DMP lifecycle

The DMP lifecycle is the framing that stops a Data Management Plan (DMP) from being treated as a one-off proposal attachment. It names three phases a plan passes through over a project’s life: the creation phase at proposal stage, an active phase during project execution, and the closeout phase at or after project end. Each phase has a different primary audience (a funder reviewer at creation, the project team and data steward during the active phase, an auditor or future re-user at closeout) and a different relationship to truth: creation-phase content is intent (“we will…”), active-phase content is commitment being tracked, and closeout-phase content is evidence of what actually happened.

A research office typically encounters the lifecycle framing in three places: reviewing a draft DMP at proposal stage against a funder template, checking in on a DMP mid-project when a data steward flags a divergence from plan (a new sensitive dataset, a repository that fell through), and reconciling the DMP against actual deposits at project closeout for reporting or audit. Institutions that treat the DMP as a single static document tend to skip the middle step, which is exactly the gap that machine-actionable DMP (maDMP) tooling and the concept of an active DMP are designed to close.

How it differs from adjacent DMP concepts

• vs. DMP creation phase / DMP closeout phase — those are the individual phases; the lifecycle is the whole sequence and the transitions between them.
• vs. Living DMP — a living DMP is a specific implementation choice (versioned, citable) for moving through the lifecycle; not every DMP that passes through creation, active, and closeout phases is “living” in that stricter sense.
• vs. Static DMP — a static DMP nominally still has a lifecycle on paper, but in practice only the creation phase is ever populated; the active and closeout phases are never revisited.

References

• Simms, Jones, Mietchen & Miksa, “Ten principles for machine-actionable data management plans,” PLOS Computational Biology (2019).

Checking this against the current guidance

Which phase of the DMP lifecycle gets skipped most often, and what risk that creates, depends on the project type and funder oversight. The answer depends on which funder oversight you are working to,
and the page above states the general rule.

Ask
CASRAI: Which phase of the DMP lifecycle is most commonly skipped or treated as a formality by research teams, and what compliance risk does skipping it create?

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

How many phases does a typical DMP lifecycle have?

Most frameworks describe four: creation, active, assessment/review, and closeout, though some funders and DMP tools subdivide these further.

Does every project need to go through every lifecycle phase?

In principle yes, but enforcement varies — funders with lighter post-award oversight may never formally check the assessment phase, which is part of why it’s the phase most often skipped in practice.

Is the DMP lifecycle the same as the broader research data lifecycle?

No — the research data lifecycle describes what happens to the data itself (creation, processing, sharing, preservation); the DMP lifecycle describes what happens to the planning document that governs that data.