Who needs a DMP?
For many researchers, Data Management Plans do not exactly spark enthusiasm. They come with the grant requirements, funder template, or institutional checklist, and are often perceived as yet another document that needs to be completed before the real work can continue. It doesn’t have to be this way, believes Niels Geudens, data steward at the VIB Data Core. He explains why DMPs should become more than a compliance exercise, and how the new VIB Data Management Plan Platform can help make the information researchers provide more useful over time.
“Researchers often feel they are writing something down because someone decided it had to be written down,” says Niels Geudens. “The frustrating part is that, afterwards, it can feel as if nobody does anything with that information.”
That is one of the problem’s the new VIB Data Management Plan Platform aims to address. The idea is to make the information in a DMP more structured, reusable, and useful, not only for researchers, but also for support teams and (eventually) institutional reporting.
From compliance task to useful planning
“Yes, DMPs are administrative,” Niels admits. “But that does not mean they cannot be useful. A good DMP makes you think about practical questions before they become problems.”
Where will the data be stored? Who needs access? Will the project involve sensitive data? What happens when the project ends? How will datasets be documented, shared, or preserved?
“These are not abstract questions,” he explains. “If a project will generate huge volumes of data, someone needs to think about storage costs. If personal data are involved, someone needs to think about access, ethics, and data protection. A DMP should help surface those questions early.”
The challenge is that many existing DMP workflows are still largely text-based. Researchers fill in fields, export a document, and move on. The information may be present, but it is difficult to reuse.

Machine actionable
In Flanders and across the broader research landscape, funders and policymakers increasingly want DMP information to be usable for follow-up, reporting, and open science monitoring. A PDF alone is not enough for that. The information needs to be structured before it can be reused reliably.
“That is what people mean with a machine-actionable DMP. The idea is that the information is entered in a structured way, so systems can understand it, reuse it, and connect it to other processes.”
In practice, this means shifting from a DMP as a static document and to a DMP as structured project information. That information could be reused across different templates, exported for different funders, or eventually connected to VIB systems that support reporting, infrastructure planning, or data stewardship.
“For researchers, the first benefit should be very practical. The platform guides you through the questions, gives support while you write, and makes it easier to maintain your DMP over time.”
The new platform also allows information to be reused: “A DMP created for one context can provide a starting point for another. Information is not locked into one specific document layout, but can be adapted to different templates.”
Reduce duplicated effort
Today, researchers and PIs may be asked for the same project information multiple times, especially when reporting deadlines approach. Structured DMP information could help make that process more efficient in the future.
“We will still need to develop the necessary integrations, but the idea is that the DMP platform and reporting will be integrated. Formal DMP policy requirements will be communicated once we’ve reached that point.”
For now, the VIB Data Core strongly recommends using their platform for new DMPs and major updates, especially for projects where VIB has reporting responsibilities.
“As always when changing platforms, there is an initial effort. Because the platform asks structured questions instead of giving you one big text field. That can look like more work at first. But the point is that the information becomes more reusable afterwards.”
Early users have already experienced that value. One of them is VIB-KU Leuven Center for Cancer Biology PI Jonas Demeulemeester. For him, the collaboration feature and flexible export option are the two most important benefits of the new platform.
“It was very straightforward to work on the DMP together,” says Jonas. “I completed it together with my postdoc Laurens, using comments, to-do’s, and assigned fields to divide the work. We could move through it quite smoothly, and the option to export the information in different funder-specific formats is a clear advantage.”
Even easier in the future
The current launch focuses on helping researchers create and maintain DMPs in a more structured way. But the longer-term outlook goes further.
One future direction the Data Core is exploring is AI-assisted DMP preparation. In such a scenario, researchers could upload existing project information, such as a grant application or work package description, and use that as a starting point for completing parts of the DMP.
“That would remove some of the initial barriers,” Niels says. “Much of the information needed for a DMP already exists somewhere else. In a grant or project application, in previous DMPs… The question is how to reuse it safely and responsibly.”
That last part is crucial. AI-assisted workflows would need careful assessment before they could be offered to researchers. Project applications may contain sensitive, confidential, or proprietary information, so questions around security, privacy, validation, and institutional approval need to be addressed first.
“A DMP should not be the end point,” Niels says. “It should be a starting point for better support.”

.png)
.jpg)
