Your DMP should describe briefly the research methodology for your project - i.e., how you will collect data for your project and what the goals of such collection will be. Will there be secondary data from a previous or existing project that you'll use? Be sure to integrate any additional information that will help reviewers understand clearly your project and the techniques you'll be implementing.
Be sure to note the types of data you'll be collecting - e.g., specimen, observational, experimental, simulation, derived, etc. The DMP should also state what formats your data will be in - will they be text files, numerical data, modeling data, software code? Use, where possible, open-source (i.e., non-proprietary) formats - or, at the very least, formats in heavy use by your research community. For example, many researchers use Excel to keep track of data they're collecting; though a proprietary format, it's almost ubiquitous usage is providing assurances that it will subsist for some time.
In addition, the DMP should describe the tools or software you'll rely upon to make sense of the data and perform analyses on them.
Finally, in writing about your research methodology and data collection practices in a DMP, try to estimate how fast (or slowly) your data will grow. Where will the data be kept? How much storage is it possible to anticipate to cover the expected rate of data growth? How often will you need to access the data you're collecting? If there are data from an existing project, then it might prove a relevant exercise to revisit data from a previous project and see if the rate of growth could be tracked over a certain period of time.