What Feasibility Questions Should You Answer Before Conducting a Definitive Study?

What Do You Need to Know Before Committing to a Definitive Study?

You are preparing a definitive research study and want to establish whether its design can be implemented successfully. You may already have identified the research question, target population, intervention or exposure, outcome measures, and planned analysis.

Yet several practical questions remain. Can enough participants be recruited? Will they remain in the study? Can the intervention be delivered consistently? Are the measurements acceptable and sufficiently complete? Can the research team manage the required data, resources, and institutional arrangements?

It would be tempting to answer every possible feasibility question before proceeding. However, an excessively broad feasibility investigation can consume substantial resources without resolving the uncertainties that actually threaten the definitive study.

Which questions should be prioritized, how should they be translated into measurable outcomes, and what evidence is sufficient to support a decision about proceeding?

Answer the Questions That Determine Whether the Definitive Study Can Produce Credible Evidence

Identifying and Evaluating the Feasibility Questions That Matter

What Is a Feasibility Question?

A feasibility question asks whether and how a proposed study or intervention can be conducted under the intended conditions.

It concerns the practicality, acceptability, or methodological workability of the research rather than directly testing the definitive substantive hypothesis.

For example, "Can the research team recruit enough eligible students within one semester?" is a feasibility question.

By contrast, "Does an AI-supported feedback intervention improve student achievement?" is ordinarily an effectiveness question.

The distinction matters because the evidence needed to answer these questions differs.

A small preliminary study may provide useful information about recruitment while remaining unsuitable for definitive effectiveness testing.

How Should Feasibility Questions Be Selected?

Begin by identifying what must work for the definitive study to answer its research question.

Then examine which requirements remain uncertain and what would happen if they failed.

For example, a randomized educational intervention may depend on sufficient class recruitment, appropriate allocation, consistent delivery, acceptable outcome assessment, and accurate data management.

Not every component necessarily requires a separate feasibility outcome. Some may already be supported by relevant evidence or established procedures.

The most important questions are those whose answers could change the study design, required resources, implementation arrangements, or decision to proceed.

What Are the Main Domains of Feasibility?

Bowen and colleagues identify several areas relevant to feasibility research, including acceptability, demand, implementation, practicality, adaptation, integration, expansion, and limited-efficacy testing.

These domains provide a useful conceptual starting point, although they should not be treated as a mandatory checklist for every study.

For research preparing a definitive trial or other substantial investigation, the following operational domains are often particularly relevant.

The domains may overlap. For example, intervention burden may influence both acceptability and adherence, while assessment scheduling may affect retention and data completeness.

What Recruitment Questions Should Be Answered?

Recruitment feasibility concerns whether the study can identify, approach, screen, and enroll enough eligible participants within the available time and resources.

Relevant questions may include how many eligible individuals are accessible, what proportion respond to invitations, and how many agree to participate after learning the study requirements.

Researchers should distinguish recruitment speed from recruitment yield.

For example, enrolling 20 participants in four weeks provides an observed recruitment rate of five participants per week. It does not reveal the proportion of approached individuals who enrolled unless the relevant denominator is known.

Recruitment estimates should also be interpreted in relation to institutional calendars, recruitment channels, and population characteristics.

For more detailed discussion, see Should Recruitment Rate Be an Outcome in a Feasibility Study?.

What Retention Questions Should Be Answered?

Retention feasibility concerns whether participants remain in the study long enough to complete the activities and assessments required by the definitive design.

Researchers may ask whether follow-up schedules are practical, whether reminders are acceptable, and why participants withdraw.

For example, a semester-long study may recruit students successfully but encounter substantial attrition during examination periods.

A feasibility study can examine whether follow-up procedures need modification.

However, retention should not be confused with outcome data completeness. A participant may remain enrolled while failing to complete a particular assessment.

Both outcomes may require separate measurement.

What Questions Concern Intervention Delivery?

Intervention feasibility concerns whether the planned activities can be implemented with sufficient consistency and appropriate resources.

Researchers may ask whether implementers can complete training, whether intervention components are delivered as intended, and whether participants engage with the activities.

For example, teachers may be able to operate an AI-supported feedback system but find its required review procedures too time-consuming during ordinary teaching.

These findings can inform training, scheduling, or workflow modifications.

However, successful delivery does not establish that the intervention improves the intended outcomes.

Implementation feasibility and effectiveness should remain analytically distinct.

What Is the Difference Between Fidelity and Adherence?

