.

This judgment becomes particularly important when research budgets are constrained or studies impose substantial burdens. It also prevents “important topic” from becoming a blanket justification for unlimited data collection.

At the same time, do not reduce the decision to a crude financial return. Scientific understanding, methodological development, evidence for underserved populations, and other contributions may be valuable even when their benefits cannot be monetized neatly.

The goal is proportionality: the resources and burdens of the study should make sense in relation to the importance of the uncertainty and the amount of useful information the research is realistically expected to provide.

Before Investing Heavily in Answering a Research Question, Check:

Questions About Whether Research Is Worth Its Cost

The Question Is Not Simply Whether Research Is Expensive, but Whether the Information Is Worth the Investment

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.