In With the Right Crowd
Getting management on board to support statisticians' roles
by Ronald D. Snee, Roger W. Hoerl and Angela N. Patterson
First, the good news. The importance of statistics related to the way the world does business has never been greater:
- Large-scale statistical training through improvement initiatives like Six Sigma has spread statistical thinking, and it has become common in fields such as finance and healthcare.
- Commercial statistical software puts the ability to do complex statistical calculations on the desktops of managers and executives who neither have nor need specialized statistical expertise.
- About half a million students each year take a general introductory statistics course in college in the United States.
- The web puts vast storehouses of statistical information at everyone's fingertips.
Now, the bad news: The statistician and quality professional might become the proverbial middle men who get cut out by these advances.
Consider the plight of travel agents. When was the last time you booked an airline ticket or hotel reservation through a living, breathing travel agent? Why should you when it's easier and cheaper to do it yourself using powerful web based programs that can instantly scour the universe for the best deals? Using the business term for "cutting out the middle man," travel agents have been "disintermediated."
Statisticians run a similar risk unless they find a way to break out of their traditional roles. During the past 40 years, much of the work done in corporate statistics groups can be described in two ways:
- Performing technical tasks, such as analyzing data and designing experiments at the request of someone directing a project.
- Advising others how to attack a statistical problem.
These types of activities have typically been referred to as statistical consulting. In the days before all of the good news, this role of enabler made sense. Corporate statistics organizations grew comfortable in this role, "owning" the organization's statistical methods. As a result, statisticians came to be viewed as narrow specialists, such as accountants or lawyers, and as passive consultants rather than equal members of business teams.
It's a precarious position. You can no longer own statistics when anyone with a computer can download and use statistical software. Further, narrow technical tasks can easily be moved to low-cost countries, and many cost-conscious businesses are asking why their people can't analyze data themselves.
As demand for statisticians' traditional services dries up, statisticians must move from the role of passive consultant to active colleague. In other words, they must do what travel agents have so far failed to do: find a way to add significant value in the face of developments that have made their traditional services widely available for free.
The world of statisticians and quality professionals / Figure 1
There are two compelling reasons for statisticians to become active partners, particularly with managers. One of those reasons is opportunistic and the other idealistic. When the infamous bank robber Willie Sutton was asked why he robbed banks, he said. "Because that's where the money is."
Managers control the resources of time, money, personnel and equipment. They are where the action is—setting direction, goals and objectives—and if you want to be in on that action, you need to do far more than regard managers as a distant end-user of your services.
The idealistic motive comes from the convergence of your vocational ideals and values with theirs. Managers are responsible not only for the day-to-day running of the business, but also for improving the way work gets done. Like the best statisticians, the best managers want to help people succeed. They want to see process performance improve and data based approaches and scientific management really work. In realizing those goals for the organization, managers want to fulfill their personal goals.
How do you create opportunities to work more closely with managers? Think marketing. Like a veteran marketer, you must first identify your customers and their needs. Where is the value you can add to make the most difference—in manufacturing, supply chain and distribution, or in a particular division, department or discipline? Which managers in the area you have identified possess the requisite resources and authority to bring you on to a project?
But before you propose to help, determine exactly what needs those managers and departments have in terms of critical issues and operational, financial and corporate objectives.
You can do this by interviewing managers about their goals and the challenges that keep them awake at night. Also, interview the managers' colleagues for their perspectives. This intelligence, combined with your knowledge of the company's objectives, should enable you to see a clear picture of the needs of potential partners.
Once you understand a potential partner's needs, then you can (again, like a marketer) create your value proposition, which details the benefits you can deliver at what cost. In creating your value proposition, be sure you can deliver what you say you can. In fact, you should underpromise and overdeliver.
For example, based on your background in finance, you might realize your company is not getting the best possible return relative to the risk it is taking when investing in a particular portfolio. Rather than lecture people on the use of statistics, you are more likely to succeed if you develop an overall value proposition that quantifies the types of returns that should be possible, relative to the risk appetite of the business.
Once you get people's attention with this high level analysis, you could then develop a more detailed business case—tailored to a specific manager—outlining a conservative estimate of the increased returns, level of effort and resources that would be required, and the general timeframe of implementation and financial return.
It would be most effective to couch this analysis and recommendations in the standard terminology and metrics of the business, such as return on investment. Simply put, the value proposition and business case enable the manager to decide whether your services are worth it.
