Benchmarking, Analytics and the Quest for Enlightenment

March 4, 2013

In the movie Jerry Maguire, Cuba Gooding Jr. frequently reminded Tom Cruise to “show me the money“. While the physical similarities between Tom Cruise and myself are no doubt apparent, the reason for mentioning this in my blog is to paraphrase Mr. Gooding and say “show me the data“.

At Netsmart, we’re all about data, measurement, benchmarking, analytics, and knitting (Okay, one of these isn’t true). Our zest for data can sometimes lead to confusion. For example, we have heard some express a lack of clarity about the difference between benchmarking and analytics.

Benchmarking compares our performance to others or to ourselves over different time periods. Knowledge gleaned from benchmarking is often aggregated data that answers the question “what”. It highlights the differences between others, and us and calls attention to those differences. In doing so, it brings our own comparative performance to our attention. If our performance is worse than a comparison group, we know only that there is a problem, but we don’t know why it is a problem.

One way to determine why something is a problem is to perform higher order analytics on the data. That is, to “drill down” into the data, often at a transaction level to dissect the problem at a more granular level. For example, benchmarking might disclose the average length of stay in my organization’s outpatient department is 50% higher than the average of all other similar organizations. I now know what the problem is, but not until I drill down into the data will I discover that one clinician is responsible for pushing up the average for my outpatient department. Once I discovered that fact, I now know why the problem is occurring and I can do something about it.

Analytics isn’t the only tool we have for uncovering the “whys”. Dr. Paul Lefkovitz, general manager of Netsmart’s benchmarking, has developed another very clever way to find out why scenarios occur. It’s called Process Benchmarking and it involves differentiating what high performers do compared to low performers. Imagine your organization has a much higher no-show rate than your peers. Paul’s Process Benchmarking attempts to determine what it is that the first group does that the second group does not. In examining the data for these activities, trends emerge that differentiate a high performer from low performers and suggest ways for poor performers to improve. Naturally, the results aren’t 100% predictive, but they do provide actionable ideas that have a high probability of success and represent a sort of “best practice”.

As so, analytics and process benchmarking are two techniques that capitalize on the knowledge gleaned from benchmarking and take it to the next level. These tools allow organizations to fix the problems benchmarking discovered. Please contact us for more information on these services.