thinkhard.SI

Prompt repository Research & learning

Write a data management plan for a research project

Your funder or institution asks for a data management plan, and you need one that follows its template and promises only what your project will really do.

Registry no.
RES-043
Collection
Research & learning
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
8

Research & learningAny chat

The prompt

Score 10 of 10, Excellent

Use it when Your funder or institution asks for a data management plan, and you need one that follows its template and promises only what your project will really do.

  • {{funder}}
  • {{template}}
  • {{project}}
  • {{data_description}}
  • {{storage}}
  • {{repository}}
  • {{consent_terms}}
  • {{retention}}
Full prompt and why it works
Prompt
Write a data management plan for my research project, following {{funder}}'s template, for the reviewers who assess it and for my team to follow during the project, so that it meets the funder's requirements and every commitment in it is one we will actually keep.

The funder's template or required headings, with any length limit:
"""
{{template}}
"""

About the project (aims, methods, timeline, partners):
"""
{{project}}
"""

The data (types and formats, volume, how it is collected, whether it includes personal or sensitive information): {{data_description}}
Storage and backup we have access to: {{storage}}
Repository we plan to use for sharing, if decided: {{repository}}
Consent terms that limit sharing, if any: {{consent_terms}}
Retention rules from our institution or the funder: {{retention}}

Follow the funder's headings exactly, in their order. If no template is given, use these headings: Data description; Documentation and metadata; Storage and backup during the project; Ethics, consent and legal requirements; Data sharing and access; Long-term preservation and retention; Roles and responsibilities.

Under each heading, say what we will do in specific terms: formats, file naming and versioning, where data is kept, who has access, how personal data is protected or removed, when and where data will be shared and under what license, and how long it is kept. Keep it within the funder's limit, or under 1,200 words if none is given.

A good plan is specific enough to check against later, for example "Interview recordings are deleted once transcripts are checked, within 3 months of each interview" rather than "Data will be handled securely." That example shows the form only.

Do not:
- Promise open sharing of data that the consent terms do not allow, or that would identify participants.
- Name a repository, license or standard we have not chosen as if it were decided; put suggestions in a note marked [SUGGESTION, CHECK FIT].
- Add storage systems, security measures, costs or staff we do not have.

If our details do not settle a choice a heading needs, write [DECISION NEEDED: what] under it, and list these decisions at the end with who should make each.

Variables

{{funder}}
The funder or institution asking for the plan.
{{template}}
The funder's template or required headings and any length limit, or "none given".
{{project}}
A short description of the project: aims, methods, timeline and partners.
{{data_description}}
What data you will collect or create, in what formats, how much, and whether it is personal or sensitive.
{{storage}}
Storage you can use, for example "the university's research drive, backed up nightly".
{{repository}}
Where you plan to share data, or "not decided".
{{consent_terms}}
What participants agreed to about sharing, or "no participants".
{{retention}}
How long data must be kept, from institutional or funder rules, or "not known".

Why it works

  • What good looks like. Each commitment has to be specific enough to check against later, which keeps vague promises such as "handled securely" out of the plan.
  • What to avoid. No sharing beyond what consent allows, and no repository, license or security measure presented as decided when it is not.
  • If unsure. Open choices become [DECISION NEEDED] markers listed with who should make each, so the plan never commits the team by accident.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/data-management-plan

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "data-management-plan"}, or just ask for it by name.