Dynamic Banding in Custom KPIs - Part 3 (Variables)

David Novick
Technical Writer
David_Novick
updated 4 mths ago

by David Novick, Pyramid Analytics Technical Writer

Part 3 in this blog series presents an example involving the use of variables for the definition of Custom KPI band endpoints. Before proceeding, we suggest you read through Part 1 and Part 2 in this series to ramp up on Dynamic Banding for Custom KPIs.

If you are new to BI Office variables, this blog helps you to get oriented quickly with variables. If you are already familiar with variables, please stick with us - you may pick up a few interesting tips & techniques along the way.

You can employ variables in BI Office to accomplish many different tasks. When defining Custom KPI banding endpoints, you can reference existing variables or create new ones.

How Variables Empower Your Custom KPIs

When using KPI gauges, the optimal display of data occurs when the KPI range bands match the source data in an efficient manner. Problems may arise when a KPI gauge is designed for certain data ranges and then those ranges shift at runtime - causing the KPI gauge to be much less efficient in displaying the data.

Let's consider a case where a KPI is designed to display "sales increases" in the vicinity of 90-110% and then the data shifts so that the sales increases are now in the 60-80% range. You can imagine the results - the gauge needle sits at dead bottom because the results are "off the map". But if you variabilize the range endpoints, you can adjust the KPI range bands at runtime by just clicking different variable values.

An Example of Variable-Driven KPIs

In this blog, we create four variables to provide runtime control over a Custom KPI gauge. These four variables help us to fine tune our KPI gauge display at runtime to match the current data. The four variables provide the following gauge functionality:

B1-START (40/60/80%) User can select 40/60/80 as band 1 start point.
B1-END (80/90%) User can select 80 or 90 as band 1 end point (and band 2 start point).
B2-END (110/120%) User can select 110 or 120 as band 2 end point (and band 3 start point).
B3-END (120/140/160%) User can select 120, 140 or 160 as band 3 end point.

When we create a new variable, we can assign it any desired name. Once we save the variable, we cannot change its name but we can still edit the variable definition.

Define KPI Endpoints with Variables

In the KPI setup dialog, we assign six band endpoints through the use of four variables, as shown below.

Create New Variables from Two Entry Points

There are two entry points for creating new variables, as described below.

Create Variables from Custom KPI Dialog

To create a new variable from within the Custom KPI dialog, we click one of the endpoints and select the New Variable command, as shown below. Once we have created a variable, it can be referenced for one or more end points.

Here we show how we created the B1-START variable.

Create Variables from Analytics Ribbon

We can also create a new variable from the Analytics ribbon as follows:

We open Analytics ribbon in Data Discovery.
We click Variables List icon.
We select Slider, Text Box or Buttons to create a variable (these are the three variables types).

View All Existing Variables

To view a list of all current variables:

We open Analytics ribbon in Data Discovery.
We click Variables List icon.
We select Variables List command.

We use the Variables List to view, edit and delete existing variables.

Adjust Runtime KPI Results using Variables

At runtime, we can make use of the four variables to control the KPI gauge display. In this example, we have designed all four variables as “button type” variables for easy single-click access. As we click on different values for each variable, the range bands are adjusted accordingly.

NOTE: The variable icon in BI Office is a purple lightning bolt.