Defeating Uncertainty (in Demand Planning)

As part of my recent innovation week, I posted Measuring Innovation which provided you some metrics that you could use to measure your innovative progress – since you can’t manage what you can’t measure is as true with innovation as it is with any other business activity. But the importance of measurement goes deeper than you may recognize, as pointed out by recent articles in the Supply Chain Management Review and Knowledge @ Wharton.

According to Wharton, supply chain measurement is a mission critical element but many companies lag when it comes to measuring how well they are doing when implementing new supply chain initiatives. Considering that procurement and supply chain departments are under continual pressure to get better results without increased resources, it’s vital that you use metrics that identify how the strategic needs of the company are being met.

Wharton recommends using the “efficient frontier” to gage capability where you plot points along a trade-off curve between multiple the performance metrics and look for a position that protects your interests and those of your customers simultaneously. For you technical folks, you’re finding the optimal point on a multi-objective pareto curve based upon the relative weightings of the metrics you are using.

However, as the SCMR points out, there are many challenges in supply chain measurement that you have to solve to effectively manage your supply chain and find the optimal point on that multi-objective pareto curve. Old data, too many metrics, constantly changing metrics, and endless debate over metric definition are just some of the difficulties you need to overcome.

The key is to know what to measure, decide on some industry standard metrics to measure it, and have a program in place to measure it that focuses on quality and not quantity. This program should be based on best practices and avoid the common pitfalls of excellence addiction (constant improvement is good, but you need to take it one step at a time), missing data (the right stuff isn’t enough), ingrained inertia (resistance to change), and analysis paralysis (don’t overanalyze or fail to act on the results).

The metric definition best practices outlined in the article are great:

  • design different metric portfolios for different goals
  • keep it small (avoid the “mushroom effect”) as each portfolio must be of a manageable size
  • address the basics: balanced, cross functional, and practical (with respect to cost, quality, time, & effectiveness)
  • align execution and strategy
  • understand the interdependencies
  • balance the need for standards versus customization (every chain should measure demand-forecast accuracy, perfect order, total chain costs, and cash-to-cash cycle time)

Finally, remember to set targets, work hard to achieve them, and retain them once you have reached them.