Getting the most value out of your data [Q&A]

The past few years have seen a boom in digital transformation as enterprises have sought to modernize their operations to take advantage of a new, more flexible world of work.

But this change has often led to unwieldy data structures that are difficult to manage, which means that extracting value from the data is harder than it should be.

So, what can businesses do to ensure that they're getting good value from their data? We talked to Wilko Visser, CEO of enterprise architecture platform ValueBlue to find out.

BN: What has led to so many organizations allowing their data to get out of hand?

WV: Data is intangible, and capturing of data is often not centralized: Each and every person in each and every organization requires data for their day-to-day job: Whether you're a salesperson looking for customer data in your CSM or a support employee looking for the last update on a ticket: Data is everywhere. The decentralized nature of data makes it very hard to control: Without central insight into all the data that is captured, and a mechanism to avoid duplicates or redundancy, and a mechanism to check for data quality, it’s getting out of hand quickly. And without a central system of record, you'll often only detect this when it's too late.

BN: Why is it important to have a 'single source of truth'?

WV: A single source of truth will allow you to track where data originates and will designate places of origination as the truth. This prevents redundancy and incoherence of data, thus improving the quality and traceability of that quality. This is extremely important because data-driven decision-making is only as good as the quality of the data it’s based on. Making data-driven decisions without a single source of truth will, because of that, often result in poor decisions. On top of that, a single source of truth that’s available always and everywhere also speeds up your decision-making process. And those organizations that can make the right decisions fast will thrive.

BN: What's the first step in understanding an organization's data landscape?

WV: Understanding all data in your organization is hard, or even say, impossible. There’s going to be more of it than you could ever imagine, hidden in all sorts of places. So the first step in understanding your organization’s data landscape is scoping and prioritizing. And for that, you need to understand what data is critical for supporting your most important business processes. Once you understand which data is critical for your most important business processes, you can capture where it originates (Master Data Management) and how it flows through your organization (data lineage). From there, you can work on improving the quality, security, and consistency of that same data, putting in place management and governance to keep that quality at the required level.

BN: What factors influence the quality of data?

WV: Data often gets corrupted when it has multiple sources of record: this creates inconsistencies. Getting your master data management and data lineage right can help with this. Next, the manual transfer of data is always dangerous: human errors are one of the biggest risks for data quality. Automation of integrations is therefore not only much more efficient, but it also improves data quality. Last but not least, putting in place a good central system of record and a good data management and governance system will act as a safety net to detect loss of data quality early when it occurs, allowing organizations to pinpoint and solve the cause of it.

BN: How important is the choice of location for data storage?

WV: The location of data storage is actually not very important as long as it is clear what data are stored where. It might be that some data that your organization manages has specific requirements for its storage, such as geographical barriers, needs for redundancy, or specific security measures. However, understanding which of these requirements apply to which data is more challenging than choosing the right location for that data once that information is known.

Image Credit: aslysun / Shutterstock