---
name: independent-research-planning
description: Use when turning a research question or idea into a feasible investigation, protocol, or project plan.
---

# Research planning

Build a plan that lets the researcher decide what to do next and what observations would change the conclusion. Respect the maturity and scope of the project: a pilot, a protocol, and a completed study require different levels of detail.

## Working method

1. State the problem, intended contribution, and decision the work is meant to inform. Separate the established context from the untested idea. If the aim, setting, or constraints are missing, make a clearly labeled working assumption and identify the information that could alter the plan.
2. Formulate a question specific enough to study. Specify the system or population, conditions or comparison, outcome, and time or parameter range where relevant. For theory or computation, identify the model, quantities, assumptions, and proposed checks. Express hypotheses as testable predictions when appropriate; do not force a hypothesis onto exploratory work.
3. Choose methods that can answer the question. Describe inputs, sampling or case selection, procedure, measurement or computational approach, analysis, and suitable controls or benchmarks. Specify relevant inclusion and exclusion rules, quality checks, and how missing or unreliable inputs will be handled. Explain how the design addresses plausible alternatives and where it cannot. Treat power or precision targets as design choices supported by assumptions, not as results.
4. Map practical dependencies: access to data or equipment, permissions and ethics review where applicable, skills, budget, calendar, and reproducibility requirements. Identify the most consequential risks and a feasible fallback or decision point for each.
5. Provide ordered milestones and concrete outputs, with estimates only when supported by available information. Where practical, state decision criteria for continuing, revising, or stopping and the evidence needed to apply them. Mark approvals, data collection, calculations, and validation as pending until they occur. Suggest the smallest useful next action that reduces a consequential uncertainty.

When a protocol will affect people, animals, sensitive data, or other regulated work, identify the applicable review and expertise required without claiming that the proposed design has been approved. Do not present a plausible gap as established novelty without a current, traceable literature assessment.