Data ecosystems can help innovate superior value propositions and tackle the world’s most pressing issues, according to Dan Klein, chief of AI and data at Zühlke. In a blog for the consulting firm’s website, the digital expert recently explored how sharing data can help deliver value to companies and their customers.
A combination of enterprise infrastructure and applications, a data ecosystem can be used by a group of organisations to utilise to aggregate and analyse information. Sharing in the data equally, the organisations involved can use ecosystems either for mutual commercial interests, or to do social good.
According to Dan Klein, such ecosystems differ from typical business-to-business models when it comes to data use, because their emphasis “is on open collaboration”, instead of selling datasets for the sole benefit of one entity. This means that to implement them effectively, they require an ethos of decentralisation, and transparency, in a way that also means they may be a good deal more socially palatable than traditional data-harvesting activities.
Of course, in a world where dog-eat-dog business practices are still the norm, some business leaders may not see the logic in an agreement which might favour a competitor. But according to Klein, data ecosystems have a unique set of benefits worth considering.
“When they work effectively, data ecosystems allow organisations to face new challenges,” the Zühlke chief of AI and data explains. “Issues that need many hands, like the Covid-19 pandemic, for example, are easier to solve when all the necessary data sources are open and readily available. Previously unsolvable problems can also be tackled, as when you bring disparate, siloed datasets together, you unlock insights that can help uncover solutions that were previously impossible to spot. And firms can also use ecosystems to make better-informed decisions, as an open data ecosystem helps eliminate glaring knowledge gaps.”
Even the term ‘dog-eat-dog’ is a limiting generalisation, which overlooks the fact that there are many species in the natural world, which have actually evolved to thrive in partnership with one-another. And in that same vein, “a data ecosystem lets businesses grow and evolve in an environment shaped by shared information.”
Klein asserts to that end, that data ecosystems “enable problem-solving in the same way you’d solve a jigsaw puzzle where the pieces are distributed among a group”. If firms continue to take an all-or-nothing approach to data, and refuse to share all parts of the puzzle, each participant would end up with “just a fraction of the full image”. In this case, ecosystems allow companies to unearth trends and patterns they couldn’t see without complementary datasets.
This insight can then be deployed to create collaborative solutions, which bridge the gaps flagged up by the data. To explore this point further, Klein points to a recent example of Zühlke’s work, with an electric vehicle (EV) infrastructure investor application. The client developed a proof of concept after pooling data from transport, energy, and geographical sources. The app used those sources to “create a map of high-density EV traffic, which can be cross-examined against things like weather, time of year, the direction of traffic, and even ferry timetables”, and to highlight where future EV charging stations are needed most urgently.
With examples like this showing how data ecosystems can underpin innovation, the concept is catching on quickly. Data ecosystems are already becoming a growing trend, as companies in every sector look to address major challenges. For example, 2021 research from Statista found 81% of telecoms, 73% of banking, and 60% of consumer goods businesses were planning to launch new data-led innovation ecosystem initiatives. Meanwhile, BCG Henderson Institute figures suggest more than half of the planet’s biggest companies now have active data ecosystem models.
Enabling data ecosystems
As clear as the business case might be for launching data ecosystem projects, though, doing so is easier said than done. After all, Klein notes that “the growth of the internet has helped create a host of thriving industries focused on selling data”. Data ecosystems – and the idea of sharing data for free – therefore require a mindset shift. Either away from viewing data as a material to barter over, or from the conclusion that “this is my data, I don’t give it to anybody”.
Klein expands, “Solving those issues requires businesses to put egos aside and refrain from any instinct to become the ‘leader’ in what should be a democratic space. Real value comes not from being sold a single stream of information, but from the gold nuggets that you find inside multiple crossmatched datasets. So you only need one break away, and suddenly the ecosystem doesn’t work.”
Beyond this, Zühlke has identified five other ways to get ready for data ecosystems. While doing so can take time, these best-practices should help to reach the desired outcome, according to Klein.
“First,” he notes, “communicate clearly. Define and communicate the rules of engagement between organisations in the ecosystem. Your aim is for transparency, and to encourage data collaboration within the context of any competition or anti-trust regulations. Second, run in-house data due diligence. Data needs to flow, and be available where and when it’s needed – not stored and static. Timeliness is crucial, and having data available in real-time means that it can influence decisions and actions, rather than just report on them long after the fact.”
On the third point, he says that ecosystems should see participants “strongly encouraged” to work on a ‘presumed open’ basis for data sharing. By doing so, data is shared by default and restricted only on an exception-by-exception basis – rather than vice versa. This is essential for accelerating collaboration. Similarly, his fourth point is that firms should also think like “a team player” – looking for opportunities to benefit fellow organisations by allowing non-sensitive data to flow back to them. Allow participants to see both the whole picture and the ‘working out’ so you can build a shared understanding and benefit from the peer review of observations and decisions.
Finally, he adds, “Formalise things. Create shared tools and techniques for working with data across the ecosystem – with curated datasets, feeds, APIs, methods, and algorithms. Then define the governance for who is trusted – and the scope of that trust.”
Risks of failing to act
There are a number of risks which firms could be exposing themselves to, if they do not look to adopt the benefits of data ecosystems now. Amid the quickly shifting economic environment, futureproofing their organisations will become more difficult, for one.
“Too often, data is something that primarily reports on the past,” Klein explains. “When you need to drive change or manage risk across a complex ecosystem, data needs to inform and direct action – so it needs to be up to date. That’s more easily done when the data you’re accessing is available at all times, rather than bought or sold as chunks that represent a set period of time.”
He added, “Data sharing between diverse players is essential for solving the biggest issues of our time and creating pioneering solutions. But complex data silos, poor data quality, and regulatory red tape can make this an impossible task.”
But if all this seems daunting, then there are professional services firms which offer external expertise on the matter, and can help companies with the cultural shifts needed to make the most of data ecosystems. Pointing to Zühlke’s own experience, Klein concluded that the firm’s “ISO-accredited strategists, scientists, and engineers” can help clients create new value at scale with the right data strategy and AI solutions.