Read - How to Create a Data Strategy - Part l 

Read - How to Create a Data Strategy: Part II – Ten Steps to Success

To survive and thrive in today’s information economy, organizations need to establish a compelling data strategy that advances the company’s key objectives and goals. This is not a job for the chief information officer or even chief data officer; it’s a task that the CEO must own and champion and assemble his best and brightest to execute.

The first article in this series examined the reasons why organizations develop data strategies, outlined a methodology for creating one, and provided a sample visual depiction of a data strategy. The second article described the first five steps in a 10-step methodology to create a data strategy, including how to assemble and educate a team to create the strategy and develop a vision statement to guide the initiative.

This article will examine the last five steps in the methodology:

1. Develop recommendations
2. Develop a roadmap
3. Develop a business case
4. Prepare for change
5. Execute the strategy

1. Develop Recommendations

With a vision statement in hand, the next step is to develop a strategy to achieve the vision. The strategy refines the vision from a high-level concept into a series of concrete actions. 

Objectives. The first step is for the cross-functional data strategy team to define three to five objectives to achieve the vision. An objective is a high-level goal that supports the vision. The best objectives are action-oriented and start with a verb. For example, “Facilitate governed self-service” or “Break down functional silos”. The group should brainstorm dozens of objectives and then consolidate and prioritize them until it has about six to ten good ones.

Typically, there is an objective for each of the foundational elements of any data program:

• The leaders who guide the data strategy
• The organization and program that will execute the strategy.
• The architecture required to support the free flow of data.
• The processes for designing, building, and managing the environment.
• The people who manage data processes and those who consume data output.
• The data that fuels the organization (volume, velocity, variety, validity).

Recommendations. The group then works in small teams to develop three to five concrete recommendations for each objective. The recommendations must be practical and specify and owner, a process, a cost, a timeline, and an outcome or mission. For example, to facilitate governed self-service, the team might create the following:

• Recommendation: “Establish a review board to evaluate changes to cross-functional reports or new submissions.”
• Owners: John Doe and Nancy Deer
• Process: Define review board mission, process, SLAs, member requirements, and communications strategy
• Cost: No capital costs or additional labor costs; all volunteer
• Timeline: Three months to create.
• Mission: Verify changes adhere to corporate data standards without inhibiting the free flow of information.

Guiding Principles. During this phase, the data strategy team might also define guiding principles. These are value statements that define core principles about how the team will design, implement, and adapt the data strategy and/or how the team will work together and interact with its stakeholders. Sometimes the guiding principles are interchangeable with the objectives. Many guiding principles start with the word, “We”. Examples are:

• “We treat data and analytics as a critical asset to our company’s success.”
• “The business owns and drives the data strategy and data analytics program.”
• “We tailor self service to each individual and role.”
• “We foster a culture of innovation and governance.”
• “We celebrate successes.”

2. Develop a Roadmap

With all the recommendations in hand, the data strategy team (or team leader) examines them to identify timeframes and dependencies. The team then creates a chart that depicts a three-year roadmap. Typically, there is a high-level chart for executive consumption (see figure 1) and a more detailed Gantt chart for managing projects.

Figure 1. Executive-level Roadmap

3. Develop a Business Case

The data strategy team then compiles the output of the data strategy team into a business plan that can be presented to the executive team for approval. The business plan should specify the plans and resources required to implement the strategy, including capital investments, new processes, new hires, and new organizational structures. The plan should use the company’s template for submitting proposals and contain sufficient description and financial projections to satisfy the executive team and board.

Before submitting a plan, the data strategy team should socialize its findings and direction with the executive team and other key stakeholders. This is important to calibrate executive interest in the data strategy, identify champions and saboteurs, and modify the proposal to reflect political realities and the current funding climate at the company.

Would you like to speak to an Eckerson Group expert about data strategy?

To get the business plan approved, the data strategy team will need a strong executive-level champion who is willing to work with the group to tailor the recommendations and then sell the plan to the rest of the executive team. Without a strong champion, the data strategy will have little chance of success since its recommendations will likely touch every part of the business and cost a significant amount of money.

4. Prepare for Change

Convincing the executive team to adopt a data strategy—as hard as that can be—is easy compared to getting the organization as a whole to manage and use data differently. This requires a well-defined change management plan to ensure a successful rollout with high-levels of adoption. Many large organizations now have a change management team that can help craft a communications plan, run workshops and educational events, and develop marketing material.

To ensure adoption, executives must vocally champion the initiative with well-timed presentations, memos, and public statements. Executives must tout the strategy as critical to the company’s success. The company may also need to tweak job descriptions and incentive plans to ensure success. Everyone needs to know how the new strategy will affect their jobs and what is expected of them once the strategy is rolled out. 

Ultimately, the change management plan must appeal to the “head, heart, and herd”. People need a rational explanation for the change (head); an incentive (either monetary or visionary) to change (heart); and knowledge that other people are adopting the change. We are shaped to a large degree by what other people around us do. Getting a small group to buy into the change and amplifying their work is an effective change management technique. Of course, if the herd doesn’t change, then it might be necessary to change the herd (i.e., hire and fire employees) to get people who exhibit the requisite behavior.

5. Execute the Strategy 

The final step is to execute the data strategy. Today, most organizations hire a chief data officer (CDO) to oversee all things data—including data warehousing, business intelligence, analytics, data infrastructure, and data governance. The CDOs often work closely with data security teams to ensure data is adequately protected throughout its lifecycle without unduly hampering user access to information. The CDOs also work closely with the office of innovation to foster a steady stream of proposals and prototypes for new data-driven products and services.

According to Gartner research, 30% of CDOs report to the CEO. This is the ideal situation; it ensures the CDO has the clout to execute a comprehensive data strategy. Reporting to the CIO is the least advantageous position since it places the data strategy within the IT department. To succeed, a data strategy must be owned and run by the business.

Some CEOs wonder if the CDO should have operational responsibilities or work across functional groups in a federated manner. The federated CDO ruffles fewer political feathers and is easier to implement, but the CDO will struggle to implement the data strategy. Ideally, the CDO has several direct reports who oversee key areas of the data delivery supply chain. With this level of operational responsibility, the CDO can better align initiatives, processes, and infrastructure to avoid data silos and ensure the free flow of data throughout the organization.

Summary 

A data strategy is a collaborative effort. Because data touches so many parts of an organization, any initiative to change the way the company harnesses data requires input from many people. A cross-functional team that follows the ten-step methodology described in this series is a good way to ensure that there is widespread support for a data strategy. The next article in this series will drill into the elements of data strategy using the visual depiction from the first articles in this series.