SND’s recommendations for Zenodo communities

An organization-specific community on Zenodo allows you to collect and showcase datasets and other research outputs produced by the organization’s researchers. This provides better oversight and control, as well as an opportunity to ensure the quality of metadata.

This guide presents SND’s recommendations for Zenodo communities and metadata harvesting from Zenodo to Researchdata.se. Similar features can be found in services such as Figshare and Dataverse.

About Zenodo communities

Setting up an organization-specific community has several advantages. Primarily, it provides an overview of the datasets that the organization’s researchers share on Zenodo. Other types of research output can also be included, for example, reports, presentations, and conference papers.

By collecting research outputs in a community, the organization can review and quality-assure the records before they are published by or included in the community. This ensures that the data and metadata meet the organization’s guidelines and achieve an agreed level of documentation quality.

A community can also be used to collect data from specific research projects, institutions, conferences, or domains. Each community defines its own guidelines and selection criteria.

For more information about creating and managing communities, see Zenodo’s website.

Make sure to inform the organization’s researchers that they can add their datasets and other research output to the organization’s community. Use the communication channels that are normally used for information about research data management.

Managing communities – general information

Only the creator of a record, or a person with permission to edit it, can request that it should be included into a community. It is not possible to “invite” someone else’s published record.

Users can submit a published record for inclusion in a community using the Submit to community function. The community’s curators will then review and accept or decline the record.
For unpublished records, the creator uses the Submit for review function to send a draft to the community for review and publication. A curator will review and, if necessary, edit the record’s metadata before publication, or decline the submission.

Both the creator of a record and the community’s curators (or anyone with editing permissions) can edit the metadata of a record.

When a record is included in a community, this is visible in its metadata. The record is then automatically included when metadata are harvested from the community, for example, to Researchdata.se.

Zenodo users can be invited to become members of a community. Members are given different roles with varying permissions, for example viewing restricted records. However, it is rarely necessary to invite individual researchers as members.

SND’s recommendations for communities

  • Describe the purpose and scope of the community
    Give a clear description of what type of material will be included and that the community is intended for data whose research principal is the organization.
    Example: The KTH Royal Institute of Technology community on Zenodo
  • Publish a curation policy
    The policy should state the requirements and recommendations that apply to data and metadata in the community.
    Example: The KTH Royal Institute of Technology curation policy on Zenodo
  • Review data and metadata
    Staff from the research data support function should verify that the data and metadata in the community are of high quality, comply with the community’s guidelines, and meet the FAIR principles.​​​

Harvesting metadata to Researchdata.se

Datasets published in organization- or domain-specific communities on Zenodo can be made findable in Researchdata.se.

By harvesting metadata, information about datasets is retrieved from Zenodo to the research data catalogue at Researchdata.se. They also become visible in Sweden’s National Data Portal, Sveriges dataportal, which is updated daily with metadata from Researchdata.se.

Recommendations for metadata and harvesting

To ensure that metadata maintain high quality and can be harvested to Researchdata.se, the following applies:

  • Only records of the resource type dataset are included in metadata harvesting to Researchdata.se. Metadata for, for example, publications, presentations, and code are not harvested.
  • Metadata should include a clear free-text description of the dataset, which allows secondary researchers to understand its content, purpose, and methodology..
  • Metadata should include at least one subject term that corresponds to a concept in the Swedish Standard Classification of Research Subjects 2025 (SSIF 2025). It is recommended to add an SSIF 2025 term to the keywords when including a record in a community.

Tips for improving metadata quality

  • Specify the research principal organization in the Publisher field. This must be changed manually, as Zenodo is set as the default. Use the organization’s English name.
  • Use ORCID for individuals and ROR for organizations in the metadata.
  • Include the organizational affiliation of researchers listed in the metadata.
  • Note that Zenodo defaults to the CC BY licence. This may need to be changed in accordance with the recommendations from the Swedish Agency for Digital Government (Digg) and the Swedish Research Council regarding licences on research data. 
    Digg’s recommendation on open licences and intellectual property rights (in Swedish)
    The Swedish Research Council's recommendations for research data: FAIR and open access
  • The default file accessibility setting (Visibility) is Public. If access is restricted, there should be clear information about how to request access to the data files, and whether the restriction is temporary (due to embargo).
  • Community managers can access restricted files, but only if the owner allows this when submitting the material to the community. This means that if the owner of a record leaves their position, the organization might lose access to the dataset.
  • In the Creators and Contributors fields, you can specify roles to clarify the functions of different contributors. Note that the roles under Creator are not included in DataCite’s metadata schema and therefore not exported to Researchdata.se.
  • Specify the subject in the Keywords and subjects field using terms from Zenodo’s controlled vocabularies (e.g., EuroSciVoc or MeSH). When metadata are harvested to Researchdata.se, these subject terms are automatically supplemented with the corresponding SSIF 2025 terms.
  • Add references to related materials in Related works. Include a PID (e.g., DOI), if possible, and specify the relation type, for example, Is cited by, Is derived from, or Continues.
  • Share documentation files, such as README files or variable lists, together with the data. You can use SND’s README file template.
  • Metadata in English increase international discoverability and visibility. If unsure about language choice, select English and indicate the language in the Languages field.

Additional tips and recommendations

Well-described data are easier to find, understand, and reuse.
A short and clear title, along with an informative description, are essential parts of the data description in a record. It is also valuable to include relevant keywords, time periods, and information about funding and licence.

As an example, see how Runverket at the Swedish National Heritage Board describes and shares its datasets on Zenodo: Runverkets community på Zenodo.

Remember that the more thoroughly data are described, the easier they will be to find, understand, and reuse correctly and efficiently. A concise and informative title and a good summary description are key elements of the data description. Other useful information includes keywords, time periods, funding, and licence details.

Runverket share well-described data on Zenodo and can be viewed for inspiration: https://zenodo.org/communities/runverket

What to consider when sharing and archiving data.

  • Plan how and where the data will be archived. For further information, see SND’s guidance on long-term preservation and FAIR data management over time on Researchdata.se.
  • Data shared via Zenodo are stored on CERN’s servers in Switzerland.
  • Sensitive or confidential data should not be published on Zenodo. Although files can be published with restricted access, data are stored unencrypted and may be accessible to Zenodo staff, even if they are not publicly accessible.