Mark Twain said “There are lies, damned lies,…and then there are statistics.” In some instances he could have just as easily been talking about metrics.
We all understand the need for metrics when it comes to running a business. If you can’t keep score, how do you know if you are winning or losing? Just remember when you start basing compensation on these metrics that it is changing the game that is being played.
Here is a good case in point. A business I knew was running in the red. A look at the product prices (and costs) showed that it was almost 20% more expensive than the competition on a cost basis. However, based on the operational metrics they were running at the peak of efficiency (99+% on the production yield targets). How could this be?
A deeper dive into the metrics showed that over time the production yield targets had been lowered (to an 86% yield target!) so that the operational team could maximize their goal attainment and incentive compensation. They were actually achieving 99+% of an 86% target. The rest of the market was attaining true 99+% production yields. The incremental 14% disadvantage in production efficiency was the root cause of the product cost differential in the market.
Over time it had become easier for the operation to change the metric that it was measured (and paid) on, than it was to improve the process. This is obviously an extreme case, and it was a metric “creep” that had occurred over many years. But it does point out how a metric can affect how you look at a business’s performance. You can go from looking seriously at exiting a business because it is thought that it could not effectively compete, to looking at the root cause of the issue: how do you make the business more efficient and continue to compete.
Stephen, you are correct in your assessment and well said. Metrics, like any measurement of your business, must be aligned with the overall corporate strategy. The metrics and strategy must be reviewed outside of the department being measured and calculation of the metric should be handled from a position of verifiable data not submission from the department itself. It has been my experience that if departments can grade themselves (manipulate the result) then they usually find a way to put the best spin on the results.
Keep up the good work on the blogs. They are quite interesting.
I agree. I’ve seen this organizational run for cover in several companies I have worked for within the telecommunications sector, before the 2000 dot.com bubble burst. I wonder if the behavior continues in that industry and if it is prevalent elsewhere?