Do not kill your business with KPIs

Metrics! KPIs! We love them!

But we need to know how to use them, or we will kill our businesses. Understanding the math behind them is not enough: we need to know how to combine the math with management systems and often, with good digital technologies.

“The road to hell is paved with good intentions”

One could argue that everything we do is always with the best of intentions, at least in relation to ourselves. Whether it’s good or bad depends on the analysis of the context and what the reference is.

If management sets quantitative goals and makes people’s work dependent on achieving them, they will probably do so – even if they have to destroy the company to do it.” – W. E. Deming, cited in “Eliminate Slogans, Exhortations and Targets” (2016) –

And that’s exactly what almost happened at a giant Brazilian industry with their equipment setup performance goal.

In this company, achieving the setup performance goal by operators, and performance in a few more indicators, resulted in a bonus at the end of each quarter. At the end of the year, there was even a second award, with an even better prize.

But after the beginning of this work mode, with this incentive, something “strange” started to happen:

  • More and more, the setups, previously problematic, always met the 95% performance target.

An interesting effect.

Was this a miracle?

The result of an extremely competent and dedicated team?

Was the leadership idea excellent?

Or was it a case of human beings doing what they do best: adapting to the system to survive in it?

The management team then began to evaluate whether the extraordinary gains were real or not, and there were no surprises. It was discovered that, in several cases, there were attempts to circumvent the metric. The operation was frequently indicating that the setup had been completed within the standard time, and the indication of problems was logged in another category of loss.”

Who was the Sherlock Holmes of the story?

The company, which chose Overall Equipment Effectiveness (OEE) as one of its indicators, realized that despite the miracle in setups, there was no increase in OEE.

OEE is an indicator generated from the multiplication of three indices – Availability, Operational Performance, and Quality – with the aim of demonstrating the percentage of the available time of a piece of equipment that is being used to produce value. Simply put, its calculation is as follows:

OEE = Availability x Performance x Quality


  • OEE: Overall Equipment Effectiveness
  • Availability: percentage of time the equipment is available for production
  • Performance: actual production speed compared to target speed
  • Quality: percentage of products that meet quality standards

As the problem notes were just changing index category within the OEE analysis, it was explicit that there was something wrong with the improvement in setup performance.

To quickly realize and resolve this issue, it was essential to have the support of the COGTIVE system.

The COGTIVE system’s setup performance indicator showed that something suspicious was happening.

What can you do to resolve (or avoid) a situation like this?

The core of the problem lies in the fact that the leadership was not able to see the trees and the forest, and only saw the branches: there was a lack of systemic thinking about the problem.

This led to the issue of setup performance being analyzed in a one-dimensional way, practiced through only one incentive, and without a system that supported the truly appropriate behavior and/or correctly verified the impact of one indicator on another.

But resolving this became much easier when they had the necessary information in the most practical way.

There were two changes:

  • With the COGTIVE reports in hand, managers started holding daily quick meetings with the team at the beginning of the day to discuss and evaluate the OEE performance and the setup situation. Incorporating this habit of using system reports drastically reduced improper recordings.

  • A second correction was made by changing the bonus method, by linking other indicators to the setup performance. One of them was the so-called Effective OEE without Setup – a difficult-to-calculate indicator, but one that COGTIVE can generate instantly. This indicator measures OEE without the impact of stops that do not depend exclusively on the equipment or operator (such as stops waiting for programming, stops waiting for previous or subsequent stages), making the measurement more consistent with the defined objective. This allows, for example, projecting how many setups would be expected within a certain volume and/or production mix demand, and creating an equivalent value between continuous lines and batch production lines.

As a complement, the reduction in setup time also became the responsibility of management, so that they could work on improving it together with the operation, which was committed to following standard times as a reference.

As you may have noticed, using the correct incentive with appropriate information was one of the key elements in enabling the team to improve how they intended to increase the performance of the setups. This was a case where we addressed only one of the needs of a specific activity type:

  • This effect may be happening in many other indicator management situations in your company.

And this consumes your time and your teams’ money.

So, don’t miss the chance to unlock the full potential of your industry with COGTIVE’s solution for maximum performance on the factory floor. Learn more about the solution by clicking here.

Want to stay up to date on the manufacturing industry and industry 4.0?

Sign up for our newsletter

Share this post

Ready to take the next step?
Contact us to schedule a meeting.