What To Consider When Building an Enterprise Data Strategy

What To Consider When Building an Enterprise Data Strategy

An enterprise data strategy (EDS) is a road map that an organization uses to determine how data will be collected, organized, and processed based on business priorities, the size and industry of the organization, and the level of data maturity.

Contrary to popular belief, an enterprise data strategy is not limited to large businesses with large amounts of data. Indeed, small businesses can benefit from investing in a data strategy early on to lay the groundwork for future growth.

Benefits of an Enterprise Data Strategy

The common pitfall that many organizations encounter is that, while they collect a lot of data, each team interprets it differently. There is no standard reporting method, and each team may report a value for the same metric differently.

As a result, everyone ends up with inconsistent data and no clear idea of what is accurate. Without a single source of truth, it becomes exceedingly difficult to trust your data and derive actionable insights.

“Data does not exist in a vacuum,” Kossowski explained, a data strategy analyst. “The marketing team will not simply use marketing-specific data over which no other team has any control. They’re going to want to gather data from a variety of sources as well.”

“Thus, an element of governance, standardization, and a common language is critical to ensuring that those teams can communicate with one another,” she continues.

Thus, by implementing an EDS, you can avoid information silos, foster data trust, and facilitate decision-making.

What To Consider When Building an Enterprise Data Strategy

1. Your Current Data Maturity Level

Kossowski recommends conducting a self-assessment prior to developing your strategy.

Consider the following: Where does your organization stand in terms of data maturity?

Dell has a widely used “Data Maturity Model” that assists businesses in determining their level of data maturity. The process is divided into four stages:

Data aware – Your organization’s reporting system is not standardized, and there is no integration between its systems, data sources, and databases. Additionally, there is a lack of confidence in the data itself.

Data proficient – There is still a lack of trust in data, particularly in its quality. You may have invested in a data warehouse but are still missing some components.

Data savvy – Your business is empowered to make business decisions based on the data it collects. However, some kinks remain between business leaders and IT as IT strives to provide reliable data on demand.

Data-driven – IT and business departments collaborate closely and are on the same page. The emphasis is now on scaling the data strategy, as the foundational work (particularly integrating data sources) has been completed successfully.

What’s critical here is to be realistic about your company’s position.

“I believe the biggest pitfall I see is not being completely candid with yourself about your company’s data maturity stage,” Kossowski explained.

She adds that it is insufficient to examine your feelings about how data-driven you believe your company is. Consider the facts.

Begin by identifying the data issues that your business is currently facing, as this is an excellent indicator of where you stand.

2. Your Industry and Company Size

Your industry and company size will dictate whether you take a centralized or distributed approach to data strategy.

However, before we dissect these approaches, let us discuss two data strategy frameworks: offensive and defensive.

While data defense places a premium on data security, access, governance, and accuracy, data offense places a premium on gaining insights that enable decision making.

Each business requires a balance of offense and defense. However, some industries lean more toward one end of the spectrum than the other.

A healthcare organization or financial institution, for example, almost certainly deals with highly sensitive data and places a premium on data privacy and security.

Obtaining real-time data and insights is unlikely to be a priority, but enforcing access controls is almost certainly. As a result, they will gravitate toward a defense-oriented framework.

On the other hand, there are technology companies, an industry that moves quickly and places a premium on quick turnaround of data insights.

As a result, they emphasize offense. With that said, certain departments within technology companies (and other fast-moving industries), such as finance, will undoubtedly prioritize defense.

Now, let us return to centralized and decentralized strategies.

The framework you use will dictate which strategy is most advantageous for your business.

In a centralized structure, the data and reports are managed and prepared by a centralized reporting or business intelligence (BI) team.

“That [structure] works much better in a smaller organization, and especially in an organization that prioritizes defense,” Kossowski explained. “You will be the bottleneck, but you will also have complete control over every component.”

On the other hand, a distributed model works better for larger teams that take an offensive approach. This way, each team can work more efficiently and is empowered to complete tasks in the manner that works best for them.

According to Kossowski, in this model, BI is simply responsible for the platforms and setting the guardrails, while the teams perform the development work.

“If you think about an organization, as it grows larger and has a more centralized team, scaling becomes increasingly difficult,” she explained. “You end up having to hire an increasing number of people to accomplish that.”

“So I believe that once a company reaches a certain size, it will naturally gravitate toward a decentralized [strategy].”

Thus, once you’ve determined which framework is most appropriate for your industry and size, you can implement the necessary strategy.

3. Your Data Management Team

According to Kossowski, data science is a hot topic in data management right now. And she is correct.

Harvard Business Review named it the sexiest job of the twenty-first century in 2012. Nearly a decade later, Glassdoor ranks it as the second best job in the United States.

However, if you’re considering adding a data scientist to your data management team, this should not be your first choice.

Kossowski emphasizes that data science is only as good as the data that fuels it. And if the data is unreliable, you will not gain valuable insights.

“Data science is not a wand that magically transforms unusable data into actionable insights. Regardless, you’ll require that data foundation “she continues. “So, I believe that jumping into something just because it’s the next big thing is a major concern.”

If your organization is still in the early stages of the data maturity model, Kossowski has some advice on where to direct your efforts.

“A data warehouse architect or even a data analyst who is comfortable writing SQL and constructing SQL tables,” she explains. “If you’re only hiring one person and don’t have a lot of data, this can be a very powerful hire because one person can accomplish a lot on a smaller scale. They can wear a variety of hats and acquire a variety of skills.”

When it comes to more technical tasks, such as data ingest into the warehouse, third-party tools can assist you.

Leave a Reply