03 · What You Need to Know
How to Assess Whether a Research Study Is Really Feasible
Feasibility Is More Than “Can I Afford It?”
Budget matters, but feasibility is broader. A project can be fully funded and still fail because recruitment is unrealistic, data access was assumed rather than confirmed, an instrument cannot measure the intended outcome adequately, or the research team lacks the necessary expertise.
The FINER framework treats feasibility as one criterion for evaluating a research question, alongside whether it is interesting, novel, ethical, and relevant. Feasibility includes practical considerations such as funding, time, institutional support, participant or data availability, personnel, and expertise.
A useful feasibility assessment therefore examines the complete chain between the research question and the final analysis. If any essential link is unrealistic, the project needs modification.
Start With the Evidence Your Question Requires
Before asking whether the study is feasible, make sure you know what evidence would actually answer the research question.
What population must be represented? What variables, outcomes, experiences, records, specimens, documents, observations, or other evidence are required? Does the question require a comparison group? Repeated measurements? Long-term follow-up? A particular experimental condition? Specialized equipment?
This matters because feasibility belongs to a specific research design. “I can survey 100 students” tells you very little unless a survey of those students can produce evidence capable of answering the question.
Watch Out
Do not redesign the research question around whatever data happen to be easiest to collect unless those data can genuinely address the research problem. Convenience can improve feasibility, but it cannot substitute for choosing a research approach that is methodologically aligned with the question.
Check Whether the Study Design Is Realistic
Once the question is clear, identify the design capable of answering it and ask whether that design can actually operate in your setting.
A protocol should specify the study population, sampling or recruitment approach, procedures, measurements, expected duration, analysis, and other central methodological elements. These decisions determine many of the practical requirements that follow.
For example, a longitudinal study creates follow-up and retention demands that a cross-sectional study does not. An intervention study may require training, intervention delivery, monitoring, comparison conditions, and adherence assessment. Laboratory research may depend on equipment availability, consumables, storage, technical expertise, and quality-control procedures.
Do not label a design feasible merely because studies using that design exist elsewhere. Feasibility depends on your question, population, setting, team, infrastructure, and constraints.
Can You Reach the Participants You Need?
For studies involving participants, recruitment is one of the most consequential feasibility questions.
Estimate how many potentially eligible people actually exist within your recruitment pool, how you will reach them, what proportion may be eligible, how many are likely to consent, how quickly recruitment can occur, and how many participants may withdraw or be lost to follow-up.
A hospital may treat thousands of patients, for example, while only a small fraction meet a study's eligibility criteria. A university may have a large student population, but that does not mean researchers automatically have permission to contact those students or that enough will participate.
When recruitment is uncertain, obtain information from the actual setting where possible rather than relying on intuition.
Can You Obtain Enough Participants for the Planned Analysis?
Recruiting some participants is different from recruiting enough appropriate participants to support the intended analysis.
For quantitative research, sample-size requirements depend on the study design, objectives, outcomes, expected variability or event rates, desired precision or statistical power, planned comparisons, and other assumptions. The appropriate calculation should be determined for the particular study rather than by a generic minimum sample size.
For qualitative research, feasibility involves different considerations, including access to information-rich participants, the scope and purpose of the study, the analytic approach, and the depth of data required. A statistical power calculation is not the appropriate test for every research design.
The central feasibility question remains the same: can you realistically obtain enough appropriate evidence to support the analysis you intend to perform?
Confirm Data Access Rather Than Assuming It
Secondary data can make a study appear easy. The data already exist, so the project must be feasible, right?
Not necessarily. You need to establish that the dataset exists in the required form, contains the variables and observations you need, has adequate quality and coverage, and can legally and institutionally be accessed for your proposed use.
Access may require an application, data-use agreement, ethics review, payment, collaboration, institutional approval, secure computing environment, or lengthy processing period. Some datasets permit only certain analyses or forms of disclosure.
If your study depends on a dataset you have not inspected or confirmed access to, treat data availability as an unresolved feasibility risk.
Check Whether Your Measurements Can Produce the Evidence You Need
Access to participants or data does not guarantee useful evidence. You also need an appropriate way to measure the concepts central to the question.
Ask whether suitable instruments, assays, coding procedures, interview methods, sensors, administrative records, or other measurement approaches exist. Consider reliability, validity, burden, licensing, language, cultural appropriateness, equipment requirements, and the expertise needed to administer or interpret them.
