Your dataset is a potential treasure chest of data and you should think of your metadata and documentation as the key. You should be able to provide the dataset(s) and the metadata in your DMP to any of your peers or colleagues and they should ideally have everything they need to access and reuse your data. Your documentation should include...
Where to find the data
During the research lifecycle you might not want to (or be able to) share the dataset widely, but in most cases there will be at least a few teammates that will need to know where the data is. After finishing, you may want to share the data with others (and receive the appropriate citation), in these cases Data DOIs (Digital Object Identifiers) can help.
How to interpret the data
What metadata schema are employed? What controlled vocabularies are used? Be explicit about what abbreviations are used, list them if necessary. What experimental assumptions were made, and what methodologies were used? If machines needed to be recalibrated, how often? In all of these try to adhere to disciplinary standards, use the vocabularies, notations etc that are accepted by your peers, but do still be explicit about them in your metadata (don't assume a peer or newly appointed teammate would "just know").
The Research Data Alliance maintains a Directory of Metadata Standards to help here.
The DCC List of Metadata Standards to help here.
Set out the reuse terms for reuse
Consider what you will want others to be able to do with your data once your finished with it. The guiding principle here is that data should be 'as open as possible and as closed as necessary'.
Where to find the key!
One for when you are finished and ready to archive your dataset. If your documentation and metadata are the key to unlocking the treasure chest then make it obvious to access. Even a simple top level readme.txt file with basic instructions to get started with the documentation can be invaluable to your peers.
