James Taylor's Decision Management

I saw this article on dashboards - Dashboards Help Align Operations with Business Goals - by Raul Gonzalez today and it made me think about the value of a dashboard to front-end, operational staff. Raul says in the article:

"The concept of the dashboard is taken from the car dashboard, which provides critical information about the operations of the car, allowing the driver to make the right decisions related to driving the car, such as to accelerate or decelerate, or to stop for gas."

This is the classic example used to explain the purpose of the dashboard. When I think back to my early cars, the decisions I took about driving the car included things like manually controlling the choke. Today cars automate many of these decisions based on rules developed by experts and models that turn data coming from various monitors into predictions. The dashboard no longer includes the information relevant to the choke.

When we think about operational decisions, we should think about a modern car, or perhaps even a future car, rather than a past one. Instead of telling people about a problem, we should do something about it and report how well we did. Instead of tracking things, we should use this information to improve things. We must consider if the right thing to do is to automate a decision, not just report information to allow someone to make that decisions. This is the essential approach in decision management - automate and improve over time the operational decisions that drive the business so that business people can control the strategy not have to deal with the details of each transaction. Raul uses a good example:

"For example, one possible KPI for a customer service call center may be the average time a customer waits for his or her call to be answered by a representative. The desired goal may be a customer wait time of less than 60 seconds. The current average wait time compared with the goal can be shown on a dial with values between 0 and 120 seconds. Values above the goal of 60 are coded in red, values between 45 and 60 are coded in yellow, and values between 0 and 45 are coded in green. The color-coded background provides users with an immediate picture of the Customer Service performance relative to the specified goal. A second indicator in a customer service dashboard may show the directional trend of customer wait time - whether it is increasing, or decreasing. "

Now I would see this problem differently, when it comes to operational decisions. I can use the information to predict the likely wait time for new callers and then use rules to perhaps change the message or IVR structure, rout calls to different desks if some are busier than others, display a "hurry up" prompt for those on calls taking a long time, turn off the cross-sell/up-sell script for reps or otherwise deal with the problem. Now the operational staff don't need to see a dashboard as their environment adapts. The managers still need a dashboard but they can control strategy, not the details or each decision, using it.

Dashboards are good for managers. Decision automation is often better for operations.

Technorati Tags: analytics, business agility, business intelligence, business rules, decision automation, operational BI, dashboard