Increase Data Liquidity by Building Digital Data Assets

Future-ready companies monetize their data assets by improving processes, wrapping products with analytics features and experiences, and selling innovative information solutions.[foot]Companies can generate economic returns from their data via improving, wrapping, and selling as described in B. H. Wixom, “Data Monetization: Generating Financial Returns from Data and Analytics—Summary of Survey Findings,” MIT Sloan CISR Working Paper No. 437, April 2019, https://cisr.mit.edu/publication/MIT_CISRwp437_DataMonetizationSurveyReport_Wixom.[/foot] MIT CISR research[foot]To date, MIT CISR data liquidity research includes interviews with 85 executives in 2020 regarding their company’s strategic data initiatives and the nature of the associated data assets; a Q2 2021 MIT CISR Data Board discussion regarding data liquidity measurement; and theoretical work regarding digital data assets, as described in Gabriele Piccoli, Joaquin Rodriguez, and Varun Grover, "Strategic Initiatives and Digital Resources: Construct Definition and Future Research Directions," Proceedings of the International Conference on Information Systems (ICIS), December 14, 2020, https://aisel.aisnet.org/icis2020/digital_innovation/digital_innovation/5/. [/foot] has found that data liquidity—the ease of data asset reuse and recombination—accelerates the pursuit of new data monetization opportunities.

Data liquidity varies along a continuum. One data asset may be more or less liquid than another, with highly liquid data being more prepared for planned and unplanned new value creation opportunities. Google Maps is a familiar service based on a highly liquid data asset. The data asset represents map information that can be reused and recombined for innumerable value creation initiatives (e.g., route optimization) by numerous firms (e.g., rideshare company Uber).[foot]The authors identified the Google Maps API as a DDA by applying their theoretical work, which conceptualizes DDAs, to the characteristics and structure of the Maps API.[/foot]

Data often is illiquid because it is trapped in local business processes, closed platforms, or obsolete data structures; it requires movement, consolidation, or translation; or it is constrained by regulation, organizational boundaries, or technical limitations. Incumbent banks, healthcare organizations, and companies with long-established business models are often mired in illiquid data assets.

Increasing the liquidity of a data asset can be a costly, labor-intensive endeavor. Yet having liquid data is critical for organizations that hope to respond quickly to novel value creation opportunities. In a previous briefing,[foot]B. H. Wixom and G. Piccoli, “Build Data Liquidity to Accelerate Data Monetization,” MIT Sloan CISR Research Briefing, Vol. XXI, No. 5, May 2021, https://cisr.mit.edu/publication/2021_0501_DataLiquidity_WixomPiccoli.[/foot] we explained that companies can increase data liquidity by investing in the company’s data monetization capabilities. The theoretical work of the first two authors of this briefing identified a second approach: building digital data assets, or DDAs.

The Structure of Digital Data Assets

A DDA is a digital resource[foot]Piccoli, Rodriguez, and Grover, “Strategic Initiatives and Digital Resources: Construct Definition and Future Research Directions.”[/foot] structured to eliminate friction in asset-level reuse and recombination. DDAs result from the purposeful design and deployment of data assets with three structural elements: value, modularity, and a programmatic interface. These attributes produce extremely high liquidity and offer companies an unprecedented degree of reuse and recombination potential.

Value: DDAs possess significant value creation potential when they are important for successfully executing data monetization use cases. Google Maps is based on a DDA that exposes map data, which thousands of companies use to carry out improving, wrapping, and selling use cases. Google captures some of the value that these companies create by selling ads served alongside mapping results or charging a fee each time the DDA is accessed. In 2019 one analyst estimated that Google Maps revenue would exceed $11 billion in 2023.[foot]Dennis Schaal, “Google Maps Poised to Be an $11 Billion Business in 4 Years,” Skift, August, 30, 2019, https://skift.com/2019/08/30/google-maps-poised-to-be-an-11-billion-business-in-4-years/.[/foot] Rideshare company Uber alone paid Alphabet about $58 million during 2016–2018 to enable mapping services reliant on the Google Maps DDA in Uber’s mobile applications.[foot]Uber Technologies, Inc., initial public offering (filed May 13, 2019), p.261, https://www.sec.gov/Archives/edgar/data/1543151/000119312519144716/d647752d424b4.htm.[/foot]

Modularity: Data asset owners abstract DDAs into modules to enable companies’ reuse and recombination of the data. Use of modular data assets requires minimal information about the assets’ inner workings; any third party can use and reuse the assets without specific knowledge about their native technology, data architecture, or other structural elements. DDA users follow explicit use specifications that the DDA owner provides to draw on the black box asset and thereby create value. In the case of the Google Maps DDA, companies such as Uber integrate the DDA into their initiatives by following use specifications for the Google Maps API. In 2020, the Google Maps modularity enabled five billion Uber trips.[foot]“Uber Announces Results for Fourth Quarter and Full Year 2020,” Uber Investor website, February 10, 2021, https://investor.uber.com/news-events/news/press-release-details/2021/Uber-Announces-Results-for-Fourth-Quarter-and-Full-Year-2020/default.aspx.[/foot]

Programmatic interface: DDAs are encapsulated within an interface that enables code-based data interchange between systems. The interface manages the DDA’s technical requirements and the governance rules by which DDA users must abide. Interfacing with a DDA is automated, so the ability of the DDA owner to scale to additional third parties is limited only by the technical context. The interfaces associated with the Google Maps DDA are combined into the Maps Software Development Kit (SDK), available for both iOS and Android, and the Maps JavaScript API.[foot]“Maps SDK for iOS,” https://developers.google.com/maps/documentation/ios-sdk/overview, “Maps SDK for Android,” https://developers.google.com/maps/documentation/android-sdk/overview, and “Maps JavaScript API,” https://developers.google.com/maps/documentation/javascript/overview, Google Maps Platform.[/foot] These interfaces automatically manage access to Google Maps servers, map display, billing, and service level agreement oversight, thus eliminating the need for manual involvement and coordination by Google.

