03 · What You Need to Know
A Feasibility Check Should Resolve a Decision-Relevant Uncertainty
Feasibility work is intended to examine whether and how a proposed study or intervention can proceed. Contemporary guidance emphasizes identifying key uncertainties and using feasibility findings to decide whether research should proceed, be modified, require further feasibility work, or stop.
This makes the starting point surprisingly simple: What do you still need to know before you can responsibly design or commit to the larger study?
First Ask Whether You Need Empirical Testing at All
Not every uncertainty warrants a pilot or feasibility exercise.
If you need to know whether your institution licenses particular software, check the license. If you need an equipment specification, consult the documentation or facility manager. If you need the cost of transcription, obtain realistic quotations. If you need to know whether a dataset contains a variable, inspect the data dictionary or consult the data custodian.
Running a study to answer a question that could have been resolved by an email is an unusually expensive form of correspondence.
Verification question
The uncertainty can reasonably be resolved through records, documentation, calculation, formal confirmation, or specialist consultation.
Empirical feasibility question
The uncertainty concerns what will happen when a study process is attempted under realistic conditions.
Before testing anything, therefore, check whether the critical resource or condition can simply be verified directly.
Good Feasibility Questions Concern What You Need to Learn
A small feasibility check is most useful when its objective is explicit.
Examples include:
- Can enough eligible participants be identified and approached through the proposed recruitment pathway?
- What proportion of approached participants appears willing to enroll?
- Can the proposed data-collection procedure be completed within the available appointment time?
- Do participants understand the questionnaire items as intended?
- Can study staff deliver the procedure consistently?
- Are the required variables actually captured with sufficient completeness?
- Can the proposed workflow move data from collection through processing and analysis without a major operational failure?
These questions are more useful than the vague objective “to see whether the study works.”
Pilot and Feasibility Studies Are Related but Not Identical Concepts
Terminology varies across fields, but methodological guidance commonly treats feasibility as the broader question of whether and how a future study can be done. A pilot study is generally understood as a smaller version or rehearsal of some or all procedures intended for the future study.
Thus, a pilot can be a type of feasibility study, but feasibility work does not necessarily require running a miniature full study.
| Approach |
Typical purpose |
Example |
| Documentary verification |
Confirm an existing fact or requirement |
Check whether a dataset contains required variables. |
| Targeted feasibility check |
Test one uncertain process or condition |
Try the proposed recruitment pathway for a limited period. |
| Pilot study |
Rehearse several components of the intended study together |
Run a small version of recruitment, allocation, data collection, and follow-up. |
| Full study |
Address the substantive research objectives |
Conduct the adequately designed main investigation. |
The terminology should be used consistently with the conventions and guidance relevant to your field. More important than the label is whether the preliminary work has a clearly defined feasibility purpose.
Run the Check Before the Uncertainty Becomes Expensive to Discover
Timing is central. A feasibility check has greatest value when its findings can still change the design.
If you are uncertain whether participants can tolerate a 90-minute assessment, test that before committing to a protocol requiring hundreds of 90-minute visits. If you do not know whether a recruitment pathway can generate enough participants, investigate before building a fixed timeline around an assumed recruitment rate.
Once the full study has begun, discovering the same problem may require amendments, additional funding, extended recruitment, replacement measures, or abandonment of collected data.
Focus on the Critical Assumption
Your project may contain dozens of uncertainties. Do not pilot all of them simply because they exist.
Prioritize the assumption that could make the study impossible if it is wrong, or another uncertainty whose resolution would materially change your design decision.
If the study depends on recruiting 25 participants per month and you have no evidence about recruitment performance, that uncertainty may deserve testing. If you are uncertain whether interviews will average 47 or 52 minutes, a formal feasibility study may contribute very little unless that difference has meaningful operational consequences.
Define What Result Would Change Your Decision
A feasibility check becomes much more useful when you decide beforehand how its findings will inform the next step.
In formal pilot and feasibility research, these are often expressed as progression criteria. The logic can also be useful in smaller research projects.
Proceed The result supports the planned design sufficiently to continue.
Proceed with modification The study appears feasible, but a procedure, timeline, recruitment strategy, or resource plan needs adjustment.
Investigate further The check leaves a consequential uncertainty unresolved and additional feasibility work is justified.
