When conversation in organizations turn to analytics, topics normally include the quality and accessibility of data, the infrastructure for storing and processing data, the necessary level of analytics sophistication, and the unique skillsets required to build a successful analytics program. Often overlooked in is the importance of having an enterprise level analytics governance strategy.
A recent GAO report titled Data Analytics to Address Fraud and Improper Payments focuses on key considerations for establishing and refining an analytics program. An organization can build a technically sound model, but if it’s answering the wrong question or the model output is not adopted and used by decision makers to effect change, the value will not be realized.
To develop an effective analytics governance strategy an organization should consider the following seven questions:
1. How much are we investing in analytics and what is our return on investment?
A common oversight is for organizations to invest in analytics software tools and data infrastructure, then prematurely decide that the investment in analytics is too expensive. Instead, it is often wise to start with a small project that addresses an interesting problem, that has relevant data, with an experienced data scientist, using open source tools; this can provide a quick win and build momentum for the program. According to Jeff Deal and Gerhard Pilcher, authors of Mining Your Own Business, A Primer for Executives on Understanding and Employing Data Mining and Predictive Analytics “Data mining projects take some time to staff up, get new data analytic systems in place, analyze the data, and create the recommendations required. Strong leaders must demonstrate patience and perseverance to champion the project. Time, money, personnel, and resources must be maintained so that the project can stay on course, and produce the success that is desired.”
2. How are we defining value for the anticipated return?
Of course “dollars” jump to mind when thinking about any return on investment, but often the work hours saved through increased productivity represent a large part of the benefit from an analytics project. A fraud detection solution Elder Research developed and deployed for the U.S. Postal Service – Office of Inspector General reduced the number of hours spent by fraud investigators on a case by 30% and increased the dollars returned per case by 35%.
3. Does our analytics strategy support our business strategy and performance goals?
Analytics project decisions are often centered on the analytics software tools that are available within the organization, or specific analytical techniques that are well understood by the analytics team. Allowing a tool or technique to drive the analytics strategy limits the types of problems that can be solved and may not align with strategic goals for the business. Using analytics to achieve a sustainable competitive advantage and generate significant return on analytics investment begins with a well-conceived analytics strategy and roadmap for success that is aligned with, and supports, the overall business strategy.
4. Are we prioritizing the right projects?
Choosing projects that benefit the entire organization or ensuring modeling efforts are focused on current or emerging needs provides insight to make more strategic decisions. The analytics team should assess the unique business challenges for the organization, match those challenges with relevant data and resources, and establish processes that grow capabilities and institutionalize analytics to ensure key decision makers have access to actionable results.
5. Should we abandon or improve our models?
If a model is not regularly updated with fresh data, business owners and end-users can become dependent on outdated model results, increasing the likelihood of biased recommendations that are not based on objective results. Once a deployed model answers the original question look for opportunities to add new data sets and ask deeper questions. Use this initial success to demonstrate that the benefits of data analytics merit the required investment of time, money, and emotional energy to solve even bigger challenges with greater potential return on investment.
6. How can we improve our analytics capabilities?
Technology continues to improve, data is being created at an ever increasing rate, and new analytical techniques continue to emerge. A successful data analytics program must always be strategic and deliberate about moving forward. Continue to invest in analytics tools, making model results easily accessible to decision makers, and use analytics training to enhance skills and institutionalize a data-driven culture. Elder Research has found that the presence of an “analytics culture” is one of the strongest indicators of future analytics success. Building an effective analytics culture can transform a business.
7. Do we have the right skillsets in place for success?
The ideal analytics team structure allows the entire organization to benefit from the insight provided by the analytics solutions. Analytic knowledge and skill turn the available business data into information that can be used to take action. Data science requires curious, analytically-minded people. These people must be technical enough to deal with the many IT aspects of data science, be familiar with the complex mathematics of statistics, and be creative problem solvers. It is rare that any one individual possesses all these skills, so it is important to build an analytic team that covers the full breadth of data science activities. As stated in Mining Your Own Business “Changing a corporate culture is never easy. That’s why successful predictive analytics initiatives demand strong leadership from one or more “champions” who are enthusiastically committed to analytics and who command sufficient respect within the organization to enlist the commitment of others”. Having a champion with direct access to a Chief Data or Analytics Officer provides the best opportunity for success.
Whether you have already started down this pathway or not, consider these seven strategic questions before making further investment in analytics to save time and money, guide project decisions based on stakeholder requirements, and inform the development of data analytics models to deliver value. Start small and move forward deliberately, as the time spent planning will more than pay for itself over the long run.