These five key food groups are found across all successful data sharing initiatives. They form the basis of a varied and balanced data sharing diet. The quantity (or emphasis on) each of these varies according to individual data sharing initiatives, as do the recipes needed to create them.
Each of the five sections of this cookbook contains “recipes,” i.e., tested methods and tools that can be replicated by others.
Data sharing in the development sector is a relatively new domain, and there are some clear gaps in knowledge and experience. For this reason, the cookbook points out where recipes or information on ingredients are missing. Highlighting these gaps helps aspiring chefs to understand what still needs to be done and paves the way for more thinking and experimentation.
Factors influencing choice and quantity of ingredients
There is no one-size-fits-all approach to effective multiparty data sharing. Instead, this cookbook offers suggestions for key ingredients and recipes to inspire creative approaches.
The importance (or quantity) of each food group, as well as the recipes to use, depends on two broad categories of factors related to:
- The characteristics of the specific data sharing initiative, such as stakeholders or sectors involved, types of data (i.e., personal, nonpersonal) being used, the stage the initiative is in, the objectives of the partnership, the number of stakeholders involved, and the openness of the data or initiative.
- The context in which the data sharing initiative operates, including, for instance, the regulatory and policy environment, other stakeholders within a particular data ecosystem, and other contextual factors such as whether data sharing happens during an emergency situation (e.g., a pandemic or natural disaster) or during “business as usual.”
What are we making?
Success in data sharing depends on establishing and maintaining shared value for all partners. It means that all partners benefit from data sharing to some extent and no partner benefits disproportionately more than others.
That said, value distribution is not set in stone and can change over time, either because the needs and expectations of data partners change or because the data sharing initiative evolves in terms of focus, activities, and level of input needed from the different partners. Just as adjusting the value distribution might be required, it is also important for initiatives to be clear about their initial approach and be able to monitor whether the promised benefits materialize for the various partners and to what extent. Doing so allows them to correct imbalances and change value propositions as needs arise.
Like other aspects of data sharing, there is no one-size-fits-all approach to creating and distributing value creation. Recipes to ensure that benefits and value are distributed as fairly as possible can take many forms, often customized to the needs of specific partners or to differing contexts and ecosystems. Three, in particular, emerged from the landscape analysis in preparation for this cookbook: lowering costs, innovation and the provision of data products, and the delivery of tailored services.
Understanding value in data sharing*
Value in economic terms is generally used to refer to “added value,” meaning the difference between inputs and outputs for a certain product. In the context of multiparty data sharing, this often translates into a focus on financial benefits that are relatively easy to quantify. At the macroeconomic level, the Organization for Economic Development and Cooperation (OECD), for instance, suggests that improved public and private “data access and sharing can help generate social and economic benefits worth between 1 and 2.5 percent of Gross Domestic Product - GDP (in few studies up to 4 percent of GDP)”. At the microeconomic level, a well known study by Deloitte on the impacts of Transport of London’s (TfL) practice of openly sharing non-personal data showed that companies using TfL data generated a gross value added between GBP 12 million and GBP 15 million per year, including directly supporting around 500 jobs.
Many benefits from sharing data are not so easily measured. For the public sector, diverse and numerous social benefits may be attained through data sharing partnerships. The OECD highlights in particular positive impacts on transparency and accountability and increased user empowerment as associated with greater data sharing. For companies, reputational or knowledge benefits, as discussed above, may also serve as motivation to engage in data partnerships. The OECD also points at the opportunity for private sector data providers to crowdsource new insights and exploit user-driven innovation linked to the emergence of a community that creates additional value that an organization on its own would not be able to create.
*more research, experimentation, and/or knowledge exchange is needed.