What constitutes Good Performance in Services?

Following up on my previous blog entry on Performance Constraints here, I would like to follow up on some more ideas on how to do better Performance Optimization in Services.

Usually in Services, a number of Key Performance Indicators like Efficiency Metrics (Average Handle Time, Average Speed of Answer, Average Hold Time, Number of Mortgage Applications Processed in an Hour, Number of Collections Calls made, etc) and Effectiveness Metrics (Total Collections in a Collections Process Compared to Target Collections, Total Number of Telemarketing Calls made vs Number of Calls that resulted in a Qualified Lead, Customer Satisfaction Index), etc are routinely measured. In the Business Process Outsourcing context, some of these KPIs are also in the Service Level Agreements (SLAs) and are shared with the end client.

Often in Services, Efficiency Metrics and Effectiveness Metrics are opposed to each other as contrasted with Manufacturing where they seem to be less conflicting with each other. Faster manufacturing may quite often mean more output as long as the manufacturing process is stable and quality precautions are taken.

In Services, faster hanging up of phones as a goal often erodes Customer Satisfaction. When your doctor spends only 10 minutes with you as opposed to spending 20 minutes with you, your past history and talking to you about your ailment, the quality of care and your Customer Satisfaction Index invariably suffers.

When you are processing Orders in an Order Processing application, when your productivity metrics such as number of orders processed per hour goes up, your effectiveness factors such as Accuracy and Precision suffer and you make errors that may require rework and cost the company money!

If you look at Service Process Performance as a set of Performance Constraints and BANDS of values for KPIs are acceptable rather than just a single value, as long as they are compensated by better performance in another KPI like Customer Satisfaction, it should be Optimal Performance.For example, let’s say that Average Handle Time could be anywhere between 10 minutes and 15 minutes on a phone call AS LONG AS lower performance in AHT is compensated by higher Customer SatIsfaction Scores!

Or if you have 10 Key Performance Indicators in a Service Process, good performance can be defined by 6 our of 10 KPIs being in certain ranges rather than an individual value for each one.

Services seem to demand bit more sanity and thought into how to measure, and what constitutes good performance. Call center managers will tell you horror stories about how the KPI tails usually wags the Agent Performance Dog with very undesirable results. There is the story of agents suddenly in the middle of the call saying" Hello, Hello, I can’t hear you anymore" and then hanging up on customers. When they realize that the call has lasted more than 10 minutes and they will be in trouble with their own Average Handle Time metric, this has been known to happen!

Performance Constraints and Bands of Acceptable Performance may provide a better way of managing performance in a saner way and not let poor metrics design drive behavior into unacceptable territory, defeating the whole purpose of measurement!

For me, goals and daily metrics are the key to keeping me focused. If I don’t have access to the right stats, every day, it is so easy for me to move on mentally to the next thing. But if I have quick access to key metrics every day, my creativity stays within certain bounds–my ideas all center on how to achieve our goals - Paul Allen