03 · What You Need to Know
Judge Protocol Detail by Function, Not Page Count
There is no universal page count for an adequate protocol
A protocol is not sufficiently detailed because it reaches 20, 50, or 100 pages. Nor is a short protocol necessarily incomplete. Length is a poor substitute for methodological specificity.
For randomized trials, SPIRIT 2025 defines a protocol as a central document providing sufficient detail to understand the rationale, objectives, population, interventions, methods, statistical analyses, ethical considerations, dissemination plans, and administration; replicate the trial methods and conduct; and appraise validity, feasibility, and ethical rigour. Other forms of research will not necessarily require the same elements, but the underlying test is useful: does the document permit meaningful understanding and evaluation of what is planned?
Before worrying about length, first determine which components actually belong in the protocol for your study.
Important methodological decisions need greater specificity
Not every sentence deserves the same degree of detail. Concentrate precision where alternative choices could materially change who is studied, what evidence is produced, how that evidence is analyzed, or what conclusions can reasonably be drawn.
Consider sampling. Saying that "students will be recruited" may be inadequate if the study depends on which students are eligible, where they are recruited, how they are selected, and how many are needed. Likewise, stating that "academic performance will be measured" leaves unanswered which measure defines performance, when it will be obtained, and how it will be scored.
The same principle applies to analysis. SPIRIT 2025 guidance for randomized trials calls for planned statistical methods to be described sufficiently clearly to support implementation, replication, and critical assessment. The appropriate details depend on the design, but an analytical label alone rarely resolves the decisions that matter.
Use the replication test carefully
A useful diagnostic is to ask whether another competent researcher could understand how the study is intended to operate from the protocol and its referenced documents. If that person would have to make a consequential choice that should have been yours, the protocol may be underspecified.
This does not mean that every research tradition aims for literal procedural replication in the same way. In qualitative inquiry, for example, researcher judgment, interaction, reflexivity, and responsiveness to emerging material may be legitimate parts of the methodology. Adequate detail may therefore involve explaining the methodological approach, sampling logic, data-generation process, analytical procedures, and researcher role rather than pretending that every future interaction can be scripted.
The test is better framed as transparency: could a knowledgeable reader understand what is planned, what remains open, and why?
Distinguish methodological decisions from routine implementation details
Some details affect the scientific meaning of the study. Others simply describe how routine work will be organized.
| Type of information |
Example |
Typical treatment |
| Consequential methodological decision |
Eligibility criteria, primary outcome, sampling strategy, measurement timing, primary analysis |
Specify clearly in the protocol or a formally referenced study document |
| Decision rule |
How incomplete responses will be handled or when a participant is considered withdrawn |
Specify when the rule can affect study conduct or interpretation |
| Specialized operational procedure |
Detailed data-entry checks, laboratory processing instructions, interviewer procedures |
May be described in the protocol or an associated plan or standard operating procedure |
| Routine administrative detail |
Minor internal workflow that does not affect scientific interpretation or participant protection |
Include only if needed for implementation, oversight, or applicable requirements |
The boundary is contextual. A detail that is trivial in one study can become scientifically important in another. The timing of an assessment, for instance, may be incidental for one descriptive variable but crucial when an intervention effect is expected to change rapidly over time.
Describe procedures at the point where ambiguity would create inconsistent conduct
Imagine a protocol stating that participants will complete a follow-up assessment "approximately one month later." Is that adequate? Perhaps, but not if a measurement at 25 days and one at 45 days could produce meaningfully different results.
Where consistency matters, the protocol should establish the relevant timing, sequence, conditions, roles, or acceptable windows. For randomized trials, SPIRIT 2025 specifically recommends documenting schedules of enrolment, interventions, and assessments, including target time points and, where appropriate, acceptable time windows.
The principle extends beyond trials. Detail should increase when ambiguity could lead different researchers, sites, or participants to experience materially different procedures.
Analytical detail matters because many reasonable choices may exist
A dataset rarely dictates a single analysis automatically. Researchers may have choices about outcome definitions, exclusions, transformations, covariates, missing data, subgroup analyses, model specifications, coding procedures, or other analytical decisions.
If those choices are central to answering the primary research question, postponing them until after the data are examined can make it difficult to distinguish planned analysis from result-contingent decision-making. Protocol detail therefore has a temporal function as well as an explanatory one.
SPIRIT 2025 guidance illustrates this principle for randomized trials by calling for specification of the main analysis, analysis population, missing-data methods, effect measures, covariate adjustment, and relevant additional analyses. A separate statistical analysis plan may contain greater technical detail, but the protocol should still identify the key analytical approach.
You do not need to put every detail in the main protocol
A protocol can remain readable while relying on controlled supporting documents. A statistical analysis plan may describe models and decision rules in greater depth. A data-management plan may cover coding, storage, validation, access, and security. Intervention manuals, laboratory manuals, interview guides, case-report forms, or standard operating procedures can provide implementation detail.
The important point is not physical location. Important information should be documented somewhere appropriate, clearly referenced when necessary, available to the people who need it, and consistent with the main protocol.
Protocol completeness
The important scientific, ethical, and operational decisions are identifiable and sufficiently specified.
Protocol exhaustiveness
Every conceivable procedural detail is placed inside one document, whether or not it helps readers understand or conduct the study.
A strong protocol aims for completeness, not exhaustiveness.
Some details should intentionally remain conditional
Writing a detailed protocol does not require pretending that every future event can be known. Certain decisions may depend legitimately on information that becomes available during the research.
The better response is not to manufacture false precision. Identify which decisions can legitimately remain flexible and, where possible, document the conditions or criteria governing them.
For example, a qualitative sampling strategy may legitimately evolve as analysis identifies cases requiring further exploration. The protocol can explain that logic in advance rather than inventing an exact final sample composition that the methodology does not require.
More complex and higher-risk studies usually require more detail
The consequences of ambiguity are not equal across studies. A single-site, minimal-risk project with straightforward procedures may need less operational documentation than a study involving multiple sites, complex interventions, sensitive data, safety risks, or several teams performing coordinated procedures.
Greater complexity can require clearer allocation of responsibilities, stronger data-management procedures, quality controls, monitoring arrangements, decision rules, and communication pathways. Applicable ethics, regulatory, funder, sponsor, or institutional requirements may also prescribe particular details.
Watch Out
Do not use a highly detailed clinical trial protocol as a universal model for all research. Protocol guidance should be appropriate to the study design and context. Adding irrelevant sections can make a protocol longer without making the research better planned.
