In every organization is a hidden data economy. The supply side comes from apps, devices, and sensors creating new observations and adding them to an ever-growing stock of data capital. The demand side comes primarily from analytics and AIs chewing on those datasets to improve actions and decisions across the firm.
But most companies don’t think of data this way. Firms tend to think of data in terms of what it takes to capture, store, and process datasets. Very few think in terms of data supply, demand, and transaction costs between the two.
Because companies tend not to see data this way, their internal data economies are hiding in plain sight. And as a result of being hidden, most of these data economies are probably under-performing.
How does an organization bring its hidden data economy into the light and encourage it to grow? It should focus on three things: 1) data liquidity, 2) data productivity, and 3) data security.
- Data liquidity
Data liquidity is the ability to efficiently get data from its point of origin to its many points of use. The essence of data liquidity is reducing the time, cost, and effort to repurpose data for new uses.
Each digital observation gets created in a particular structure that, from the app’s perspective, makes it fast and easy to capture. Think of these structures as different shapes. Some are born as a row in a table, others as a set of attribute-value pairs bundled into an object, others as nodes and edges in a graph, still others as a line of text appended to a list. But every analytic or AI, as well as other apps, need the data in a slightly different shape for its own purposes.
The challenge in increasing data liquidity is to turn your data into a shapeshifter. To allow the app to create data in the shape it needs while making it quickly and easily appear to other points of use that the data was created in a different shape that it needs for its own task. The faster, easier, and cheaper a dataset can be repurposed for a new use, the greater its liquidity.
- Data productivity
Data productivity is dollar output per data input. That is, when you apply a given dataset in a particular action or decision point, what is the incremental revenue generated or incremental cost avoided? It’s tempting to think of data productivity in terms of how much money you have to spend to generate that incremental value, but this misses the point. This comparison focuses on the hardware and software costs involved, ignoring the data coursing through the technology.
Data productivity aims to measure the value-generating capacity of the information signal contained in the data. In essence, you’re estimating the revenue generated or cost avoided by using a given dataset in a process or decision point. To do this, you’ll need either a way to do A/B testing where some instances of the process or decision get the data input, the others don’t and you compare the performance of one group against the other. Or, you’ll need a baseline and a measure of the change in performance after adding the data. In practice, this is harder than it sounds. But the basic method is straightforward.
- Data security
Data security must now provide protection for both the observer and the observed over observations about them. The datafication of nearly every activity in personal, commercial, and civic life rewrites the social contract between people, companies, and governments. As a result, keeping data secure means not just authorization, access, encryption, and auditing, but also transparency and upholding data property rights declared in a patchwork of overlapping regulations worldwide.
The next step: Data trade
A firm’s reward for increasing its data liquidity, data productivity, and data security is both an increased return on data capital and reduced legal and reputational risk from data collection and use. By focusing on proprietary data assets and using them in unique ways, firms can use the growth of their internal data economy as a source of sustainable competitive advantage.
However, even this is not the biggest prize. Increasing data liquidity, data productivity, and data security sets up a firm to take the next step—increasing data trade with built-in protections for all data stakeholders.
Global supply chains, collaborative research efforts, even distribution and channel partner relationships tie firms together in webs of interactivity. These cross-company activities create and use data just like those inside each firm. But they often struggle to get the data they need because of the costs and risks of sharing information across firms.
As companies get their own internal data economies in order, they’ll find a new ability to license, contract, and trade datasets with business partners. This will open the door for new digital goods and services these firms could not create on their own, enabling new forms of value-creation while protecting the privacy of its customers.