How to get started?

Based on your previous work on identifying data sources, you ideally have a good idea on who to build a data partnership with and what data you are looking for. The next question is: what type of partnership would you like to build?  

There are different data sharing partnerships that come with different features across public sector entities or between public bodies and private sector, and so on. While there are many other types of data sharing agreements, this article focuses on the following most commonly use frameworks:

• Individual data-sharing agreement: This type of partnership involves a direct agreement between two parties to share specific data. It is suitable when there is a mutual benefit in sharing data and both parties have the necessary resources and capabilities.
• Join existing data partnerships: Joining established data partnerships allows organizations to leverage existing data ecosystems and infrastructure. This option is advantageous when seeking to access diverse datasets or when lacking the resources to build an independent partnership.
• Data collaboratives: These are often used when private sector data is combined to help inform public sector decisions. Data is shared strictly between partners, with an independent third party who manages access.
• Data commons: A data commons is a shared pool of data resources accessible to a defined community of users. It provides an environment for collaborative data analysis, experimentation and innovation. Data is pooled and shared as a common resource, addressing power imbalances. There are some examples like OpenStreetMap and DECODE, a consortium of European partners who work to make communal data available to innovators and civic groups.

There are also partnerships where governments facilitate the data exchange, while not accessing or sharing data themselves. While that is less relevant for the purpose of policymaking, we are sharing one example that the UK established the Open Banking Limited to steward standards and data infrastructure for the open banking initiative. Although the government itself may not share data, it guides and supports businesses and organizations in sharing data effectively. This kind of partnership is led by policymakers but does not involve direct government data sharing.

Peer learning networks are another form of partnership, where organizations working on the same topic come together to learn from each other's experiences and insights. These emergent partnerships enable policymakers to play a role in establishing connections and facilitating knowledge exchange in the wider data landscape.

Some data sharing agreement examples can be given among different sectors. One example is an agreement on pooling data about spawning and fisheries in Belize which is signed by a number of international NGOs, academic institutions and government offices. Another example is an agreement signed by the Utah State Board of Education and Utahns Against Hunger for sharing school-level data about student breakfast access and participation.

How to set up a data sharing agreement?

Establishing a data sharing agreement is a key foundation in formalizing a data partnership. To ensure a successful agreement, one of the critical steps is to be very clear about the data you need. Moreover, your partner needs to understand for what purpose you would like to access the data. This includes identifying the stakeholders to include in the process and to share data access with.

You further need to communicate high-level timeline with the data provider. It is also important to address security concerns, especially with regards to data storage (see also this article of Data to Policy Navigator).

Prior to these steps above, you may realize that actors may not be willing to share data without a clear incentive or business model. Sometimes, the sustainability of data sharing is questioned, as recurrent sharing requires additional effort. Advocacy and aligning incentives are needed for establishing agreements and promoting data sharing to create win-win solutions that motivate actors to exchange data. This can be achieved through discussions that address timelines and purposes, where the focus lies on highlighting the incentives and benefits for all stakeholders.

Another key point is to ensure consent and data privacy (see this article of Data to Policy Navigator). It is important to prioritize consent and data privacy to protect the rights of data providers and individuals involved. This includes ensuring compliance with relevant regulations and guidelines, and clearly defining how data privacy will be maintained throughout the partnership.

Sample agreements can be used for drafting your data sharing agreement.  Moreover, a data sharing agreement often benefits from including the following important terms: 

• the grant of the licence/permissions to use the data for the intended purpose 
• restrictions to the permitted use of the data, such as territorial or time limitations, exclusivity or commercialisation rights 
• warranties or other assurances provided in relation to the data provider’s rights in the data 
• allocation of liability for contract breaches and other liabilities between the parties
• confidentiality 
• term/duration of the agreement 

In addition to these points, when setting up a data sharing agreement, it is often advisable to work with the lawyers in your organization. Moreover, it may be useful to make data sharing agreement available to all users of the data in the ministry so that they can assess and manage their compliance.

How to build a trusted (data) partnership?

Building a trusted data partnership requires specific practices that foster collaboration and synergy. Some good practices to consider can be given as follows:

How do I know I have built a successful data partnership?

The following checklist may provide you with some guidance on understanding when you have established a successful data partnership

‍

What's next?

After identifying all possible data sources, collecting and accessing data, and building data partnerships, you are well on your way to implementing data-based policies. There is still work to be done in understanding, analysing, and visualising the data before policy can be crafted. Next, learn how to prepare and process the data.