Story Leadership Next Generation Leadership Advisors #HeadsUp Newsroom Careers
Real-time Case Studies
Natural Resources Mining + Metals Oil + Gas Private Equity Small Business Center Energy Building Materials Aerospace + Defense Industrials Agribusiness Automotive + Assembly Chemicals Consumer Packaged Goods Engineering Food + Beverage Industrial Goods Pharmaceuticals Private Equity Financials Services + Insurance Distribution + Logistics Technology Transportation Utilities
Operational Strategy Operational Improvement Supply Chain Maintenance Repair + Operations Working Capital Risk Management Customer Experience Sales Force + Support Pricing + Cash Transformation Design and Program Management Digital Transformation People Solutions™ Cultural Change Leadership Talent
Publications Videos Blog
SpinningWheels

Is your continuous improvement program stalled? Avoid these analytics mistakes

Implementing a continuous improvement program is an important step toward increasing productivity. Having your plant manager and executives look at data from your processes hourly gives you greater visibility of performance issues across your operation. This added insight is valuable, but is your team properly interpreting the data in order to know the right course of action for a given problem?

Statistics can be subjective. Even people trained in statistics and process control can easily misinterpret performance results, leading them to make a change in the process that actually does more harm than good. This is one of the reasons continuous improvement programs often end up spinning their wheels, expending significant time, effort, and money without making progress.

Neglecting variance

businessman hand working with new modern computer and business strategy as concept-1

A common mistake when interpreting performance data is to rely on averages without accounting for variance. Jumping to conclusions based on averages alone probably won’t solve your issue and will more likely make the problem worse.  You must analyze variation, not just the difference in sample averages, and understand what might be driving the variation before adjusting the process.

Interrelated variables

Statistical challenges also arise when processes have multiple inputs because there are relationships and interactions between the input variables. In these cases, changing one factor at a time will not lead to a solution because there are a combination of variables at play. Not only do you need to nail down the primary inputs and how they should be set but also control the interactions that are important to that process.

Issues like these can cause a continuous improvement program to go in circles, chasing symptoms without addressing the root cause of the issue. Proudfoot helps our clients perform a statistical scientific evaluation and assessment of complicated processes and make better use of data to ensure that changes are truly leading to improvement. Not only will we accelerate your continuous improvement program, we’ll drive the behavioral transformation that will ensure results are sustained.

Contact Proudfoot at 404-260-0600 or info@proudfoot.com to discuss ways to get more from your continuous improvement efforts and unlock the full potential of your operation.