Working with managers
Not all managers are alike. In seeking out managers to ally yourself with on proposals or projects, you will find at least four common types:
- Low-risk managers think small, avoid outside help to keep the budget down and avoid rocking the boat at all costs.
- The resisters come up with infinite reasons why your proposal won't work, always know a better way and don't really want help.
- The do-it-yourselfers gladly accept all the help they can get and take all the credit they can.
- The collaborators want help, appreciate everyone's contribution and will challenge and reward you throughout the course of a project.
During your career, you will likely work with all four types at one time or another. It goes without saying that the collaborator is the most desirable type to work with. But even beyond the collegiality of the collaborator, the ideal manager will have some additional attributes:
- A significant need for your help.
- Adequate funding and a willingness to spend it.
- An appetite for leading change.
- A desire for a win-win relationship.
In addition, the ideal manager will exhibit the qualities of loyalty, trustworthiness and high integrity that inspire everyone on the team.
Unfortunately, not all managers are ideal or even collaborative. Further, not only are all managers different, but they are also not equal—some have more power than others. Therefore, to find your way to the most beneficial alliance—for you and the organization—you should thoroughly understand where real power resides. Who are the movers and shakers with influence to command resources (including you) and to lead or initiate the kind of projects to which you can add value? Such knowledge of the real dynamics of the organization is power, but it also can mean making hard choices: working with a credit hogging do-it-yourselfer manager with real power versus a collaborator manager with little leverage in the organization. See the sidebar "Click With Managers," for tips on making your relationships with the managers succeed.
Working with teams
Of course, most significant projects involve not just one-on-one relationships with managers, but also entail work with teams. If your role has been primarily consultative in the past, then working as an integral part of a team outside your department might be a relatively unfamiliar experience. Like work with managers, teamwork requires some skills beyond purely technical statistical ability.1, 2
Your statistical expertise, however, is part of what got you on the team. You have the opportunity to help the team put statistical thinking to work and to make sure the team gets the maximum value from data and statistical techniques.3
You should therefore promote a standard problem solving approach, such as define, measure, analyze, improve and control (DMAIC) that systematically structures data-driven techniques.4 Simply resorting to data in an ad hoc fashion greatly diminishes the power of statistical methods and the value you can bring to the team. By contrast, frameworks like DMAIC enable the team to build successively on each stage of its work, use statistical methods to get at the root causes of problems and solve them once and for all.
During a project, there might be teachable moments. These times are when you can educate the group about precisely how statistical thinking can solve the problem at hand and similar problems that might arise in the future. But don't play the role of expert witness or use statistical jargon to intimidate or dominate other team members.
Remember, too, that you are trying to help the team find the best business solution, not the best statistical solution. Certainly, as the team's primary interpreter of data, you are in a position of power, but you should use that power responsibly and diplomatically.
Participation as an equal team member also requires additional skills.5-7 You will need an understanding of group dynamics and the ability to facilitate and manage team meetings. You should also be able to teach statistical principles to nonstatisticians in a way they can understand and put to practical use.
You should possess general business understanding, especially of finance, as well as a specific understanding of your company's business: its markets, customers, strategic objectives and key success factors.
As you acquire or exercise these skills and gain additional trust and authority from managers and teammates, look for opportunities to assume broader, more consequential roles, such as project or initiative leadership.
In summary, successful collaboration with managers and teams results when you:
- Learn about the subject, issues and business of the client group.
- Focus on helping people be successful by delivering results and positive financial impact for the organization.
- Provide the best business solution, which might not be the optimal statistical solution.
- Promote the scientific method and the effective use of the right data.
- Move from a consulting mind-set to a collaborative, leadership mind-set.
Insofar as you put these principles to work, you will succeed in transforming yourself from a merely nice-to-have resource to a valuable must-have colleague, thought leader and change agent.
© Ronald D. Snee, Roger W. Hoerl and Angela N. Patterson, 2008.
Click with managers
It is unlikely you will match well with every manager. Personal chemistry with a colleague is mysterious. Nevertheless, once you have made a compelling case for your participation in a project, there are some useful principles you can adopt to help the relationship with the manager succeed:
The manager is the boss. In all likelihood, you sought out the manager, not the other way around. And even though you might have been included on a project because you proposed it or demonstrated how you could add value to an existing activity, that doesn't mean you are in charge. Even the most collegial managers will expect to call the shots, because they are ultimately held accountable for results.
