The Long Tail of Application Usage

A long tail is a distribution of some data point where a few data point have a high value, but most of the data points have a much lower value. When sorting the data points by value, there will be a large “head” and a long “tail”.

Application usage displays the “long tail” characteristic: A few screens are used much more than others.

Long tail of app usage

If you are re-developing your application, for example moving from Oracle Forms to ADF, focus on the head – the 10% of the application that is used 90% of the time.

For more graphics and discussions like this, subscribe to my weekly Technology That Fits newsletter.

2 thoughts on “The Long Tail of Application Usage

  1. Found your site when researching the cost of Oracle perpetual vs term licensing. In your article http://www.vesterli.com/2010/07/27/oracle-forever-re-consider-perpetual-licenses/, you assumed the user would renew the term license under the same terms (e.g., a 2-year term renewed under another 2-year term). In a more realistic scenario, a user tries to predict the expected useful life of the software; if they underestimated this duration, they would typically renew for shorter terms to cover the additional time needed. So, the user under a 5-year term might only need a couple of years more, and would renew under a 1- or 2-year term instead. That cost model changes the comparison dramatically.

    1. You’re correct that I present a simplified scenario. Under your more realistic scenarios, the organization would pay 90% for 5+1 and 105% for 5+2. Buying your own license does present the possibility of overpaying if you get the life of the software wrong – if you have great uncertainty, the option to buy Oracle cloud services might be the way to go today.

Leave a Reply