Metadata Management Roadmap: Four Key Considerations

Metadata Management Roadmap

This is part of Solutions Review’s Premium Content Series, a collection of contributed columns written by industry experts in maturing software categories. In this submission, ZL Technologies Co-Founder and CEO Kon Leon offers a brief metadata management roadmap template via several key considerations.

SR Premium ContentTechnologist David Weinberger once wrote, “Metadata liberates us, liberates knowledge.” Indeed, for those seeking to discover just how much they don’t know, metadata is the place to start. And in the enterprise context, where data grows at a breakneck pace, including mountainous volumes of hard-to-wrangle unstructured data, metadata may just be a source of salvation.

Enterprise Data Today

Let’s first examine the state of affairs at today’s enterprise. Spurred by the emergence of remote work, employees create and share more data now than ever, typically accumulating across email, cloud-based file sharing, collaboration platforms such as Teams and Slack, and video conferencing platforms such as Zoom. In fact, the workforce is said to spend as much as 80 percent of its working hours communicating, and today much of that is digital. As a result, when pieced together, this workforce-created data represents the “human side” of the enterprise.

As of yet, companies have done a very good job of ignoring this data—and in some ways, rightfully so. The unwieldy volume, paired with its inherent lack of a defined structure, makes it especially challenging to scrub, process, and extract insights. This is where metadata enters the story. By processing unstructured data at the enterprise-wide level and generating key metadata, patterns begin to emerge that reflect the activity, intent, and dynamics of the workforce, and we start to see a much more vivid picture of the company’s human element.

The Human Element

There is in fact very little we cannot infer about the company’s human players by looking at, what some have called, “digital footprints.” For instance, early use cases include the ability to identify the company’s “go-to” people, since they tend to exhibit a definitive set of digital communication patterns that clearly identify them as a focal point during organizational network analysis. Identifying employees that may be likely to quit based on their digital behavior is a similar proposition, characterized by increased latency in email replies over time. Or, companies going through a post-merger integration can use message metadata to identify any bottlenecks or “dark spots” that may pose an obstacle to a successful integration.

The use cases delivered by harnessing unstructured data extend across various disciplines, including people analytics, HR, security, and operations. Some use cases require metadata management software alone, while others become much more powerful when paired with the capability to search and process content. Sentiment analysis, for example, combines the two by transforming the content of messages into metadata about the messages. This becomes a powerful indicator for companies seeking to understand how their employees feel.

The Role of Governance

Needless to say, metadata should not alone be enough to determine important management decisions. Analytics should always be layered with “human touch,” contextual awareness, and most crucially, oversight. Now more than ever, access to information must be heavily guarded by a governance layer, dictating who can see what. Information governance—the process of harnessing and managing information while ensuring that it is only used for its required functions—must be applied synergistically with analytics. In fact, they can be seen as two sides of the same coin, illuminating the human aspect of the company while also protecting it. This duality is exemplified by new privacy regulations such as GDPR, which paradoxically require complete intrusion into employee data; in essence, in order to control personal data, you must first find it.

The Path Ahead

Given the rate of advancement in analytics, and the unprecedented access to data and metadata that enterprises now possess, we’re tantalizingly close to realizing the full meaning of Weinberger’s original thought. However, even with the intention of enlightenment, the pursuit of knowledge can also bind us. As we continue down this path, another prescient observation must be resurfaced, this time about the perilous cliffs on either side: “absolute power corrupts absolutely.”

At this stage in our technological evolution, the path toward more knowledge is not one from which we can easily depart. However, with careful planning, governance, and humanity, we can shine a light on the path ahead when the night grows dark.

Kon Leong
Follow Kon