Most employees sincerely want to perform well for the company. There are some exceptions of course, but the true saboteur is a rare thing. People usually will perform to the expectations and priorities they perceive as important to management. However, they always will strive first and foremost to meet their published goals or performance measures. This is why developing truly useful and well-integrated performance measures is crucial.
Measuring people and departments is a dangerous undertaking. Deming went into this with his "Seven Deadly Diseases" that plague corporations. Even the best-intentioned measures can work at cross-purposes to the overall organization's performance.
Many organizations have poorly scoped and compartmentalized, or what I like to call "silo-ed," measures and goals. Indeed, industry tends to develop goals for departments and individuals in isolation.
Such performance measures easily can encourage behaviors that look good for one area of the organization but create chaos for another. Typically, organizations track the outcomes of these actions separately for each department. So, the performance reviews of the party that caused the chaos rarely reflect the unintended consequences.
Let me share my experience with three real-world examples from a chemical plant to highlight this phenomenon.
In 2005, Company A's procurement department's bonus goals were linked to raw material cost reduction. The procurement agents were given the authority to substitute raw materials that were chemical equivalents of those currently approved. (Previously, the product line manager or the principal chemist was responsible for approving new material streams.) The agents found a much cheaper source for a key raw material in China. The material safety data sheet and certificate of analysis matched those of the current material. So, the new material was ordered.
However, the new material came in 55-lb paper bags. By the time it reached the port by boat, it had absorbed so much moisture that it no longer was usable. Unfortunately, the plant generated several batches of off-spec product before it detected the problem and rejected the new material. Procurement was pursuing its goals but this resulted in operations missing one of its goals, namely, to produce 99% on-spec product.
Later, the same company created a logistics department to contract transportation rates, select shippers and schedule trucking shipments. In the past, operations would contact a firm on an approved list of shippers a few hours before a bulk batch would be ready for pickup. There rarely was deadtime.
The logistics department, which was graded on cents/mile, started scheduling truck pickups two-to-three weeks in advance to get the lowest rates. This created a new operational problem -- deadtime due to truck delays.
Typically, this stranded reactors for an average of 18 hours, costing $400/hour in lost revenue. That's around $7,000 per batch. Saving a nickel per mile driving all the way around the equator would amount to only $1,300! The change resulted in a loss of more than $6,000. Logistics met its goals but operations began missing one of its key measures, pounds produced/hour.
We're all familiar with the "run equipment to failure" approach that casts the maintenance group in an unfavorable light while absolving the operations shift team of one of its primary responsibilities -- taking good care of the plant and equipment. Company A gave that a different dimension. To reduce production cost, management rolled out a program that awarded a gift card every month to the shift producing the most pounds the quickest. There's nothing wrong with that, right? After all, product cost directly relates to raw material cost and cycle times. Reducing cycle time would increase the sales margins the company enjoyed.
However, something less admirable occurred. Shifts began to figure ways to take shortcuts and create hurdles for competing shifts. This caused a sharp increase in quality issues and process safety concerns. The plant saw no significant gain in saleable product but the more ingenious/devious shifts were able to claim their gift cards.
On first glance, the goals and performance measures were well intentioned -- but their results damaged overall performance. However, because of the organization's silo-like nature and the manner in which the goals and measures were developed, communication never reached management in a way that could prevent these outcomes.
OPTIMIZING PERFORMANCE MEASURES
As the examples highlight, it's essential to develop well-integrated goals and measures. Several steps can help you come up with appropriate ones:
• First, define the overall objective of the corporation.
• Second, find your corporate gurus who have all the tribal knowledge. With them, build cross-functional teams made up of the key stakeholders in the organization. Recognize that a small percentage of your teams will try to use the measures you design to keep from being held accountable for their poor performance.
• After that, "just do it."
In the beginning, the measures don't have to be perfect. You just work together to decide on what measures and goals will help the company reach a new level of excellence that's aligned with the overall corporate vision. Kaizen techniques are useful here to ensure asking the right questions.
Track the performance measures' effectiveness. If they're not quite right, feel free to change them. Strive for continual improvement -- the key performance measures should evolve and improve over time.
Most importantly, keep in mind bad performance measures can be worse than no measures at all. Remember the old Navy adage: "You don't get what you expect, you get what you inspect." In the end, people will perform to the goals you establish.
To summarize, the best way to prevent unintended consequences of poorly scoped goals and measures is to: get your best cross-functional team, ask "what is the company trying to accomplish and how can we do that?," then set goals and periodically check them for effectiveness. Change them later if needed. Keep keeping up the improvement.
JASON G. LAWS, P.E., is plant manager at Gulbrandsen Technologies, La Porte, Texas. E-mail him at [email protected].
Fellers, G., "Why Things Go Wrong," Pelican Publishing, Gretna, La. (1994).
Laws, J., "Management Driven Reliability Philosophy," p. 9, Solutions, Vol. 5, No. 4 (Aug. 2010).
Laws, J., "Beyond Safety: Using Root Cause Analysis Tools to Improve All Your Business Processes," Proceedings, TapRoot Summit, San Antonio, Texas (2010).
Laws, J., "Equifactor and Process Troubleshooting: Managing Your Crucial Tribal Knowledge," Proceedings, TapRoot Summit, Nashville, Tenn. (2009).
Parmenter, D., "Key Performance Indicators: Developing Implementing and Using Winning KPIs," Wiley, Hoboken, N.J. (2007).
Price, B., "Error Reduction = Cost Reduction in Plant Processes," p. 1, Solutions, Vol. 4, No. 2 (2009)