“Strategy without tactics is the slowest route to victory. Tactics without strategy is the noise before defeat.”—Sun Tzu.
This wisdom from Sun Tzu, the ancient Chinese military strategist, summarizes what I see happening in the world of data. Organizations talk about data as one of their most valuable assets, but then are ad-hoc and chaotic about their use and application of data.
An organization must treat data as an asset, in actuality, not just as an ideal. If it does not, that coupled with an associated lack of ownership and purpose surrounding data-driven functions, makes it much harder to reliably consolidate data from multiple sources to derive fast, high-quality actionable insights.
When considering how to put data to work for your organization, it is essential to go beyond talking about data as an asset to actually valuing and treating it as one. As Sun Tzu advised centuries ago, to do so means you need a strategy.
A data strategy consists of a vision and an actionable roadmap. It outlines a set of governing methods, approaches and tools for everything an organization must do with its data. It defines the roles and responsibilities of people within the organization as they produce, manage and consume data. It culminates with a plan for how an organization collects, stores, processes, controls, disseminates and consumes data.
Developing a data strategy is a complex operation, but you can think of it as having four key pillars:
1. Changing the culture
People lie at the center of the data strategy. People make decisions, solve business problems, perform actions, create opportunities, manage risk, measure success, learn to adapt, and communicate with each other.
To successfully manage data as an asset, organizations need to change their culture—a difficult process that must start at the top. Leaders should broadcast their data strategy so that people at every level in the organization understand how the use of data can serve their mission and objectives. Senior leaders should seek out experts on their teams and learn from them—and experts should persuade their leaders to adopt best practices.
Leaders should then encourage partnerships and collaboration across offices and include talent across policy, program, operations and technology teams.
Also, leaders need to deliver quick wins to demonstrate the value of applied data analytics and gain support for their strategy. Without this quick win, the organization may disregard the broadcasted data strategy and best practices. Developing a roadmap of data initiatives that can lead to these quick wins will help accomplish this.
2. Sorting out skills
To value data as an asset, leaders should integrate the data team within every part of the business. To thrive in this data age requires building skill sets that might not be common in the organization’s environment.
A gap analysis of the skill sets available can uncover the ones the organization still needs. Data architects, engineers, scientists and storytellers may be needed to build a data team that can glean insights from data and communicate them effectively to the decision-makers of the agency.
3. Choosing technology
Technology is at the forefront of successfully managing data as an asset. By assessing the options and finding the right tools, leaders should lay a foundation for technologies of tomorrow, including artificial intelligence (AI), machine learning, intelligent automation, and blockchain among others.
The tools chosen should be easily accessible to everyone and empower teams to act quickly, and tailored to meet the organization’s needs. In some cases, commercial off-the-shelf (COTS) technologies will fill the need. In others, custom tools are a better choice. That decision depends on the specific requirements of an organization’s data. Any solution, whether COTS or custom, must be secure and privacy-aware when connecting data from multiple sources.
4. Monitoring metrics
Metrics and benchmarks are an important point of reference in determining an organization’s effectiveness toward achieving strategic goals, implementing methods and efficiencies, and establishing processes for continued improvement. They reflect and support the strategy and vision and provide a quantitative method for measuring success.
Metrics provide accountability, transparency, auditability and an understanding of data-driven adoption within an organization. A full discussion of what you can measure and what the metrics mean is beyond our scope here, but it takes time to explore the choices and develop processes for monitoring and acting on the measurements.
Data is too important to leave its application to the whims of the moment. Developing—and adhering to—a strategy takes some time and determination, but organizations who make the effort to do it are rewarded in time saved, efficiencies gained and goals met.
For additional insight on this topic, read one of my earlier posts, “Not your granddad’s government data sharing.”
About this author
Director, Consulting Services
Sumit serves as the chief data scientist in CGI Federal’s Emerging Technology Practice. He leads a team of data engineers and data scientists who are defining and driving technical direction of prospective business opportunities, and work with business stakeholders and client executives to develop overall ...