Digital Data Assets at Santander: A Case of Open Banking

In 2018, two European regulatory bodies introduced banking regulations that required financial institutions to share customer-permissioned data about individual and business bank accounts with licensed third-party providers (TPPs). In effect, the European Commission’s Revised Payment Services Directive, referred to as PSD2,[foot]The PSD2 is the second, revised version of the original Payment Services Directive (PSD); the “2” in PSD2 refers to the second version.[/foot] and the Open Banking initiative from the UK‘s Competition and Markets Authority (CMA) encouraged banks to create digital data assets by requiring that banks share their data assets via open, secure, and standardized interfaces.

To comply with the regulations, most incumbent banks had to consolidate multiple systems of record, resolve data and technical conflicts, and abstract DDA users from the internal complexity of highly integrated legacy systems.

Initially, TPPs struggled to access and use the shared data assets because they were not yet offered as DDAs. Swedish fintech Tink, for example, reported the need to resolve unexpected interdependencies with banks’ systems by manually coordinating with the banks “through lengthy email threads, conference calls, and WhatsApp text exchanges.”[foot]“Could the poor readiness of APIs put the success of PSD2 in jeopardy?” Tink blog, July 4, 2019, https://tink.com/blog/open-banking/status-of-psd2-production-apis/.[/foot] When first exposed, banks’ account data was neither modular nor fully encapsulated in a programmatic interface, leaving innovators like Tink to “make a connection, call the bank, and wait for their approval.”[foot]“The blood, sweat and laughs shared on our way to the PSD2 deadline,” Tink blog, August 14, 2019, https://tink.com/blog/open-banking/tink-integration-efforts/.[/foot]

Over time, however, banking institutions increasingly succeeded in structuring customer account and transaction data as DDAs. For example, Santander Group is a Spanish financial group founded in 1857 and headquartered in Madrid. The company serves more than 140 million corporate and private customers worldwide.[foot]Banco Santander, 2020 Annual Report, February 22, 2021, https://www.santander.com/content/dam/santander-com/en/documentos/informe-anual/2020/ia-2020-annual-report-en.pdf.[/foot] Santander UK, the British subsidiary of the group, made its customer account and transaction data available through a digital data asset product called Account and Transactions V3[foot]“Account and Transactions V3,” Open Banking Account Information, Santander Developers, https://developer.santander.co.uk/sanuk/external/account-and-transactions-v3.[/foot] (see figure 1).

Value: Account and Transactions V3 creates value by providing transaction-level and account-level information for use by TPPs, which include startups and direct competitors. For example, the startup Coconut’s core product is a bookkeeping app that accesses Santander UK’s customer account data to help the self-employed track income, claim expenses, and calculate taxes.[foot]Coconut Platform Ltd, https://www.getcoconut.com.[/foot] Barclays, a British banking competitor to Santander UK, leverages the Account and Transaction V3 product to offer its own customers access to accounts held with other financial institutions via the Barclays mobile app.[foot]“Manage accounts with other banks in your Barclays app,” Ways to bank, Barclays, https://www.barclays.co.uk/ways-to-bank/account-aggregation/.[/foot] Moreover, like TPPs, Santander UK’s own initiatives can access the DDA to create value via reuse and recombination.

Modularity: TPPs can integrate Account and Transactions V3 into their initiatives without specific knowledge of the technology used by Santander UK or the internal structure of the account and transaction data being exchanged. TPPs refer to Santander UK’s published data exchange protocols to make use of the DDA,[foot]“Account and Transactions V3.”[/foot] with expectations shaped by pre-specified service level agreements.

Programmatic interface: TPPs access Santander UK’s Account and Transactions V3 product without requiring physical or manual involvement from the bank’s engineers, managers, or operational staff. Rather, Account and Transactions V3 exposes data programmatically via an API that adheres to the Open Banking API specifications. The interface details technical specifications (e.g., REST endpoints) and governance rules (e.g., GDPR compliance) that TPPs must follow when using and reusing the DDA.

Data Liquidity Built via Digital Data Assets

Historically, companies have increased their data liquidity by investing in enterprise data monetization capabilities. Today, technological advancements such as cloud computing, multicluster technology, API gateways, and shared data architectures enable liquid data at the data asset level. At the core of such technologies is the ability to scale storage, compute, and service layers both independently and interdependently, as opposed to traditional data management technologies that have fixed compute and storage resources. Such breakthroughs are central enablers of data liquidity.

Our theoretical work and early observations suggest that digital data assets unlock unprecedented levels of data liquidity. Before engaging in DDA creation, however, companies need to be clear about how they will monetize the data—and how they intend to capture value from associated monetization activities. One option is to expose DDAs for internal use before unleashing them externally. For example, before exposing DDAs for external consumption, Banco Bilbao Vizcaya Argentaria, S.A. (BBVA) experimented years ahead of the introduction of PSD2 with configuring data assets for internal consumption. This approach has the dual advantages of ensuring that a data asset can be manipulated without requiring manual or ad hoc coordination and that the company can capture some of the value the data asset creates. Configuring a DDA initially for internal use facilitates the identification and resolution of existing interdependencies between the data asset, other data assets, and business processes. And a DDA used internally permits full visibility of the data asset’s reuse and recombination potential by way of comprehensive performance metrics that monitor internal initiatives.

© 2021 MIT Sloan Center for Information Systems Research, Rodriguez, Piccoli, and Wixom. MIT CISR Research Briefings are published monthly to update the center's patrons and sponsors on current research projects.