Fidelity commonly concerns whether the intervention is delivered according to its intended design.

Adherence may refer to whether implementers follow the prescribed procedures or whether participants complete intended activities, depending on the framework used.

For example, a teacher may deliver all planned feedback sessions while some students access only a portion of them.

The intervention may therefore have high delivery fidelity but lower participant engagement.

Researchers should define these constructs operationally rather than treating them as interchangeable.

See Should Intervention Adherence Be an Outcome in a Feasibility Study?.

What Acceptability Questions Should Be Answered?

Acceptability concerns how participants, implementers, or other relevant stakeholders experience the proposed procedures or intervention.

Researchers may examine perceived burden, appropriateness, ethical concerns, and willingness to participate.

For example, students may value a proposed online learning intervention but object to extensive monitoring of their digital activities.

Teachers may consider an intervention educationally worthwhile while finding its workload impractical.

These responses may suggest modifications to the design or implementation arrangements.

However, acceptability should not be inferred solely from high participation rates or general satisfaction.

Participants may enroll despite concerns, and favorable initial attitudes may change during implementation.

What Outcome Assessment Questions Should Be Answered?

Outcome assessment feasibility concerns whether the intended measurements can be obtained under appropriate conditions.

Relevant questions include whether participants understand instructions, whether assessments can be completed at the intended times, and whether scoring and recording procedures operate correctly.

For example, a study may require students to complete a performance assessment immediately after an intervention. A feasibility investigation may reveal that the schedule conflicts with institutional examinations.

Researchers may need to modify the assessment window.

However, successful administration does not automatically establish measurement validity or reliability.

The suitability of the instrument for the intended interpretation requires appropriate supporting evidence.

Why Is Data Completeness a Feasibility Question?

A study may recruit participants and administer assessments successfully yet obtain incomplete or unusable records.

Data completeness concerns whether the information required for analysis is available.

For example, a longitudinal study may have complete baseline questionnaires but missing follow-up scores for a substantial proportion of participants.

Researchers should examine the location, frequency, and possible causes of missingness.

However, a small feasibility study may not establish the missing-data mechanism expected in the definitive study.

The objective is to identify important risks and determine whether the proposed data collection procedures require modification.

See Should Data Completeness Be an Outcome in a Feasibility Study?.

What Questions Concern Research Data Management?

Data management feasibility concerns whether research information can be captured, organized, protected, linked, and prepared for analysis.

For example, a mixed-methods study may require quantitative survey responses to be connected with follow-up interview records.

Researchers should determine whether suitable identifiers can be used, whether the necessary linkage is permitted by consent arrangements, and whether data transfers preserve the intended information.

Technical tests using synthetic records may identify problems before real participant data are collected.

However, technical success does not guarantee methodological suitability or compliance with applicable data governance requirements.

What Questions Concern Randomization and Allocation?

For randomized studies, feasibility questions may concern participant acceptance of randomization, correct implementation of the allocation sequence, allocation concealment, and the practicality of assigning individuals or clusters.

For example, a cluster-randomized educational trial may need to determine whether classes can be recruited before allocation and whether intervention materials can be kept separate between conditions.

A feasibility study may identify operational weaknesses or contamination risks.

However, similar baseline characteristics between groups do not establish that randomization was implemented correctly.

The relevant evidence concerns the allocation mechanism and its implementation.

What Questions Concern Participant Understanding and Consent?

Researchers should consider whether prospective participants understand the study and can make voluntary participation decisions.

This may be especially important when procedures are complex or recruitment involves relationships of authority.

For example, students recruited by their own teachers may perceive participation as connected to academic assessment.

Pretesting may identify misunderstandings, but ethical safeguards should be established before recruitment begins.

Consent completion rates should not be treated as evidence of adequate comprehension.

Researchers should distinguish recruitment feasibility from the ethical adequacy of informed consent.

What Questions Concern Resources and Costs?

Resource feasibility concerns whether the study can be implemented within available staffing, funding, infrastructure, and time.

Researchers may need to estimate recruitment costs, intervention delivery time, assessment workload, technical support, and data management requirements.

For example, an educational intervention may require more teacher training and administrative coordination than anticipated.

A feasibility study can identify these demands and inform more realistic resource planning.

However, resource estimates obtained in a small study may not scale proportionally to larger implementation.

Fixed costs, economies of scale, and institutional differences may affect definitive-study requirements.

Should Institutional Readiness Be Evaluated?

