To optimize or not to optimize, that is the question
Recently, I had a conversation around optimization for a banking customer — specifically, about whether they could optimize a particular activity. It sparked an interesting debate: What is optimization, and where can it be applied? This post offers some ideas about where and when to apply optimization. This discussion cuts across industries and is a valuable tool that can help companies drive down their costs in a challenging economic environment.
First, optimization is a term that is used so often that its impact and meaning have been diluted. In industrial engineering, it has a very precise meaning. For our purposes, we are going to define optimization as the improvement of an activity that produces specific and measurable business benefit. So, minimizing windshield time for a field service technician would be an example of optimization, as that would produce measurable cost savings. By contrast, optimizing the number of people in a lobby would not.
With that foundation behind us, let’s get into optimization.
There are six conditions that must be present to employ optimization. Those are:
- Pre–or post–sale. If it’s before the sale, you’re working with demand and controlling customer behavior. If the activity is after the sale, then you’re working with fulfillment and managing delivery or service activity.
- Constrained asset. Anything in limited supply would fall into this category: To use an airline example, a constrained asset would be a seat on a given flight. Appointments with key personnel would be another example.
- A way to attach a value to the constrained asset. Airlines use terms and conditions to stratify seats on flights to appeal to different buyers. A field service visit will typically have a defined cost.
- A way to compel or influence behavior. This may be the most controversial, but it is also essential. Unless I can change either the buying or service behavior, there is no way to optimize.
- Lack of choice. If there is only one choice, there is no value for optimization. Opportunities to optimize increase with the number of choices.
- Goal or objective. Identify exactly what you want to optimize that someone else will recognize as valuable.
In the airline case, availability in different fare classes is used to influence buying behavior. If you take the late flight, they will reduce the price of the ticket by $200. I recall a friend many years ago, who, after evaluating an Italian airline’s ticketing process, told them that he bet they had a huge number of last-minute passengers. They agreed and wondered how he knew. He said it was because they had no advanced purchase fares. This is an example of influencing demand-side behavior to optimize revenue attainment.
On the service side, if you can control the assignments of a field service worker, then you can tell them to visit one customer as opposed to another. That ability to control workers’ schedules is what allows them to wring operational efficiencies—and hence costs—out of the service delivery process. There are many other ways to wring value, and these kinds of outcomes are also optimizations to reduce costs.
Unless your situation meets the above conditions, it is not a suitable candidate for optimization, regardless of the worthiness of the activity. There may be other things to apply, but optimization would lead to confusion as well as failure to achieve your overall goal.