Money is the language of management. You must be able to help translate process and operational improvements into financial results. What will shorter cycle times, the eradication of root causes of problems or a reduction in process variation mean in terms of faster speed to market, increased equipment uptime and higher productivity? And what will those improvements mean to the top or bottom lines?
Deliver useful results and information. Advertisers long ago figured out that consumers are interested in the benefits of a product, not its features, no matter how impressive. Similarly, managers don't care about the complexities or technicalities of your statistical analysis; they just want information they can use to be successful.
Focus on their needs, not your brilliance. That means developing your skills clearly and concisely presenting statistical studies to managers.1
Challenge, don't attack. Think of yourself as a thought leader who has a unique perspective. Adding that perspective might involve disagreeing with the manager, but do it dispassionately and in a spirit of mutual inquiry. Be especially on guard against trying to dominate or win by high-handedly invoking statistical arcana that only you understand.
Get to know the managers. Understand their issues and concerns and the unique pressures of their jobs. For example, senior managers have different needs than middle managers. They are typically more focused on business results and less on operational specifics.
Middle managers have the toughest jobs of all. They must balance direction from above with the realities of getting the work done. Remember, too, that managers are complex people, and you should try to get to know them from those perspectives as well.
Deliver bad news thoughtfully. Because of the predictive and analytical power of statistics, it often falls to the statistician to be the first to understand that something won't work or isn't working as planned. In delivering bad news, take care not to give the manager ammunition to shoot the messenger and ignore the message.
Always review your conclusions with the manager before presenting them to a wider audience. Be sure your conclusions are based on rock solid, clearly articulated analysis. Don't merely present the problem, but whenever possible offer a potential solution.
Fulfill the opportunity. Focus on solutions and results, not on methods. The idea is to make a difference, not showcase a discipline. Always strive to deliver results in scope, on time and within budget.
Share the success story together. Join managers in publicizing the results of a successful project, inside and outside the organization. Invite managers to co-author articles and speak at meetings and conferences.
Help the manager succeed beyond the scope of the project. Little things mean a lot. For example, you can pass along articles and books relevant to the manager's needs and interests.
You can provide the manager with useful intelligence about your discipline and the organization. Big things mean even more: suggesting high-performing job candidates, performing above and beyond the call of duty, and identifying additional opportunities. —R.S., R.H. and A.P.
- Marsha Redmon, "Presenting to the Big Dogs," Training and Development, October 2001, pp. 32-34.
- Ronald D. Snee, "Non-Statistical Skills That Can Help Statisticians Be More Effective," Total Quality Management Journal, Vol. 9, No. 8, pp. 711-722.
- Ronald D. Snee, Kevin H. Kelleher and Sue Reynard, "Improving Team Effectiveness" Quality Progress, May 1998, pp. 43-48.
- Ronald D. Snee and Roger W. Hoerl, "Statistical Leadership—As Traditional Workplace Roles Change, Learn to Transition from Consultant to Leader," Quality Progress, October 2004, pp. 83-85.
- Ronald D. Snee, "Adopt DMAIC: Step One to Making Improvement Part of the Way We Work," Quality Progress, September 2007, pp. 52-53.
- Allan R. Cohen and David L. Bradford, Influence Without Authority, John Wiley and Sons Inc., 2005.
- Snee, "Non-Statistical Skills That Can Help Statisticians Be More Effective," see reference 1.
- Roger W. Hoerl and Ronald D. Snee, Statistical Thinking—Improving Business Performance, Duxbury Press, 2002.
Ronald D. Snee is principal of performance excellence and lean Six Sigma initiative leader at Tunnell Consulting in King of Prussia, PA. He has a doctorate in applied and mathematical statistics from Rutgers University in New Brunswick, NJ. Snee has received the ASQ Shewhart and Grant medals and is an ASQ fellow.
Roger W. Hoerl is manager of General Electric's corporate applied statistics lab. He has a doctorate in applied statistics from the University of Delaware. Hoerl is an ASQ fellow, an ASQ Brumbaugh Award recipient and a GE Global Research College fellow.
Angela N. Patterson is a member of General Electric's corporate applied statistics lab and a certified Six Sigma Master Black Belt. She has a doctorate in applied statistics from Virginia Tech, where she is an adjunct faculty member. Patterson is a member of ASQ.