Data Ecosystems: enabling competitive advantage

Many organizations are aware of data ecosystems’ business potential but lack the knowledge to realize optimal financial benefits. With insights based on real-world experience, we show how businesses outperform thanks to the full engagement in data sharing.

In the last decades, both data providers and data consumers have increasingly noticed the enormous impact that data can have on all aspects of life, reaching from the private life of individuals to business innovations and politics. This has been caused by the rise of new technologies, ensuring a revolution in the way that data is collected, structured, and utilized. As a consequence, the general view on data has transformed from being a necessity to keep track of things, to being a valuable and tradable asset which can be priced and sold. This new economic status quo of data, combined with the potential of data-sharing, has led to a new concept that has been evolving in recent years: the data ecosystem.

A description of data ecosystems

Despite the nonexistence of a common definition, it can be stated that there is a core idea of data ecosystems as platforms or socio-technical networks between organizations in which partners collaborate to capture, aggregate, consume, exchange, reuse and enrich information from multiple sources in order to produce value and foster innovation. This collaborative concept distinguishes itself from classical siloed data systems whereby collections of data are held by a single actor without an information sharing mechanism. A central architectural characteristic is the platform’s transaction involving client and broker actors. Data aggregated by the broker is consumed by clients who subsequently compensate the broker for the used data with monetary value or non-monetary incentive such as reduced costs, risks, or improved customer insights for instance. From a conceptual perspective, data ecosystems effectuate the exploitation of diverse data and its combinations that should lead to enhanced customer experience, more effective marketing, and operational decisions.[1][2][3][4]

Four types of data ecosystems can be identified

In its latest report on data ecosystems – ‘Data Sharing Masters’ – the Capgemini Research Institute distinguished and analyzed four broad types of data ecosystems:

1. Data brokerage and aggregation ecosystems

In this relatively simple structured data ecosystem, data is collected directly from end-users with the help of mobile apps, web applications, or smart connected devices (i.e., health trackers, home devices, TVs, watches). Subsequently, the data is aggregated and/or processed by the broker actor and often commoditized before client actors derive insights and value from it.

Example: Thinknum, a US-based data-aggregation platform, captures data from government sources, social media, and company websites to help clients such as retailers to select store locations based on potential profitability.

Figure 1. Data brokerage and aggregation ecosystems

2. Reciprocal data sharing

A data sharing network can be assigned to this category if various organizations join forces with the objective of boosting efficiency. In this reciprocal system, the data transactions result in a more intense collaboration between the organizations. Though one of the partners manifest dominant traits, smaller players are still able to obtain value from information sharing.

Example: A parts manufacturer provides logistics information to the original equipment manufacturer (OEM). In return, the OEM shares results of quality testing with the parts manufacturer.

Figure 2. Reciprocal data sharing

3. Federated analytics ecosystem

This model presents a solution to run analytics on data which cannot be shared directly between actors for a variety of reasons (i.e., regulations, data volumes). The solution includes a central platform from which ‘decentralized data’ (stored in various places in different formats) can be accessed and analyzed without the data having to be moved to a central environment. Not having to move the data, but keeping it decentralized, means a reduced risk of the data being hacked or leaked, and avoiding costs of moving large quantities of data.

Example: NVIDIA’s Clara is an open framework enabling hospitals and healthcare- and life science developers to securely collaborate, train and contribute to a global AI model which can be used for automatic diagnoses of patients. Since AI requires massive amounts of data in order to build robust AI algorithms that provide for accurate diagnosis, Clara offers a platform where local data can be securely shared and combined.

Figure 3. Federated analytics ecosystem

4. Collaborative data-supply chain

In this model, the collaboration between multiple actors itself becomes the deliverable, creating value for a single client or market. As such, it is the most collaborative of the four data ecosystem models.

Example: The Open Carbon Footprint is an initiative where multiple actors in a value chain are brought together enabling them to determine and track the environmental footprint of end products more accurately.

Figure 4. Collaborative data-supply chain

Data ecosystems generate new revenues, save costs, and increase productivity

Organizations engaging in data ecosystems are building a sustained competitive advantage – not only for themselves but also for their partners – over late movers and less involved peers. The latest report of the Capgemini Research Institute concludes that data sharing ecosystems have the potential to deliver substantial financial benefits. Engagement in such networks can lead to an annual revenue increase of up to +9% in the next five years. Moreover, businesses with data ecosystem engagements demonstrate improved customer satisfaction by +15%; improved productivity/efficiency by +14%; and reduced costs by -11% annually. Research also showed that organizations which extensively use external data generate superior financial performance such as higher fixed asset turnover (up to 14x) and higher market capitalization (2x). In general, the more organizations engage in data sharing, the more they profit financially. Collaborative leaders with extensive data sharing yield an additional ten percentage points of financial advantage. Although organizations involved in extensive data sharing significantly outperform the majority of organizations (61%) which primarily exhibit low level collaboration, only 14% engage in complex data ecosystems.[5] Therefore, it can be concluded that in spite of the commitment to data ecosystems, a vast amount of financial potential remains unused in most organizations worldwide.

Many thanks to the co-authors Jarno Baumgartner and Marvin Pfister for their contribution to this blog post.

[1] Capgemini Research Institute. (2021). Data Sharing Masters: How smart organizations use data ecosystems to gain an unbeatable competitive edge.

[2] Oliveira, M. I. S., Lima, G. D. F. B., & Lóscio, B. F. (2019). Investigations into Data Ecosystems: a systematic mapping study. Knowledge and Information Systems, 61(2), 589-630.

[3] Oliveira, M. I. S., & Lóscio, B. F. (2018). What is a data ecosystem? In Proceedings of the 19th Annual International Conference on Digital Government Research: Governance in the Data Age (pp. 1-9).

[4] Rantanen, M. M., Hyrynsalmi, S., & Hyrynsalmi, S. M. (2019). Towards ethical data ecosystems: A literature study. In 2019 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC), (pp. 1-9). IEEE.

[5] Capgemini Research Institute. (2021). Data Sharing Masters: How smart organizations use data ecosystems to gain an unbeatable competitive edge.