Should you publish your data or shouldn't you? And if so, which part of it? Sometimes this question is rather straightforward to answer. For example, because your research funder requires it. Or because you strongly believe it's not up to you to withhold any latent knowledge hidden in your data from future researchers.
Even so, not all data are created equal and data publishing does involve an investment of time and money. Some datasets have a more obvious reuse potential than others. By following the arguments below (Tjalsma and Rombouts, 2011) you can see for yourself whether (part of) your dataset is an obvious candidate for data publication.
If your data aren't 'good to go' yet but do have reuse potential, don't worry. Most of the time it's not too late to document your data and other issues like gaining consent for sharing might also be tackled retrospectively (see 'Informed Consent').