What Really Is a ‘Stretch’ Objective?

When working harder or longer won’t do the trick

Published: Monday, April 18, 2016 - 13:34

One poorly understood concept in lean Six Sigma is how much to “stretch” when setting S.M.A.R.T. goals. These letters are defined as S—specific; M—measureable; A—assignable, attainable, or achievable; R—realistic, reasonable, or relevant; and T—time-based or time-bound. Regardless of the different interpretations, what do we really mean by these terms?

Clearly, we don’t want to set ourselves (or others) up with impossibly high goals or low-level objectives that discourage motivation. If you have worked in sales or other jobs where at least a portion of your wages were based on pay-for-performance, you know what I’m talking about. How can a manager know what an individual is capable of selling? How can a team leader know what is applicable to an improvement project?

Earlier in my career, I was the sales forecasting manager at Coca-Cola USA. Each quarter, I was responsible for producing an unbiased, quantitative sales forecast for our brands, which would be used in establishing budgets and sales goals. My principal customers were the executive vice president of sales and the chief financial officer. My job was to help them come up with a set of forecasts that they both felt good about. It was very helpful if my “objective” forecast fell between the desired values of the two senior executives. Otherwise, I was left with the dubious task of convincing both of them that their numbers were both too high, or both too low.

I assumed going in that the EVP would want high numbers and the CFO would want low numbers. I was dead wrong. The EVP wanted low forecasts, so he could set “stretch” goals for the sales force that would be easier to attain. In turn, the CFO was a true businessperson who believed that you couldn’t grow unless you invested in the market, and he wanted higher numbers. What was reasonable and achievable depended on the minds of these beholders.

Later in my career, I held sales positions, and was subjected to the dysfunctional process of having managers set my personal goals and objectives. When I eventually became a regional sales manager, I found myself on the flip side of the equation.

This is tricky stuff, folks: If objectives are set too low, you can easily achieve the numbers by working a little harder, or a little longer, or both. This is insidious; it does nothing to encourage innovative thinking, and eventually workers become overburdened and unmotivated. Set objectives that force thinking outside the box, where working longer or harder won’t do the trick.

For example, would anyone who starts her own company consider doing her job exactly the same way as she did at her previous job? Would she stretch for only a 3- to 5-percent improvement using previous methods? Of course not. Then why don’t we approach things the same way when we’re trying to get the best business results and keep our people motivated?

Setting goals

Make sure that establishing goals and objectives is a two-way process. Goals often begin at the top with senior management making commitments to shareholders. However, meeting these objectives hinges on many things, including the time and resources available, and employee skills, experience, and motivation. For objectives to be meaningful to everyone, they need to be negotiated.

One way to establish meaningful goals is to benchmark against processes used by similar businesses or even best practices in other industries. Benchmarking is a good way to determine what is possible. Fortunately, different organizations perform similar processes (e.g., hiring people, paying bills, budgeting, promoting marketing strategies, creating legal contracts), and can therefore be researched through industry publications, formal benchmarking studies, custom studies, networking, and of course, the Internet.

Another important consideration is that “stretch” objectives need to be statistically significant from the baseline of the metric we are trying to improve. Otherwise, the result could appear to be better when the difference is due to random variation. For instance, if a manager’s goal is to reduce overtime hours, he might think that a 10-percent reduction would be a good target. But as the control chart in figure 1 shows, a 10-percent reduction wouldn’t be statistically significant from what the company is already doing. 

Figure 1: A control chart showing overtime hours. Click here for larger image.

We need to set our objective far enough away from the metric baseline to ensure that the results are, in fact, significantly better. If we don’t have enough baseline data to create a meaningful control chart or do an hypothesis test, we must find other ways—qualitatively, if necessary—to estimate the current-state variation before we establish objectives.

Don’t begin new lean Six Sigma projects without a discussion about the most appropriate level of “stretch.” Then proceed only after a thorough analysis of the market, customer requirements, available resources, employee feedback, and other factors, including the variation of your proposed “y.” Spending the needed time on setting S.M.A.R.T. objectives goes a long way toward achieving success in lean Six Sigma projects.

About The Author

Ken Levine

Ken Levine is a Lean Six Sigma management consultant. He retired as director and lead instructor in the Lean Six Sigma certification program at Georgia State University in Atlanta in 2019. He holds a doctorate in business administration. Levine retired from The Coca-Cola Company in 2000 where he held the position of director of continuous improvement in the Coca-Cola USA division for three years. Levine is a Six Sigma Master Black Belt and Certified Purchasing Manager. He has previously published “Root Cause Analysis Takes Too Long,” “How to Determine the Worst Case for a Process,” “Recycling Your Meeting Waste”, “What Really is a Stretch Objective”, and “Ensuring LSS Success with a Robust Define Phase” in Quality Digest.

For more articles by Levine, click here.