In today’s digitally powered world, data is increasingly being generated about us and the world around us. The benefits of data can be immense, as we’ve seen its potential to improve trade efficiency, build digital economies, and broaden access to information for the public good. Movements such as open data, responsible data, and data free flows with trust all share an intention to unlock data for broad value.
For the promise of data to be meaningfully realized, it is essential we understand some of the different approaches and models that enable data to be accessed – and effectively shared – in the first place.
So, let’s get started.
Models for data sharing can be understood by the: type of data shared, governance and rules around data sharing, technologies used to store and access data, and finally, business models that sustain and support the infrastructure. There are countless ways that these different considerations can come together. As such, each permutation can offer different benefits and challenges, decisions and trade-offs, and ultimately, outcomes for people.
In this article, we outline four different approaches to exchanging data, primarily distinguished by their governance structure.
Integrated national data exchange systems.
Integrated national data exchange systems allow governments to share administrative data – such as educational records, tax filings, and records of birth and death – across ministries, departments, and agencies. The architecture of national data exchange systems varies from country to country. Many employ a mix of newer technologies along with older, legacy systems. However, a number of countries have moved to establish a layer of technology that communicates with these disparate systems and technologies simultaneously, presenting a unified interface upon which data exchange is made easier. With such an ‘e-Service’ layer, rapid development of new services can be achieved.
National governments are not the only potential beneficiaries of integrated national data exchange systems. Private sector actors, academics, and civil society organizations can all benefit from accessing administrative data sets and adding value by combining this data with additional data sets to create personalized information, for example through citizen portals.
Promise. When designed for broad use, integrated national data exchange systems intend to provide a range of benefits, including improving efficiency in government operations, rooting out corruption, and strengthening planning. In the era of artificial intelligence (AI), integration of government data sets can pave the way for training data to develop locally relevant AI models and algorithms. Beyond benefits to the public sector, integrated national data exchange systems can also provide citizens with an array of benefits. These include fostering transparency in data flows within government and improving the cost, time, and experience of accessing government services. For the private sector, integrated national data exchange systems can smooth the way for authenticating people to access financial services, healthcare, and more.
Considerations. There are several challenges to overcome, such as bureaucratic red tape, connecting different (sometimes legacy) software systems and technologies, and costly and time-consuming procurement processes and vendor agreements. Additionally, much administrative data is personal information about residents and citizens of a country. Security and privacy are therefore crucial in ensuring these systems do not leave people vulnerable to threats of government overstepping, disenfranchisement, surveillance, or abuse of power. In addressing these risks, many governments have found that optimizing citizen transparency leads to improved adoption and trust. One way to promote this transparency is by requiring citizens’ meaningful consent anytime data is shared with non-governmental actors.
Example. With its UGhub platform, Uganda’s National Information Technology Authority (NITA-U) offers one relevant example of an integrated national data exchange system – along with many insights and learnings. Originally, UGhub was established as a means of improving intragovernmental information-sharing, as departments sought to mitigate the inefficiencies and bureaucratic hurdles of connecting multiple, disparate systems. Today, the platform has seen an uptake among government departments, agencies, and private sector partners. To date, it has facilitated over 100 million transactions. As with any system of this size and scope, challenges include sustainable financing, ongoing support and maintenance, and adoption. UGhub is taking steps to promote trust, by engaging in participatory processes and facilitating change management sessions for potential new users.
Data Trusts.
Data trusts, as defined by ODI, are a legal structure that manages data for a clear purpose and with clear rules for data sharing. They are independent structures with trustees who govern the rules of data access and use. In some cases, data trusts seek a revenue model, thus improving financial sustainability over the long term. Data trusts can be used for many different types of data – for example, climate information, language, and even very sensitive data like genomic information that requires carefully managed access.
Promise. A main feature of data trusts is the fiduciary responsibility that the trust takes on to manage data for the benefit of the stated purpose. This legally binding responsibility requires the trust to operate with transparency and undivided loyalty. In the case of a Club Goods model that generates member fees, the data trust has an added benefit of strengthened financial durability.
Considerations. One key challenge that data institutions face when deciding whether to operate as a data trust is finding the right jurisdictional home. While England, where data trusts first gained popularity, has a long history of trust law, the same is not true in many other jurisdictions. Thus, for data institutions based in regions without such legal structures already in place, operating under a trust model may be unfeasible. Another challenge is in finding an approach to sustainable funding. Like other data institutions, data trusts require staff, technology, and other operational costs that need financial support.
Example. PLACE is one example of an organization that has implemented a data trust structure to responsibly manage its data. As a non-profit organization, PLACE collects high-resolution mapping data for cities in Africa and small Pacific Island states, which may otherwise be unavailable due to financial and resource constraints. To ensure its data remains accurate and high-quality, the PLACE trust operates under a legal stewardship framework, which allows governments to own mapping data, while still ensuring it’s accessible for the public good. To access the data, relevant trust members, such as non-profit organizations, academics, and researchers, pay a subscription fee, which helps PLACE remain financially sustainable.
