416.467.9100 | Dundas Data Visualization | Login
Welcome Guest
This is the support page for the legacy Dundas Dashboard application. For assistance with the current Dundas BI application please click here.
Dashboard v5.0

This site makes extensive use of JavaScript.

Please enable JavaScript in your web browser and reload the page before proceeding.

Defining States

Hide navigation
RSS
Modified on Wed, 22 May 2013 12:10 PM Categorized as KPIs
Click to return to: Documentation | Defining KPIs


Overview

For a given metric (⪚ a measure), you can divide the entire range of possible metric values into a set of mutually exclusive sub-ranges or states. This set of mutually exclusive states is called a state group. For example, you can define a state group consisting of:

  • a Bad state representing metric values from 0 to 99
  • a Good state representing metric values from 100 to +Infinity.

By defining states and state groups for your KPI (or dataset), you can drive certain data visualizations on your dashboard, such as a state indicator control.

A state group must have at least one valid state defined. You can also add more than one state group to your KPI (or dataset) but each state group will be evaluated independently.

To manage states and state groups for your KPI (or dataset), use the State Groups tab in the KPI/Dataset Designer. This tab displays two grids related to states:

  • The State Groups grid shows the state groups that are currently defined for the KPI/dataset.
  • The States in Group grid shows the states that belong to the currently selected state group.

The State Groups tab in the KPI Designer.

The State Groups tab in the KPI Designer.


As an example, the chart in the following figure displays a KPI with two states defined. The background of the chart has a different color based on the state (⪚ the Good state covering the range from 2000 to +Infinity is displayed in green using a chart strip line).

A chart displaying a KPI with two states.

A chart displaying a KPI with two states.


Managing state groups

Adding a state group

To add a new state group:

  1. Click the Add State Group button which is located above the State Groups grid. The State Group Definition wizard is displayed.
  2. In the first step of the wizard:
    • Enter a name for the state group in the Name box.
    • Enter a description for the state group in the Description box.
    • Choose a measure/contextual metric/data column from your KPI/dataset to analyze.
    • Click Next.

State Group Definition wizard - Step 1.

State Group Definition wizard - Step 1.


In the second step of the wizard:

  1. Add the first state to the state group by clicking the Add State button. A new state (&ie; row) appears in the grid. Configure the state as follows:
    • Click the cell in the Label column and enter a name for the state.
    • Enter a lower boundary value for the state in the second column of the grid. By default, this will be a static (or fixed) boundary. You can enter a numeric value directly or use the token menu on the right to set the lower boundary value to -Infinity.
    • Click the cell in the third column to set the lower bound comparison operator.
    • