by John Edwards

7 tips for modernizing data management

Jan 25, 2021
AnalyticsData ManagementMaster Data Management

Data is an enterprise's lifeblood. Don't let poor management drain away your organization's most precious resource.

A user reviews data and statistical models. [analytics / analysis / tracking / monitoring / logging]
Credit: Laurence Dutton / Getty Images

Data is an enterprise’s most valuable and enduring asset, serving as the foundation for both digital strategy and transformation. Yet maintaining a strong grip on rapidly spiraling amounts of data scattered across public and private clouds, as well as in on-premises environments, requires a new and innovative management approach.

Modernizing data management to keep pace with growing application and security demands isn’t only important, it’s essential. Managing data safely and effectively requires creating a strategy and reliable methods to access, integrate, cleanse, govern, store, and prepare data. The following seven tips can help make this challenging process faster and easier. Read on.

1. Update your existing data management strategy and architecture

Begin modernizing by developing a solid understanding of the organization’s business strategies, data needs, and data analytics objectives, suggests Yan Huang, assistant professor of business technologies at Carnegie Mellon University’s Tepper School of Business.

“Then design a data management architecture that can integrate current data management tools and systems, leverage state-of-the-art models and methods, achieve the organization’s current objective, and adapt to its future needs,” she says.

Huang notes that a strong architecture will allow data management modernization to be approached in a systematic and integrated manner, avoiding both compatibility issues and data silos.

“The process of redesigning the data management architecture requires carefully evaluating the organization’s data analytics objectives, and identifying areas of improvement and new opportunities,” she explains. “A well-designed modern data management architecture ensures that the organization’s data management systems will work effectively and efficiently, can constantly deliver value to the organization, and are flexible enough to incorporate enhancements and new capabilities.”

2. Inventory and map all data assets

Before moving forward, return to the basics. “If you can’t nail down where your data assets are and what’s protecting them, you won’t be able to answer whether the access granted is appropriately limited or wide open to the internet,” warns Mike Lloyd, CTO at cybersecurity technology developer RedSeal.

Peter Mottram, managing director and leader of consulting firm Protiviti’s enterprise data and analytics practice, concurs. “Understand what comes into the firm, what you create, and what you send out — this is the foundation,” he says. “Then layer into the equation where you want to be and how modern data management technologies can simplify the organization’s data/analytics operating model.”

Mottram notes that data management has become maddeningly complicated over the past several years. “Breaking down [management] into the core building blocks and simplifying the problem statement is the best way to start meaningful transformation,” he states.

A modern data management strategy should also include a hybrid cloud strategy. “Having a clear inventory across the environments is a great start,” Mottram advises. “Then, deploying principles and controls can help firms get a handle on its data across environments.”

3. Aim for data democratization

Just a few years ago, enterprises had a single, overriding reason for modernizing their data management ecosystems: to manage rapidly growing volumes of data.

“Today, the new brass ring is the ability to ‘democratize’ data — getting the right data to the right people at the right time,” notes Luc Ducrocq, vice president of the data management practice at business and technology advisory firm Capgemini North America.

Democratizing data gives enterprises the ability to deploy self-service analytics, empower large data engineering and data science teams, create data exchanges and collaboration zones with trading partners, and pursue other mature data management initiatives.

“By democratizing data, organizations can also achieve true data trust,” Ducrocq says. “This affords them greater freedom to focus on business value and transformative outcomes.” Trust is another important attribute. “Those [enterprises] without trusted data face a continuous struggle to find and deliver the right data to business customers,” he warns.

A governance strategy should be developed and deployed to ensure that data remains current and accurate. “The democratized data needs to be identified, catalogued, standardized, and classified in order to manage the data that consumers use across the organization,” Ducrocq says.

Strong governance also allows enterprises to reduce data prep time, giving data scientists and other power users the ability to focus their time on analysis. “If organizations don’t take the time to finally fix their data, they won’t succeed in the new world of clouds and ever-increasing reliance on data,” he adds.

4. Invest in data modernization technologies

Continue investing in cloud computing and data management technologies. The most successful data modernization projects run in lockstep with these tools, says Frank Farrall, AI ecosystems leader at professional services firm Deloitte. “CIOs should start with the low hanging fruit — the legacy technology that’s on premises and out of capacity; the aging decision support system that will be out of contract in the next 12 to 24 months,” he notes.

Investing in master data management and governance technologies and processes is an ideal way to maintain holistic control over data, Farrall observes. “Knowing where data is sourced from, key definitions, and how it moves through different systems is table stakes in a heterogenous environment,” he says. “Strong ownership of data processes and elements, with key leadership support, is often overlooked in data management programs, but [it’s] a key enabler to managing a complex environment.”

5. Tap into the management benefits created by data unification

Organizations with a hybrid/multi-cloud strategy may wish to consider exploring and investing in management platforms that can unify data.

“This can include combining enterprise data with third-party data sources available via SaaS providers, data subscription providers, and others in the ecosystem,” says Faisal Alam, emerging technology leader at technology and business advisory firm EY Americas.

Virtualization is an option that’s been available for some time, yet has only recently matured to the point where it can perform at scale with minimal latency. “All the major cloud providers have federated query capabilities (a subset of data virtualization) that allow cross-cloud and hybrid cloud data querying and unification,” Alam explains.

Data unification approaches are constantly being refined, and federated querying is just one option. Other choices, such as data lakes, feature stores, and modern data warehouses, also exist, Alam notes.

6. Designate data accountability

Clarifying data accountability is a fundamental step in reimagining data governance, states Steve Bates, principal, CIO advisory, at business and technology advisory firm KPMG.

“Successful organizations move beyond policy and process and put the responsibility of certain insights and quality measures in the hands of senior leadership,” he explains. Bates suggests clarifying the precise data responsibility roles of senior executives, including the CIO, CTO, and chief data officer (CDO).

In today’s rapidly evolving business world, virtually everything is digital and connected. “Every piece of data collected on transactions, customers, and internal processes becomes an asset that can be mined to improve the product or customer experience,” Bates notes.

The critical issue facing IT leaders is that while digital points are proliferating, many remained chained to monolithic legacy systems. “A more holistic look at modern solution development and delivery that includes agile, DevOps, cloud, and many other approaches is required,” he advises. “Modern delivery is a model that can help an organization more rapidly deliver value, reduce failed deployments, and create a culture of continuous improvement and customer centricity while helping the business win in the market.”

7. Stay on top of evolving data management methods and practices

Data management is one of the most fluid and demanding issues IT leaders will face in 2021 and beyond.

“Data management strategies and their effectiveness should be constantly monitored, and a systematic review should be performed at least every year to identify concerns and opportunities for improvement,” Huang says.

She believes that an enterprise’s data management approach should match its overall business strategy. “Therefore, it’s also a good idea to review the data management strategy when the business strategy is being reviewed.”