Personal data stores.
Personal data stores (or wallets) allow people to collect, store, use, and manage their information themselves. In many ways, personal data stores provide individuals with a means of taking back control over their personal information, be it administrative data such as certificates and medical records, or commercial data like mobility, social network, and fitness information. This is a change from today’s reality in which individuals have little control over – or even knowledge about – the data they generate online. While personal data stores differ in who creates them – whether nonprofits, commercial firms, or research organizations – they all have a similar goal: to make it easier for consumers to aggregate a variety of data.
Promise. The main benefits of these models, at least in theory, are that they promote greater personal agency and security within the exchange of data by requiring meaningful informed consent. When thoughtfully designed, personal data stores can empower people to take control over their own information – by managing how, when, and by whom it is accessed. These design features not only promote personal choice, but also foster trust and transparency.
Personal data stores allow people to benefit from their own information in several ways. Not only do they allow people to better understand their habits and histories, but these systems also set the stage to shift the data economy from a model where companies and governments have outsized control to one where people do too.
Considerations. Because personal data stores put control back into the hands of individuals, if people do not sufficiently understand how, why, or by whom their information is being used, they could be vulnerable to breaches of privacy, stolen information, or security concerns. Personal data stores do not have aligned technical standards – meaning their privacy conditions, operating systems, terms of service, etc. can differ. This can make them difficult for individuals to navigate, which is why digital and data literacy are crucial to the success of these models. Furthermore, shifting control to people can burden them to the point of making that control useless (not unlike consenting to terms and conditions today). Efforts to create personal privacy representatives as in the case of India attempt to reduce the burden for individuals.
Example. Digi.me provides one example of the ways in which data stores can be used to consolidate personal information and put control back into the hands of the individual. This platform allows users to download, aggregate, and share their medical records digitally. This way, individuals have access to important health information quickly and seamlessly when needed. Digi.me is also secure, requiring user consent before sending any information. And, by using encryption, that platform ensures that data is protected and only accessible to outside parties when granted direct access to a specific record.
Decentralized networks.
Blockchains, decentralized distributed databases, ledgers, and peer-to-peer data-sharing protocols are both a technological means of sharing data and an approach to governance. As opposed to centralized networks, decentralized share in the processing and storage of the data across all the nodes in the network, meaning that there is no single authority in control. Perhaps, the more powerful vision of these networks is the hope to foster competition and disrupt market incumbents by enabling data portability and interoperability outside of proprietary platforms.
Promise. Decentralized data sharing networks promise several benefits, such as greater transparency, security, and provenance. They also ensure no single point of failure. Additionally, they can help reduce the friction and costs associated with data transactions, opening markets that would otherwise be captured by proprietary platforms because of data economies’ network effect.
Considerations. Despite these opportunities, there are also challenges to consider. Because decentralized protocols are still in the early stages, they face barriers to viability and sustainability – especially considering the pace at which digital technology is evolving. As such, these systems will need to establish a means through which to stay both relevant and financially stable. Additionally, the success of decentralized protocols could be undermined by resistance to change, as some users may feel comfortable with the status quo despite the potential for a better system. Finally, architectural drawbacks of decentralized networks – if not thoughtfully designed and implemented – could include issues with quality of service, recourse, and trust.
Example. The Beckn Protocol, a decentralized peer-to-peer network for digital commerce, allows businesses across sectors and borders to be found on different platforms. Food delivery is one example. Today, there are a number of different apps, from UberEATS to Doordash to Grubhub, and if a restaurant wants to offer delivery, it will have to subscribe to one – or often – several of these platforms. Otherwise, consumers may be unaware of the services offered. This same conundrum is present across the world. To help solve this problem, Beckn has created a protocol that would make it possible for every participating restaurant to be searchable and retrievable on any relevant platform. So, for example, it would not matter which app a person uses, as all participating restaurants would be available on their app of choice – and for the restaurants, they would be discoverable wherever their customers are. Of course, this approach is not limited to the restaurant industry. Beckn allows for transactions across sectors, from food service to healthcare to agriculture.
These models – along with many others – can help promote an open and inclusive data ecosystem.
Data is critical, and it will only continue to increase in value as people, governments, and organizations across the world seek to harness its power. While this article provides an overview of several common models, it is not exhaustive. There are a number of other mechanisms, such as data spaces, for exchanging data – each with their own benefits and challenges.
When implemented with the Principles for Digital Development in mind, especially the principle of Establish people-first data practices, the different approaches and models that enable data to be accessed and effectively shared can contribute to a digital future where individuals and communities meaningfully benefit from unlocking data.
At the Digital Impact Alliance, we’re focused on promoting trusted, effective data exchange – with research spanning data models, country-specific case studies, and user experience. You can learn more about our work here.