If a central construct cannot be measured adequately, the study may need methodological development before the main research question can be answered convincingly.
Build the Timeline From Tasks, Not From the Submission Deadline
A statement such as “I have six months” does not demonstrate that a six-month study is possible.
Work forward through the project. Include protocol development, literature review, permissions, ethics or regulatory review where required, contracts or data agreements, instrument preparation, recruitment, data collection, follow-up, data cleaning, transcription or processing, analysis, interpretation, writing, revision, and contingency time.
Some stages can occur simultaneously, while others cannot begin until an earlier step is completed. A delay in approval can shift recruitment. Slow recruitment can delay analysis. Long follow-up can make an otherwise simple design incompatible with a fixed graduation or funding deadline.
WHO's recommended research-protocol format specifically calls for the duration of each phase and a detailed timeline for project activities, as well as anticipated problems and possible solutions.
Calculate the Full Cost of Conducting the Study
Research costs extend beyond obvious purchases. Depending on the project, expenses may include personnel, participant reimbursement, travel, laboratory tests, equipment, consumables, software, data access, transcription, translation, licensing, secure data storage, specialist services, publication or dissemination, and administrative charges.
Also consider whether resources must be available at particular times. Having access to a laboratory does not help if it is unavailable during your data-collection period.
If the project depends on funding you have not secured, distinguish between “feasible if funded” and “feasible with the resources currently available.”
Audit the Expertise Required at Every Stage
Researchers sometimes ask whether they personally know how to conduct the study. A better question is whether the research team collectively has, or can realistically obtain, the expertise required.
You may need knowledge of a specialized research design, qualitative interviewing, laboratory procedures, statistical modeling, programming, clinical assessment, instrument development, data security, language translation, community engagement, or another technical area.
Identify expertise gaps early. Some can be addressed through training. Others require a collaborator, statistician, methodologist, technician, clinician, librarian, data manager, or other specialist.
Adding a collaborator is not merely putting a name on the protocol. Make sure the person has agreed to participate and has enough time to perform the role on which the project depends.
Check the Infrastructure and Technology
Some projects require more than people and money. Consider physical facilities, laboratory capacity, storage, computing power, secure servers, internet connectivity, recording equipment, specialized software, clinical space, specimen transport, backup systems, or access to particular platforms.
Test critical technology where possible. A data-collection platform that works in your office may fail in a field setting with unreliable connectivity. Software may require licenses your institution does not hold. A computational analysis may exceed available memory or processing capacity.
Feasibility problems are often mundane. They still stop studies.
Investigate Permissions and Institutional Dependencies
A project may depend on people or organizations outside the research team. You may need permission from schools, hospitals, companies, government agencies, community organizations, laboratories, database custodians, archives, platform providers, or other gatekeepers.
Do not interpret informal enthusiasm as formal access. Ask what approvals are actually required, who can grant them, how long the process normally takes, and whether restrictions could affect recruitment, data collection, analysis, or publication.
Where human participants are involved, applicable ethical review and informed-consent requirements must also be considered. WHO emphasizes ethical review for research involving human participants and the protection of participants' dignity, rights, and welfare.
Ethical Feasibility Is Part of Research Feasibility
A method does not become feasible simply because you can technically perform it. It must also be ethically defensible.
Consider risks and burdens to participants, informed consent, privacy and confidentiality, vulnerable populations, data security, recruitment practices, compensation, conflicts of interest, and the scientific value of the proposed research as relevant to your study.
Scientific validity and ethics are connected. Research using inadequate methods can expose participants or consume resources without a reasonable prospect of producing useful knowledge. Ethical questions therefore belong in early feasibility planning, not merely in paperwork completed after the study design has been fixed.
Ask Whether Participants and Other Stakeholders Will Accept the Procedures
A protocol can look excellent on paper and still be impractical for the people expected to participate in or deliver it.
Will participants complete a 90-minute questionnaire? Return repeatedly for follow-up? Wear a device for the required period? Provide the requested specimen? Discuss the proposed topic in an interview? Will clinicians, teachers, field workers, or other staff have time to implement the study procedures?
Acceptability is a recognized component of feasibility research. If participation or implementation creates excessive burden, recruitment, adherence, retention, or data quality may suffer.
