Analytics in action: Where do you land on the data maturity curve?

It’s a seismic shift for any organization to jump from business reporting to business intelligence. It doesn’t happen overnight, and it requires a strategic effort, investing in a new kind of data team—one that is empowered to drive company growth by accessing and sharing insights companywide. In contrast to a traditional BI team that provides only a data model, today’s data team operates as a strategic pillar of the business and recommends a strategy grounded in data. The group has a leadership voice, offering informed opinions—based on data—about what the company should be doing. If your competition is here and you are not, looking at your data maturity is a smart move that can help identify next steps to advance your data advantage.

Know the stages and where you land

Data maturity is represented by a five-stage model. Where your organization lands is dependent on how important data is to your operation, as well as your investment in analytical resources. The five stages consist of business reporting; business intelligence; ad hoc analysis and unexpected insights; integrating hybrid centralized data teams; and augmenting analytics with machine learning.

Stage 1

Consider a firm focused on business reporting; it gains insight from key factors, one data team at a time, with no collaboration from other teams. Inconsistencies in this process can cause trouble, for example, when data from one domain counters another.  The data is either different or misinterpreted across departments, teams, or applications. This organization needs to move to stage two, which focuses on bringing data comprised of the same information presented similarly into one place for consistent reporting.

Stage 2

Some type of BI is common in almost every organization, generally defined as a data model plus a visual tool using data from multiple sources for business reporting. Now it is possible to ask questions about the data—for example, tapping into a unified approach to data analytics to gain insight on customer experience or retention rates. This stage utilizes data analysts, who are essential to making correlations among the data and driving the predictive reporting that helps apply data to future strategies.

Stage 3

The third phase, ad hoc analysis and insights, lets data professionals examine organizational data, observe unexpected questions, and actively seek optimization. Questions regarding business priorities can be answered, improving scoring strategies for sales and account health. Firms move to the fourth stage by scaling up this effort.

Stage 4

Utilizing one data team is not a scalable strategy. Bottlenecks happen as the group supports multiple distinct company departments. This fourth phase counters that problem by creating hybrid centralized models, establishing a core data team that defines standard practices, supported by separate analytics teams for specific departments. The business scales as a data-driven organization, as each team works with the centralized team.  

Stage 5

The fifth phase relies on data scientists and predictive technologies, adding value through the application of augmented analytics and machine learning. Data scientists anticipate behaviors by either inventing or leveraging machine learning, statistical technologies, and predictive capabilities. These methodologies help optimize operations at scale when shared with the centralized data team and can answer foundational questions like whether the business needs to change, if forecasting is optimal, or if sales priorities need reevaluation. At this stage, time to value is generally a longer cycle and requires more significant investment and patience. In step with these greater commitments and in contrast to earlier stages of data maturity, the corresponding payoff is generally much greater as well.

How to get through the stages, successfully

Determine your firm’s needs by asking key questions. Do you use different reporting platforms for separate functions or integrated reporting across various tools? Is data siloed or integrated and accessible companywide? A differentiator between stages one and two is whether organizational data is compiled in a single data warehouse, optimized for shareability, and structured for multiple business queries. A firm has reached the second phase if it has centralized reporting and an ETL process providing good data models.

Focus on data breadth to help advance your data maturity. It really is unusual for any organization to have all its data collected within that central source of truth, but what you do have in a central repository is an excellent marker of your data operations. Ideally, new system data can be merged with existing data to build on this and answer potentially unanticipated questions. A more modern approach here is to blend data post warehousing.

Data governance is vital for determining data maturity. Do relevant company roles have necessary access to the right data? Early-stage businesses often have unruly data playgrounds, accidentally sharing slides that are not necessarily backed by verified calculations. Alternatively, a well-governed system under control of the data team releases only approved data or data clearly identified as preliminary or developed under playground rules. As your organization matures, removing sensitive information such as personally identifiable information (PII), salary data, or other proprietary facts may also become a ruling data management guideline.

Hybrid centralization models occur when business units embed their own analysts to use tools from the central data team. In these scenarios, machine learning models can be leveraged by existing tools and incorporated into the analytics workflow.

Data maturity fosters the data-driven organization

It is important to mature naturally through the five stages; jumping from basic reporting to augmented analytics means potential loss of some key foundational capabilities, possibly diminishing the desired impact from more advanced analytics. Without a strong BI foundation and an established and practiced hybrid team, organizations could potentially develop advanced insights but may be unable to apply them easily throughout business operations. The communications framework suffers, as does the infrastructure that allows shared insights, demonstrating the power of traveling through the maturity model step by step. Firms that follow the curve are empowered with both infrastructure and experience—poised for data-driven success for the long term.