A data management plan or DMP is a document that outlines how research data will be handled during and after a research project. Creating a DMP is considered good practice for any research project that will use or generate data. Decisions made early on affect what you can do later, so good and timely planning can save you a lot of time and problems in the longer run. It also helps you consider the necessary resources and costs for data management - so you can budget for these and even include them in your grant applications. Planning can also help you to increase the FAIRness of your data and to make data FAIR by design. Data management plans are increasingly required by research institutions and funding agencies. These stakeholders recognize the intrinsic value of data and intend to increase scientific integrity and transparency and the accessibility and reuse of data collected by publicly funded research.

So what should be included in a data management plan?

If you are writing a DMP for your institution or funder it is important to check if a specific template is available or even required. Although DMP templates might slightly vary from organization to organization they usually cover the following topics: a summary or description of the data. You will need to indicate whether you will generate new data or use existing data sets and provide some basic characteristics about them - such as their origin, their type, their format and estimated size.

Metadata and documentation: what information will be provided to make data understandable and findable? And will this be provided following agreed standards?

Other DMP questions will deal with data storage and security: where will the data be stored? And how will you ensure that data is protected against risks? For example data loss due to human errors, computer failures or malicious attacks. And how to make sure that it cannot be accessed, deleted or modified by unauthorized persons? What about preservation? What data will be preserved after the end of the project? For how long and where?

Sharing and Reuse: what data can and will be shared? With whom and under which access and reuse conditions or licenses? Where and when will data be made available?

Another content block involves the possible legal or ethical issues related to your project's data. For example if you work with personal or otherwise confidential data, or any other situation where ethical approval is needed. Or if you work with third party data or you are planning to seek patent protection for your research results. These are all issues that will affect how you should collect, store and process your data and whether you can share them or not and under which conditions.

Responsibility should also be addressed: who bears the overall responsibility for implementing and regularly updating the dmp? Who is responsible for each of the different data management activities including documenting, storage and backups, preserving and sharing data?

And finally what about costs and resources? Are there any costs related to the implementation of the dmp? Will you need additional resources to cover these costs?

So that's for the content of a DMP but when are you supposed to submit a DMP? Well, it depends on the specific requirements of your institution or research funder. Usually, the life cycle of a data management plan looks like this: the proposal stage: this is where the various requirements diverge most. Your institution or funder might not have any specific requirements with regards to data management. Another possibility is that the proposal needs to address some RDM questions or include a light version of a DMP. In other cases, a preliminary but full version of the DMP might be required. When your project is approved. A first version of the DMP will usually be required shortly after - often no later than six months after the start date - but things can change over the course of the project. Or maybe certain things were not known at the beginning - that is why you are supposed to keep your DMP up to date throughout the course of the project. In some cases updated versions might be a deliverable with a specific deadline. At the end of the project it is likely that a final version of the DMP is requested - this should reflect the final decisions and approaches taken. For some funders the final DMP might be part of the project's final evaluation.

As we've mentioned, many institutions or research funders will provide a DMP template with a series of RDM related questions. However, there is a range of tools that can help you with the process of creating a data management plan. These tools will incorporate these templates. They will also provide you with some tips as you go along and allow you to export your final DMP in a format ready for submission. So before you start - check whether your institution provides a DMP tool. For example researchers from Ghent university and other belgian institutions can use dmponline.be. In our website you can find more information about data management plans and a video tutorial about dmponline.be. Why don't you have a look?