Institutional readiness may be consequential when the research depends on organizational support, permissions, infrastructure, or staff coordination.

For example, a multisite study may require each university to provide access to a learning management system, assign local coordinators, and permit specified data collection procedures.

Researchers should examine whether these requirements can be met consistently.

However, institutional willingness to participate does not necessarily establish operational readiness.

Formal permission, staff capacity, and practical implementation arrangements may require separate evaluation.

How Do You Turn a Broad Feasibility Concern Into a Researchable Question?

Broad questions such as "Is recruitment feasible?" are useful starting points but may be insufficiently specific for designing an assessment.

Researchers should identify the relevant population, procedure, outcome, and decision context.

For example, the question could be refined to: "Can the study enroll the required number of eligible students per month using the proposed institutional recruitment channels?"

The resulting outcome might be the number enrolled per month, accompanied by eligibility and enrollment proportions.

The interpretation would then consider whether the observed rate supports the definitive recruitment timeline.

Specificity helps connect feasibility evidence to a practical decision.

Should Every Feasibility Question Have a Numerical Outcome?

No. Some feasibility questions are best investigated through qualitative evidence or a combination of methods.

For example, a numerical adherence rate may indicate that intervention completion is lower than anticipated, but interviews may be needed to understand whether the problem reflects workload, technical difficulties, or perceived irrelevance.

Similarly, institutional readiness may involve organizational conditions that are not adequately represented by a single percentage.

Researchers should select evidence according to the question rather than assume that every feasibility concern must be converted into a numerical indicator.

How Many Feasibility Questions Should a Study Include?

There is no universal number.

A focused study may need to investigate only a few consequential uncertainties, while a complex multisite intervention may require a broader feasibility assessment.

However, including too many outcomes can make the study difficult to interpret and may divert resources from the most important questions.

Researchers should prioritize questions according to the consequences of failure, uncertainty about the procedures, and whether the resulting evidence could influence the definitive study.

A smaller set of well-justified questions may be more useful than an extensive list of loosely connected indicators.

How Should Feasibility Outcomes Be Operationally Defined?

Each outcome should specify what will be observed and how it will be measured.

For example, participant retention might be defined as the proportion of enrolled participants who complete the primary outcome assessment at the designated follow-up point.

The numerator, denominator, and timing should be clear.

Similarly, intervention adherence should specify which activities count as completed and whether the outcome concerns participants or implementers.

Operational definitions reduce ambiguity and make the findings more useful for planning.

What Is the Difference Between a Feasibility Outcome and a Progression Criterion?

A feasibility outcome is the quantity or observation collected to assess a feasibility question.

A progression criterion specifies how that evidence will contribute to deciding whether the definitive study should proceed.

For example, the observed recruitment rate may be five participants per week. The progression criterion may concern whether that rate is sufficient to achieve the definitive sample within the available timeline.

The criterion should be justified by the study's requirements rather than selected arbitrarily.

Should Progression Criteria Be Established Before the Feasibility Study?

Where progression decisions depend on specified feasibility outcomes, establishing criteria in advance can improve transparency.

For example, researchers may identify the minimum recruitment capacity needed to complete the definitive study within the available resources.

However, progression criteria should not be interpreted mechanically.

An observed rate may be uncertain, and contextual evidence may indicate that identified barriers can be addressed.

Researchers should distinguish prespecified criteria from subsequent judgments and explain consequential departures.

For a detailed discussion, see What Progression Criteria Should Determine Whether a Feasibility Study Leads to the Main Study?.

Can Green-Amber-Red Criteria Be Used?

Yes. A traffic-light approach can distinguish favorable feasibility findings from those requiring modification or substantial reconsideration.

For example, a recruitment indicator may fall within a range supporting progression, a range requiring corrective action, or a range suggesting that the current design is impractical.

However, the thresholds should be justified by the study's requirements.

Researchers should not assume that the same percentages are appropriate for recruitment, retention, intervention fidelity, and data completeness.

The framework supports structured judgment rather than replacing it.

What If Feasibility Findings Are Mixed?

Mixed findings are common because different aspects of a study may present different levels of feasibility.

For example, recruitment may be adequate while outcome completeness is poor.

Researchers should examine the consequences of each finding rather than calculate an arbitrary average across indicators.

A serious problem affecting the primary outcome may outweigh favorable results concerning less consequential procedures.

The appropriate decision depends on whether the weakness can be addressed without undermining the definitive design.

