The concept of “data as an asset” is no longer new to companies. They have realized that data is valuable not only for future purposes (e.g., predicting customer churn) but also for daily operations (e.g., monitoring daily sales of a grocery group to prepare for the next day’s inventory).
Additionally, data has become “democratized.” With advancements in data storage, data transformation, data visualization, and other technologies, companies are eager to maximize the value of their data and become “data-driven” leaders.
However, despite this eagerness and the availability of technological enablers, many companies struggle to make the best use of their data. There are multiple reasons for this struggle, one of which is the “people behind the data.”
“Data” itself cannot run a business independently (at least not yet). Businesses are run by people, and behind the abstract data entities, structures, and models are the individuals who create, manage, and interpret them. In the process of “producing” and “maintaining” data, people make choices regarding:
- How to represent concepts: People often create different ways of representing the same concepts.
- How to define granularity and levels of detail: Individuals have preferences regarding the level of detail, which depends on the metrics that matter to them.
- How to present ideas within an organization: Different business units across an organization may have their preferred ways of presenting data.
In large organizations, these factors multiply the complexity, resulting in struggles with data management.
What Is Data Management?
Given these challenges, companies realize that without proper data management, it is very difficult to use data assets efficiently. This inefficiency may lead to missed competitive advantages and opportunities.
But what exactly is “data management”? Data management is not very different from other forms of management: it requires intention, planning, coordination, and commitment.
In the lifecycle of data assets, data management involves:
- Developing, executing, and supervising plans, policies, programs, and practices;
- Controlling, protecting, and enhancing the value of data;
- Managing information assets throughout their lifecycles.
Data management activities can be divided into two parts:
- Technical: e.g., technical deployment and performance of databases.
- Business: e.g., making consistent decisions on how to extract strategic value from data.
The goals of data management are to:
- Understand and support the organization’s information needs;
- Capture, store, protect, and ensure the integrity of data assets;
- Ensure the quality of data and information;
- Protect the privacy and confidentiality of stakeholder data;
- Prevent unauthorized or inappropriate access, manipulation, or use of data and information;
- Ensure data can be used effectively to add value to the enterprise.
- …
The list above could be much longer. While technology can play a role in achieving these goals, making the system and processes sustainable requires buy-in from people; it’s not optional.
However, securing the buy-in of relevant stakeholders is not guaranteed. Many companies struggle with data management even when well-defined and enforced rules are built into the system. If people are unwilling to follow the rules, they can always find ways—or excuses—not to comply.
Gaining people’s buy-in is part of change management. It involves helping everyone recognize the value of data and encouraging them to contribute to maximizing that value.
To achieve this, several actions should be taken:
- Everyone must understand that the impressive capabilities of generative AI are not just based on technology but also on the quality of the data assets they work with.
- It’s essential for everyone to build a basic understanding of data—how it works, its value, what doesn’t work, and what does.
- People don’t work in silos in data; instead, collaboration between IT and business across the organization is mandatory to create a successful story.