How to Write a Research Protocol

A research protocol is the plan you commit to before you start — screening studies, collecting data, or running analysis. It exists to lock in your methodology before you've seen results, which is exactly what makes it credible: a protocol written after the fact can always be quietly shaped around whatever the data showed.

What a protocol needs to include

• Background & rationale — why this question matters and what gap it addresses.
• Objectives — a specific, answerable research question, often framed with PICO.
• Eligibility criteria — exactly what will and won't be included, defined before screening starts.
• Methodology — study design, search strategy (for reviews), or data collection procedures (for primary research).
• Planned analysis — the statistical or synthesis methods you intend to use, decided in advance.
• Timeline — realistic milestones for each stage.

Why registering it matters

For systematic reviews, registering a protocol — commonly with PROSPERO — creates a public, time-stamped record of your plan. See our PROSPERO registration guide for the specific steps. For other study types, an institutional ethics or protocol review board often serves the same function.

Common mistakes

• Writing the protocol loosely enough that eligibility criteria can be "adjusted" once screening reveals inconvenient results.
• Leaving the analysis plan vague, then choosing a method after seeing which one produces a significant result.
• Skipping registration entirely when it's expected for your review type or field.

A tight protocol is more work upfront and considerably less work — and less risk — everywhere after.