DMP narrative

The DMP narrative is the human-readable prose portion of a Data Management Plan (DMP), typically organised under the section headings of whichever funder or institutional template applies — data description, standards and metadata, storage and security, sharing and access, responsibilities, and resourcing are common headings across NIH, NSF, UKRI, and Horizon Europe templates. For many funders the narrative remains the primary deliverable even where a machine-actionable DMP (maDMP) also exists behind it, because the narrative is what explains and justifies decisions — why this repository, why this embargo length, why these particular access controls — in a way that a structured field alone cannot.

A research office deals with the narrative constantly: it is the document reviewed, commented on, and ultimately submitted with a grant application, and it is usually the artefact a programme officer actually reads. In hybrid workflows that use a machine-actionable DMP (maDMP) tool such as DMPRoadmap or Argos, the narrative and the machine-readable version are generated from the same underlying answers, so editing the narrative and exporting a structured JSON file are two views of one dataset rather than two documents to keep in sync by hand.

How it differs from adjacent DMP concepts

• vs. DMP machine-readable expression — the narrative is prose for a human reader; the machine-readable expression is a typed, structured serialisation for software.
• vs. DMP template — the template is the question set; the narrative is the set of answers written out in prose.

References

• Science Europe, Practical Guide to the International Alignment of Research Data Management (2021).

Checking this against the current guidance

Whether the narrative DMP still has to be submitted alongside a machine-readable version depends on the specific funder’s submission system. The answer depends on which funder submission system you are working to,
and the page above states the general rule.

Ask
CASRAI: If a funder requires a machine-readable maDMP expression, does the narrative DMP still need to be submitted separately, or does the machine-readable version replace it?

It searches CASRAI’s indexed corpus of research-administration guidance and cites
the passage behind each claim, so you can open the source and check it rather than
take its word — and it says so when the corpus does not cover something instead of
guessing. Two questions a day are free while you are signed out, no
account and no card. Everything CASRAI publishes stays free to read.

Frequently asked questions

Is the narrative DMP the default format funders expect?

Yes, for most funders today the narrative, prose-based plan is still the primary or only format required, with machine-readable expressions as an emerging supplement rather than a replacement.

What’s the typical length of a narrative DMP?

Length limits vary widely by funder, from roughly one page for some NSF programs to several pages for larger, multi-investigator or international projects.

Can a narrative DMP reference an external document instead of restating everything inline?

Some funders allow this for lengthy technical protocols, but the core commitments (repository, access terms, retention period) generally still need to appear in the narrative itself, not only by reference.