It's All in the Numbers - KPI Best Practices

Forbes recently published their list of the top billionaires in the world. Fantasy sports leagues could not survive without statistics. Professional baseball has been forever altered by what is known as Sabermetrics. Every week the top performing theater movies and TV shows are shared in order of popularity. We are inundated with TV commercials extolling the miles per gallon a car can achieve or the tonnage a truck can pull.

Metrics. They are ubiquitous. A search within the Book section on Amazon.com using the word "metrics" results in over 17,000 choices. An equal amount appears when searching for "key performance indicators." By whatever name, I am willing to wager that everyone reading this article is at least aware of metrics/ Key Performance Indicators (KPIs) and accepts that there is value in a metrics-driven organizational structure.

With so much already written about the basics of metrics creation and implementation, this article will take a different approach. Using real life experiences, I will share some best practices that I believe are fundamentally critical to achieving success in the world of metrics.

Be True to the Numbers
My client, an aerospace and defense manufacturer, determined that keeping track of on-time delivery is an essential KPI. Every day the staff gathers to review open orders, to identify what can ship that day and what orders are likely to miss the promised shipping date.

At one meeting, it is clear that a particular order will not ship on the promised date. From a metrics perspective, it is either all or nothing. Meet the date, receive a 100% grade. Miss the date and it is a zero. Nothing in between. So imagine my astonishment when I heard the "relief" question, the manager asking the rep, "Can we call the customer and ask for some relief?"

If the customer is asked to extend the shipment date..."relief"...and agrees, and now the manager is able to meet the revised, extended ship date, does that now constitute an on-time shipment?

Not in my book. If the customer had requested the extension, then yes, it is timely. But this manipulation...there really is no other way to describe what was happening...is just not acceptable. Who is harmed here? Senior management for sure. Their reports will reflect on-time numbers that are artificially higher than reality. How can they properly manage with inaccurate data?

I cannot stress this enough. Live with reality. Do not fudge numbers to meet expectations. Doing so hurts everyone involved.

Sit in Your Customer's Chair
Once again on-time delivery is the KPI being measured. In this instance, my client, an audio video and computer rental company, regards success as making the delivery up to the minute requested by the customer. If the request is 10:00 am, then having the product on-site at 10:00 was considered OK.

Imagine you were the customer. It is 9:55 and the shipment has yet to arrive. Are you confident that your timetable will be met? Probably not. If me, I would be pacing the floor. I would be checking my watch. I would be distracted.

The key here is a question of perspective. From my client's point of view, having the shipment arrive at any point up to the requested delivery time was acceptable. From the customer's viewpoint, arriving at the exact moment desired was unacceptable. We changed the expectation moving forward. A successful delivery was now one that arrived no later than 10 minutes prior to the requested delivery time.

What a difference that made. Customers appreciated the new approach. Customer satisfaction grew. And on-time deliveries, even when using the new rule, spiked to record percentages. Always remember that KPIs cannot be myopic but rather must be viewed from your customers' perspective.

When Is Enough, Enough?
While working with another client distributor, the owner proudly directed me to his computer screen. "Look," as he pointed to the monitor. "These are all the metrics we're capturing." Was I to be impressed? The specially designed Dashboard was chalk full of numbers and graphs. Appealing to the eye, but really, how useful?

I believe that Dashboards are a useful tool for data collection. Research suggests that short-term memory only holds about seven chunks of information, and those fade from our brains in about 20 short seconds. A Dashboard overcomes this memory constraint. What it does not overcome is the feeling of being flooded with information. When is the sheer mass of data too much to effectively manage and respond to?

There are no hard and fast rules when it comes to the question of "how many?" And there is probably no "right answer" either. But in my experience, there is the "Rule of Three" that I have found to work quite nicely. (The United States Marines are big believers in the Rule of Three, especially when it comes to keeping people alive. The Corps apparently experimented with a rule of four, and retention and effectiveness took a nose dive.)

Back to my client. We discussed the merits of streamlining the Dashboard. I asked him to whittle down the vast number of metrics to the top three in each organizational category, for instance operations, production, employee performance, and then work with those KPIs during the coming weeks.

When we reconvened to review his experience, he offered this one word reaction, "Whew." It was much easier to manage and be productive with the reduced number of KPIs. More is not always better.

Unintended Consequences
KPIs evaluate the success of an organization or of a particular activity in which it engages. They are chosen...or should be...to support and promote the company's strategic goals and objectives. Most times the selected metrics are productive. However, where poorly structured, implemented or supervised, they can lead to unintended consequences detrimental to the organization.

One of the metrics established at a precision measuring tool manufacturer (worked there early in my career) was number of units produced per hour. Management felt that more could be accomplished by the staff if given the "motivation." So one day the workers on a particular production line were told to produce 20% more every hour. With no material changes in terms of equipment, training, procedures, the only way to accomplish the new objective was to simply work faster.

And they did. The numbers rose. But as the production increased, quality declined. As a result of creating a metric that was not achievable without other changes, training and support, the results were counterproductive. David Parmenter, self-described "King of KPIs," believes that "over half the measures in an organization may be encouraging unintended behavior."

Once it was recognized that the KPI was doing harm, the whole process was reevaluated. The expected increase was reduced to a more reasonable level. A Continuous Process Improvement initiative was triggered that identified ways to improve the processes leading to faster times. Fortunately, before too much harm was done, the problem was corrected.

Key Best Practices
There is much value in creating a culture that relies on KPIs/ metrics to measure organizational productivity and accomplishments. However, doing so is potentially fraught with danger, as we've seen through the various experiences shared above.

These key best practices will ensure that you and your organization will have a productive and successful KPI program.

Read this article on our website

View the newsletter archive

I welcome your comments at lee@schwartzpro.com.

You are welcome to forward Keys to Profitability to colleagues as long as you credit the Schwartz Profitability Group and its source at www.schwartzpro.com.

If this issue of Keys to Profitability was forwarded to you, you may begin getting your own by subscribing to this newsletter.