The data sets are ready to go, and this blog post marks the beginning of the third phase of the KPI assessment process, the Test Run. Congratulations on now finishing the first two phases (Project Plan and Data Retrieval). The first aspect of the test run is to do the initial run through. Here you will be using data from various data sets that you prepared during the data retrieval process. You will need to identify the needed data for each KPI. The initial run through will relay heavily on the use of business rules. A business rule is made up of formulas and procedures to complete a calculation. We have all been using business rules to solve things our entire life. Remember the math problem about installing a fence? Area and perimeter. Those are the first two formulas that I remember learning in school. Applying the formula was empowering. Being able to take numbers and generate another new value almost seemed like magic. It was also quite satisfying to solve a word problem that included certain pieces of information and then was prompted to find the solution. Key Performance Indicators (KPI) are like word problems. A more accurate label would be “word opportunities” that once explored will lead to a higher level of insight. Word problems that involved the concepts of area and perimeter were based on formulaic concepts that were specific to different families of shapes. Each of these shape families had specific business rules associated with how the values of area and perimeter are determined for them. Each KPI will have one or more business rules associated with them that outlines how the actual indicator is calculated. At the time of KPI development, best practice would involve the identification of the data needed to measure the KPI, the availability of that data, where the data will be housed, and a business rule that explains how to complete the calculation. If all of these elements are in place and you have discussed them with your team, you are ready to run your initial analysis procedure (and you can skip ahead to the last paragraph.) It is very common that getting the KPIs developed, agreed to, and data collection protocols developed and documented at the beginning of the year is a heavy lift and getting to develop the analysis business rules is left to another time. In order to move forward in the analytic process, that time has come. To begin the exploration of business rule construction, lets begin with an example KPI to work with. We will use our fictional company, Reusable Cup Coffee (see blog post #3 entitled, Be Part of a Team). One of Reusable’s KPIs for the 2017 performance year was the following: Beginning in the second quarter of 2017, all stores will have trained all team members on how to make the Squared Latte, such that 90% of store staff achieve the employment level of barista associate. A couple of notes about the example KPI:
Table 1: KPI Broken into parts with corresponding purpose to each part Now that the KPI has been broken into parts and a purpose of each part is stated, we can add one more column to the table that provides a data source description for the applicable parts of the KPI. Table 2: Table 1 with additional column, Data Source Description When reviewing Table 2, the right most column labeled “Data Source Description” has a note specific to the corresponding part for the KPI. When developing the business rule, begin with the algorithm. For this KPI: KPI Value (%)= (Number of Employees with “Barista Associate” as job title/Total Number of Store Employees) X (100) Simplified: KPI % = (B/A) X 100 KPI Value (%) meets KPI target in > 89%. After the algorithm has been reviewed, construct the data calculation set by implementing business rules. The data calculation set is the data that is necessary to answer the question. The following business rules will be used to create this calculation data set for our example KPI: ==Beginning of Business Rules
Let’s run through these business rules and try it out.
KPI % = .94
Once business rules have been developed for each KPI, you are all set to conduct the initial run through of calculating the KPI values. Don’t worry if you get weird looking results or encounter errors during the initial run through. Any of these special cases will be discussed a greater length in the next blog post. Blog Post #8 Question: How did the Test Run go? Blog Post #9 Preview: Pit Stop
0 Comments
Leave a Reply. 
AuthorThis blog is written by our founder/principal consultant Dr. Brandan Keaveny. Learn more about Dr. Keaveny here. ArchivesCategories
All

© 2018 Data Ethics LLC

