Data analytics has been called the most powerful decision-making tool of the 21st century. Even though it has come of age only within the past twenty years, thousands of businesses, governmental agencies, and nonprofit organizations have already used it to dramatically increase productivity, reduce waste and fraud, enhance quality, improve customer service, boost revenues, optimize strategies, combat crime and terrorism, and solve a host of other tough challenges. Elder Research, CEO, Gerhard Pilcher, and Vice President of Operations, Jeff Deal, coauthored Mining Your Own Business to provide an easy-to-read overview of data mining and predictive analytics for organizational leaders who want to know more about these powerful tools and develop an analytic capability in their organization.
This blog, drawing from chapter 3 of this book, reviews the three most important keys to leading a successful data science initiative.
Choosing the Right Project
The first step with analytics is deciding what problems to solve. 
Begin by identifying a narrowly defined problem that is widely acknowledged within the organization as a pain point. Organizations of any size have many problems data analytics can help solve. In picking your initial project, consider what problems keep you awake at night. If there are no burning issues keeping people in your organization awake at night, and if no one is asking for the results you’ll create, the project will probably turn out to be a waste of time and money. Think about what forces are impacting your organization that you need to understand better? What issues would you very much like to resolve? To ensure success, leaders should view the initiative as a vital step toward attaining the organization’s overall goals. All involved in the project should buy into its success, and they should be willing to act on the insights the analytics will uncover. The project should be practical, with adequate funding and access to data.
Some beginning projects that have worked well for our clients are reducing fraud, reducing customer turnover (churn), predicting costs, and improving production quality. Limit the focus of your effort. If your issue is fraud, for example, begin by identifying and reducing one particular type of fraud, rather than going after every form at once. A primary purpose of your initial project should be to prove the value of data analytics and get people on board about its potential benefits. Obtaining buy-in is key, because analytics ultimately will change the way people in the organization make decisions.
View this initial data analytics project largely as a sales tool for future full-scale initiatives. Use it to demonstrate to all levels of management that the benefits of data analytics merit the required investment of time, money, and emotional energy.
Starting Small
Some companies new to data analytics try to push ahead too fast. Instead of investing $75,000 or so in a modest initial project, they rush out and hire two or three people who have some experience with analytics, spend $500,000 on software, and announce that the company is now “data-driven.” However, pursuing a data analytics initiative without proper planning and organizational buy-in is like purchasing an expensive piece of home exercise equipment without sufficient commitment. The equipment may seem exciting at first, but without a dedicated regimen, it will soon end up sitting idle in the basement or serving as a clothes rack in the corner of the bedroom.
Several years ago the managers of a very large company asked us to help them with their vision for using data analytics as a vehicle for transforming the entire healthcare industry. After our team of four people met with their team of twelve people for two days, it was clear that the organization’s analytics strategy was too grandiose to get off the ground. It was as if they were trying to reach the moon with a hobbyist’s model rocket. After two days of meetings, the potential project collapsed under the weight of its unrealistic goals.
In contrast, another of our clients, one of the largest insurance companies in the United States, did everything correctly. When we came in for the initial kickoff meeting, the leader of the project had already assembled the key SMEs, executives, IT people, and other stakeholders. The project this company presented to us involved a widely acknowledged, well-defined pain point associated with a particular line of insurance. Its clear focus and narrow scope led to highly productive meetings and substantial buy-in.
Although committing all of these people to a day and a half of meetings was expensive, the investment paid off. The initial project was very successful, and since then the company has applied data analytics to other problems associated with this same type of insurance. In the future, they plan to expand data analytics into other lines of their business.
Something else impressed us about this client. We were delighted to find that the person responsible for data security was very forward-thinking. She was determined to do everything possible within the law to make the information our initiative required available. In our experience, far too many data security people are afraid to share any information that contains personal data about customers or clients. Fortunately, this person put forth the extra effort to get us the data we needed, without violating customer confidentiality or the law.
Leadership is Key
A study our firm conducted of the projects we had completed in our first decade since our founding in 1995 showed that 90+ percent had been technical successes, but only 65 percent of those had been implemented; that is, only two-thirds had been business successes. [1] Many causes can contribute to a lower business success rate, but the biggest one is a lack of organizational commitment to implementation. Unfortunately, many organizations simply are not willing to operationalize the recommendations that the analytical work provides, even when it’s obvious that these recommendations will lead to significant improvements.
Why do so many organizations invest significant amounts of time and money in a data analytics project and then fail to implement the resulting recommendations? By this stage, they’ve paid all the costs, and they have proven returns on out-of-sample data. Without implementation, however, they realize no gain.
We believe the two major reasons are the absence of strong leadership and a lack of buy-in by key decision makers. (A minor reason is failure to understand the results, which is a failure on the part of both analyst and client.) To implement data analytics recommendations, organizations often must develop new policies and procedures, change long-standing processes, retrain personnel, and even transform corporate cultures.
Increase the probability of success by involving key stakeholders from the beginning. A data analytics project is likely to fail if you conduct it without involving the stakeholders, and then upon completion tell them, “Here’s what the data shows. Now make use of it.”
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.
Because the results of a data analytics project can take quite some time to manifest, staying the course can require considerable patience and perseverance. As an organization continues to invest time and money into the initiative, some leaders may get anxious or even fearful. “What if this project doesn’t pan out?” they may start thinking. “Are we simply pouring money down the drain?” Sometimes other people will suggest other problem-solving techniques, and leaders may be tempted to divert funds from the data analytics initiative to alternative approaches that may appear to be faster and cheaper. Leaders should continually remind themselves and all involved that quick fixes rarely provide effective long-term solutions to complex problems. Courageous, positive leadership is necessary to ensure that the data analytics initiative stays on course until successes begin to manifest. At the same time, the analysts need to keep those working on the initiative encouraged with reports of early findings.
Summary
A management revolution is underway in the world of business and government. In the years ahead, the most successful organizations will be analytically competent. Leaders of analytics initiatives need to gain the knowledge, vision, and passion to be on the cutting edge of this revolution. If you are starting or leading an analytics initiative keep in mind these factors to ensure success:
- Choose a project that supports your organization’s overall goals and make sure you have key stakeholder buy-in.
- The goal of your initial project should be to prove the value of analytics and get people on board about its potential benefits.
- Make sure the project has adequate funding and access to data.
- Start small and use the results to build momentum and leadership support for follow-on projects with greater potential.
- Positive leadership is necessary to ensure that the data analytics initiative stays on course until successes begin to manifest.
- Assign a project champion with leadership support to shepherd the implementation and adoption of the results.
