As the research community embraces data sharing, academic journals can do their bit to help. Starting this month, all research papers accepted for publication in Nature and an initial 12 other Nature titles will be required to include information on whether and how others can access the underlying data.

These statements will report the availability of the ‘minimal data set’ necessary to interpret, replicate and build on the findings reported in the paper. Where applicable, they will include details about publicly archived data sets that have been analysed or generated during the study. Where restrictions on access are in place — for example, in the case of privacy limitations or third-party control — authors will be expected to make this clear.

The new policy (full details of which are available at go.nature.com/2bf4vqn) builds on our long-standing support for data availability as a condition of publication. It also extends our support for data citation, the practice of citing data sets in reference lists in a similar way to citing papers. Authors are encouraged to cite data sets that have digital object identifiers (DOIs) assigned to them.

The introduction of data-availability statements follows a trial at five Nature journals — Nature Cell Biology, Nature Communications, Nature Geoscience, Nature Neuroscience and Nature Physics — that began in March 2016. The pilot confirmed differences in the culture of data sharing and access between different disciplines, and that the lack of obvious, public, community repositories can pose a significant barrier to public data deposition. Nevertheless, even in disciplines that are not yet so able to embrace openness and sharing, there is increasing awareness and appreciation that data deposition can enhance the visibility and reuse of published research, and that data citation can increase the recognition of those who create and share data.

This new policy will be implemented across the diverse range of Nature journals by early 2017. We expect that its implementation will shed more light on the reasons for disciplinary differences in data sharing, identify challenges and help to promote the practice more widely.

It’s not just journals. A broad drive across the research, funding and publishing communities is under way to make the availability of research data more transparent. Funders, for example, are also introducing data-availability statements. The seven UK research councils require their grant holders to include them. And the US National Institutes of Health is asking researchers to provide management plans for their research data.

We expect that offering consistent information on data availability in our papers will promote data reuse by future researchers. And where public data archiving is a mandatory requirement of journals, there is some evidence that including data-availability statements with persistent links to data in published articles is an effective approach to ensuring public data availability and policy compliance (T. H. Vines et al. FASEB J. 27, 1304–1308; 2013) .

This new policy follows the launch, in July 2016, by our publisher Springer Nature of an ambitious project to introduce and standardize research data policies across all of its journals (see go.nature.com/2by6l6x). The project sets out a defined common framework for data policy — which Nature policies align with — that enables different journals to encourage data sharing in a way that reflects the circumstances of respective specialist communities.