Why

Data sovereignty

Data spaces

International standards

We

Become a member

Members

Donate

Board

Head Office

IDSA ambassadors

Contact

Make

Working groups

Task forces

Network

Open source

Projects

Communities

Offers

Reference Architecture

Dataspace Protocol

IDSA Rulebook

Certification

IDS Reference Testbed

Data Connector Report

Adopt

Data Spaces Radar

Implementation partners

Professional qualifications

Training catalog

Knowledge Base

Publications

Most important documents

Papers

Magazine

Legacy

Events

Upcoming events

Calendar

Archive

News

Blog

Newsroom

Infohub

Newsletter

March 16, 2020

Data sharing in industrial ecosystems (part 4 of 4)

Many industrial data ecosystems are still in their nascent stages and many players are only taking their first steps in this area, often in the form of bilateral data sharing. However, the early successes that do exist provide a basis for identifying steps to increase the chances that an industrial ecosystem will prosper. As more companies participate in data ecosystems, this list will surely evolve:
Prof. Dr. Boris Otto, Prof. Dr. Nikolaus Mohr, Matthias Roggendorf, PhD and Tobias Guggenberger

Success ingredients of successful data sharing models

  1. Systematically assess business opportunities from sharing your own data and gaining access to new data: In a first step, it is important for organizations to understand an ecosystem’s potential in terms the business it will generate rather than the data the company will share. Such an approach requires looking broadly at both internal applications and opportunities that can be co-created with customers, or even new business they can build outside the core. In some cases, additional value may come from selling data alone, while in others it may involve some give and take. Companies often fail to articulate their expectations clearly and underestimate the time needed to achieve benefits. Also, it is crucial to understand the main actors in the potential business ecosystem – for example, who will need to play the orchestrator and who are participants – and what mechanisms would incentivize them to share their data (for example, if value creation is unevenly distributed). 
  2. Define your own role and options: After understanding the business opportunity offered by the ecosystem and its operating protocol, companies need to thoroughly assess their strategic options. These might involve becoming a data owner, a participant in an emerging ecosystem, or a provider of services. The choice of this role largely determines the required investments, the organizational setup, and the resources needed to drive implementation from both the business and technology side. 
  3. Create a value- and risk-driven approach to data: A systematic scan of corporate and transactional data can reveal the value that can be created and the intellectual property that might be at stake. Options should be considered for sanitizing or aggregating the data so it can still be used with partners. Methods for managing data value and risk should be put in place for data assets assessed as strategic. 
  4. Establish sustainable data governance: Once relevant data has been identified, scaling up the approach requires suitable data governance. This involves establishing business ownership of internal data so that it can be shared externally and integrated quickly with data from new sources. Companies may also want to set up a set of data management tools, such as a data catalog (or internal data marketplace), as well as a basic tools for topics such as tracking data lineage, classifying data risk, or documenting data access policies. Data partnership managers may be needed to drive date identification and commercialization. 
  5. Embark on a data platform strategy: Often, a company’s legacy IT systems are highly complex, and its data is siloed in operational applications. Inflexible systems and data models determined by software vendors make the data production side of the equation difficult to change. The data consumption side, in turn, is easier to evolve. The key to gaining traction is a flexible data consumption architecture that enables easy blending of data, real-time data processing, and the use of analytics and AI. In many cases, the main work entails making existing data accessible to such an analytics environment rather creating the environment itself, as cloud-based analytics solutions are fast to set up and use.
  6. Apply a test-and-learn approach before scaling: Companies should not be afraid to learn by doing. Set an aggressive goal of taking a first minimal viable product live with internal or external customers within a few weeks. Start with use cases of limited complexity that contribute to overall strategy. Find a friendly partner or customers who is excited to learn with you. Communicate your successes and grow your capabilities along the way.

The market for data sharing platforms that guarantee data sovereignty is still nascent. Furthermore, commercial approaches are difficult to establish (except in niches) due to the need for a neutral supplying entity (a prerequisite that may interfere with commercial interests).

Players that engage early can shape the market and build highly innovative business models and products or services that competitors cannot easily copy. In some cases, such solutions become add-ons to the existing business; in others, the superiority of the new solutions could fully disrupt the industry.

A key requirement for the proliferation of industrial ecosystems is the existence of standards for data interoperability, portability and sovereignty. Promising standardization initiatives such as International Data Spaces (IDS) and Trusted Cloud – as well as the recently started GAIA-X project – are steps into the right direction.

Author: Prof. Dr. Boris Otto, Prof. Dr. Nikolaus Mohr, Matthias Roggendorf, PhD and Tobias Guggenberger

Stay updated with us