Map the Study's Critical Dependencies
Some requirements are helpful; others are non-negotiable. Identify the things without which the study cannot proceed.
| Dependency |
Question to Verify |
Potential Response |
| Participants |
Can enough eligible participants realistically be recruited and retained? |
Broaden recruitment, add sites, modify scope, or reconsider the population. |
| Data |
Is access confirmed, and do the data contain what the study needs? |
Secure access before committing or identify another valid data source. |
| Measurement |
Can the central variables or phenomena be measured appropriately? |
Use a suitable instrument, conduct measurement development, or revise the question. |
| Expertise |
Does the team have the skills needed for the design, procedures, and analysis? |
Train, collaborate, obtain specialist support, or simplify the study. |
| Time |
Can every dependent stage be completed before the real deadline? |
Reduce scope, shorten appropriate procedures, or revise the schedule. |
| Budget |
Are all essential costs covered? |
Secure funding, reduce costs without compromising validity, or redesign. |
| Infrastructure |
Will the necessary facilities, technology, equipment, and storage be available? |
Reserve access, identify alternatives, or modify procedures. |
| Permissions |
Can required institutional, data, site, regulatory, or ethical approvals be obtained? |
Begin approval processes early or develop a realistic alternative. |
Separate Known Facts From Optimistic Assumptions
A feasibility assessment becomes much more useful when you distinguish what you know from what you hope.
“The hospital sees approximately enough eligible patients” is not the same as having verified screening data. “My supervisor knows a statistician” is not the same as having a statistician committed to the project. “The dataset should contain that variable” is not confirmed data availability.
For each essential requirement, label it as confirmed, probable but unverified, or uncertain. Then investigate the uncertain items with the greatest potential to stop the project.
Identify Your Single Biggest Feasibility Risk
Many studies have one assumption that dominates all the others. Perhaps the study works only if a company provides proprietary data, a rare population can be recruited, a laboratory technique performs reliably, or participants remain in follow-up for a year.
Identify that dependency explicitly and test it as early as possible. Solving minor details while leaving the project-killing assumption unexamined creates an illusion of progress.
Know When a Pilot or Feasibility Study Is Appropriate
Sometimes you cannot resolve important feasibility questions from existing information alone. A feasibility study may be useful when uncertainty remains about whether or how a future study can be conducted.
Feasibility studies can examine issues such as recruitment and retention, data-collection procedures, outcome measurement, acceptability, resources, implementation, and delivery. In preparation for randomized trials, a pilot study is commonly treated as a subset of feasibility work in which all or part of the future study is conducted on a smaller scale.
The purpose should be explicit. A feasibility study asks questions about whether and how the main study can proceed. It should not simply be a small underpowered version of the definitive study whose main purpose is to make premature claims about effectiveness.
Feasibility assessment during planning
Checking whether the proposed research can realistically be conducted using available information, consultation, estimates, and verification.
Feasibility study
A research study specifically designed to resolve important uncertainties about whether or how a future study can be conducted.
Pilot study
In commonly used methodological definitions, a subset of feasibility research in which the future study, or part of it, is conducted on a smaller scale.
Define Go, Modify, and Stop Conditions Before Problems Occur
When major uncertainty remains, decide what evidence would make you proceed, modify the design, or stop.
For example, you might define acceptable recruitment and retention rates, data completeness, intervention adherence, measurement performance, or operational milestones before feasibility testing begins. Predefined progression criteria can make decisions less dependent on optimism after problems appear.
Not every student project requires a formal traffic-light progression framework. The broader principle still applies: know which assumptions must be true for the project to remain viable.
A Smaller Study Is Not Automatically More Feasible
When a project seems difficult, reducing the sample size is often the first response. That may reduce cost and workload, but it can also make the study incapable of answering its research question.
Similarly, removing a comparison group, shortening follow-up, replacing a validated measurement with an easy questionnaire, or recruiting only whoever is convenient may make data collection simpler while weakening the scientific value of the study.
A feasible study must still be capable of producing credible evidence. If simplification destroys answerability, you have not solved the feasibility problem. You have created a different, weaker study.
Be Willing to Change the Question
Feasibility assessment is not merely an exercise in proving that your preferred idea can work. Sometimes the correct result is that it cannot work in its current form.
You may need to narrow the population, change the outcome, use an accessible data source, reduce the number of objectives, choose another method, collaborate with another site, or investigate one component of a larger problem.
That is useful research planning. A focused study you can complete rigorously is generally preferable to an ambitious design that cannot deliver credible evidence.
