Skip to main content

Data Management Plan (DMP)

Data Management Plan (DMP)

The University of Florence is committed to Research Data Management practices inspired by Open Science principles, with the aim of making scientific outputs transparent, traceable, and collaborativeWithin this framework, the Data Management Plan (DMP) is the key instrument for planning the entire research data lifecycle.

A DMP is a formal document outlining how data generated or collected during a project will be managed, analyzed, preserved, and—where possible—shared. Far from being a mere bureaucratic "box-ticking" exercise attached to funded projects, the DMP is a methodological best practice applicable to all research activities.

Through this tool, researchers plan the collection, preservation, description, and dissemination of their data and metadata in accordance with the FAIR Principles (Findable, Accessible, Interoperable, Reusable), ensuring:

  • Traceability and availability;
  • Authenticity and citability;
  • Appropriate long-term preservation;
  • Compliance with ethical standards and security protocols regarding future data reuse.

The DMP is designed as a living document that spans the full data lifecycle. Effective planning at the onset saves both time and resources later on. The plan should be drafted at the start of the research and updated iteratively throughout the project—particularly when changes occur in data types or management protocols.

A well-structured plan allows researchers to define from the very beginning:

  • Data types and file formats;
  • Metadata standards and descriptors;
  • Security measures and protocols for personal data protection;
  • Long-term preservation strategies and licensing for reuse.

To support the academic community, the University has adopted the Guidelines for drafting Data Management Plans (Rev. 1 - February 2026). This document provides the institutional framework ensuring data is managed according to the FAIR principles.

Unifi provides specific resources to facilitate the drafting process, including the DMP Online. This web-based tool guides users through pre-defined templates and generates machine-actionable documents, facilitating information exchange between systems and monitoring over time.

Crucially, the platform hosts the Unifi template (fully compatible with the Horizon Europe model), which includes operational instructions and practical tips tailored to each section of the plan.

It is advisable to archive data in archives or in institutional data repositories; where these exist, in disciplinary-specific repositories used by the various scientific communities, or in multidisciplinary repositories such as Zenodo which is managed by CERN, Dryad or Figshare.

Most are free of charge up to a certain size of dataset, and through databases such as re3data.org and OpenDOAR it is easy to find the most suitable repository for your data.

However, it will be necessary to verify that the selected repository meets certain requirements, in particular the following:

  • must have public governance
  • must guarantee the long-term storage of data
  • must support open licenses, such as Creative Commons
  • must be compliant with standard metadata requirements of international aggregators such as OpenAIRE
  • must assign a persistent identifier to data sets (DOI, Handle, URN)
  • must allow cross-linking with scientific publications
  • must manage the deposit of updated versions of the same data set linked together (versioning)

Along with the data, you should submit documentation and instructions (read-me files) for the tools and software used to generate and process the data.

The following descriptive metadata must be deposited with the data:

  • author(s) and contributor(s)
  • title
  • date of publication
  • abstract
  • references to funding, if any
  • citation of publications to which they refer, if any
  • distribution license
  • level of access
  • any embargo period.

Depositing software, for example on GitHub, and protocols,for example on Protocols.io would also be good practice.

Software must use the appropriate license, such as GNU or MIT licenses.

Other licenses are available on the Open Source Initiative website.