Do not proceed as planned The result indicates that the current design is unlikely to be workable under available conditions.
The exact criteria should fit the uncertainty being tested rather than being borrowed mechanically from another study.
A Feasibility Check Does Not Need to Estimate the Main Effect
A frequent mistake is treating a small pilot primarily as an opportunity to see whether the intervention “works” or whether a substantive hypothesis reaches statistical significance.
Small feasibility studies are generally not designed or powered to provide definitive estimates of effectiveness. Their primary value lies in questions such as recruitment, retention, acceptability, delivery, measurement, procedures, and the practicality of the intended evaluation.
Watch Out
Do not turn a small feasibility exercise into an underpowered hypothesis test and then use a nonsignificant result as evidence that the substantive research question is unimportant. Feasibility and effectiveness are different questions.
Recruitment Is a Particularly Useful Target for Feasibility Testing
Recruitment frequently determines whether a planned study can achieve its required sample. Useful preliminary information can include the number screened, the proportion eligible, the proportion consenting, recruitment per site or time period, reasons for nonparticipation, and whether the recruitment strategy is acceptable.
This can complement your broader estimate of how much time recruitment and data collection will realistically require.
Test the Process Under Conditions That Resemble the Real Study
A feasibility check is only as informative as its relationship to the future study.
If recruitment is tested using a population or communication channel that will not be available in the full project, the observed recruitment rate may be misleading. If an instrument is tested only among colleagues when the eventual participants have very different characteristics, important usability problems may remain hidden.
The preliminary conditions do not need to reproduce the future study perfectly, but they should be similar enough to inform the uncertainty being investigated.
Do Not Forget Ethics and Governance
Calling an activity a “feasibility check” or “pilot” does not automatically exempt it from research ethics, consent, privacy, data-protection, organizational permission, or other applicable requirements.
The appropriate requirements depend on what the activity involves and on the relevant institution and jurisdiction. If the check involves people, identifiable information, protected data, interventions, or other regulated research activities, determine what approvals are required before conducting it.
Plan What Happens to Feasibility Data
Decide whether data from the preliminary work are solely for feasibility purposes or whether there is a defensible possibility of incorporating them into a later study. In some designs, an internal pilot may form part of the main study. In others, changes made after feasibility testing mean that combining data would be inappropriate.
This should not be decided casually after seeing the results. Consider the design, protocol, analysis implications, and relevant ethical requirements in advance.
Sometimes the Best Result Is Learning Not to Proceed
A feasibility check is not unsuccessful because it reveals that the proposed study cannot work as planned.
If a small test shows that recruitment is far below what the project requires, participants find the procedure unacceptable, essential data are systematically missing, or the operational workflow cannot be delivered within available resources, the check may have prevented a much larger failure.
That information can support redesign, collaboration, a different site, a different method, postponement, or the decision to preserve a valuable research idea that is not feasible right now.
04 · A Practical Example
Testing Recruitment Before Building the Full Study Around It
Hypothetical Example
A study requiring a specialized participant group
A researcher plans a study requiring 180 participants from a relatively narrow professional population. The project has six months for recruitment, which means an average of 30 enrolled participants per month. An external professional organization is willing to distribute recruitment invitations, but there is no previous information about how many members are likely to respond.
Identify the critical uncertainty The organization can distribute invitations, but the researcher does not know whether the recruitment pathway can produce anything close to the required enrollment rate.
Check what can be verified first The researcher confirms the number of potentially eligible members, the organization's communication procedure, how frequently invitations can be distributed, and any restrictions on recruitment.
Run a limited feasibility check Subject to applicable approvals, the recruitment process is tested over a defined period using procedures that resemble those planned for the full study.
Record feasibility information The researcher examines how many potential participants could be reached, how many expressed interest, how many were eligible, how many enrolled, and what practical problems occurred.
Use the result to make a decision If recruitment is close to the required rate, the original plan gains empirical support. If it is substantially lower, the researcher can add recruitment pathways, extend the timeline, revise the sample strategy where methodologically defensible, or redesign the study before committing to the full project.
The purpose of the check is not to estimate the study's eventual substantive findings. It answers the question that currently threatens feasibility: can the proposed recruitment mechanism realistically deliver the study?