Should Feasibility Questions Include Intervention Effectiveness?

Feasibility studies may collect preliminary outcome data, but they are generally not designed to provide definitive evidence of effectiveness.

Small samples commonly produce imprecise effect estimates.

A statistically nonsignificant difference does not establish that an intervention is ineffective, while a statistically significant preliminary difference does not automatically establish effectiveness.

Researchers should distinguish the feasibility of collecting outcome data from the substantive effect of the intervention.

Leon, Davis, and Kraemer emphasize the importance of interpreting pilot studies according to their preliminary objectives rather than treating them as underpowered definitive trials.

How Should Feasibility Questions Inform Sample Size?

The sample size should reflect the feasibility outcomes and the precision required to support relevant decisions.

For example, estimating a recruitment or retention proportion with useful precision may require more observations than identifying an obvious technical workflow problem.

Qualitative feasibility research may instead prioritize participants who can illuminate relevant implementation conditions.

There is no universal sample size suitable for every feasibility investigation.

Researchers should justify the scope according to the questions being investigated rather than using an arbitrary fraction of the proposed definitive sample.

Can Existing Evidence Answer Some Feasibility Questions?

Yes. Researchers should review relevant previous studies, institutional records, implementation reports, and methodological evidence before deciding what additional information is needed.

For example, recent recruitment data from a comparable study may provide useful evidence about likely enrollment.

However, the applicability of existing evidence depends on differences in eligibility, participant burden, institutional conditions, and recruitment procedures.

Researchers should not assume that a previously observed rate will recur under substantially different circumstances.

The objective is to identify genuine evidence gaps rather than repeat feasibility assessments unnecessarily.

When Should a Separate Feasibility Study Be Conducted?

A separate feasibility study may be appropriate when critical uncertainties cannot be resolved through existing evidence, routine preparation, or focused pretesting.

For example, a proposed multisite intervention may depend on institutional cooperation and staffing arrangements that have not yet been demonstrated.

However, some questions can be answered through technical verification, stakeholder consultation, or an appropriately planned internal pilot.

The choice should reflect the consequences of proceeding without further evidence.

See When Is a Separate Feasibility Study Necessary Before the Main Research?.

How Should Feasibility Findings Be Translated Into Decisions?

The final interpretation should explain what the evidence indicates about the proposed definitive study.

Researchers may conclude that the design can proceed, that particular procedures require modification, that additional feasibility work is necessary, or that the study should not proceed in its current form.

These conclusions should reflect the prespecified objectives, observed findings, uncertainty, and contextual information.

For example, low recruitment may justify expanding recruitment sites if suitable additional sites are available.

However, if recruitment remains inadequate despite feasible modifications, the proposed definitive study may need substantial redesign.

A useful feasibility study produces an informed decision rather than simply a collection of preliminary percentages.

How Should Feasibility Questions Be Reported in a Proposal?

The proposal should identify the unresolved feasibility questions, explain why they matter, and specify how they will be investigated.

Each important question should be connected to an outcome, data collection method, and intended decision.

Where progression criteria are used, the rationale for their thresholds should be documented.

The proposal should also distinguish feasibility objectives from definitive effectiveness or explanatory objectives.

This makes the preliminary study's purpose transparent and helps readers assess whether its design can provide the required evidence.

Developing Feasibility Questions for a Multisite Educational Trial

Illustrative Feasibility Question Matrix

The matrix identifies what needs to be investigated and how the findings will influence the definitive design.

It does not assume that all feasibility outcomes must be favorable. A study may be able to proceed after addressing remediable problems.

Conversely, a serious unresolved problem affecting the primary outcome or essential intervention components may justify further feasibility work or substantial redesign.

Common Misconceptions About Feasibility Questions

How to Build a Focused Set of Feasibility Questions

Begin by identifying the procedures without which the definitive study cannot answer its research question.

Then assess the uncertainty surrounding each procedure and determine whether additional evidence could influence a consequential decision.

For each selected question, specify the outcome, measurement procedure, expected uncertainty, and decision it will inform.

Where appropriate, establish progression criteria before collecting the feasibility data.

The resulting plan should be focused enough to guide research decisions while remaining sufficiently comprehensive to address the study's critical vulnerabilities.

Before Finalizing Your Feasibility Study Questions, Check:

Frequently Asked Questions About Feasibility Study Objectives

Ask Whether the Definitive Study Can Work, and What Evidence Would Establish That

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.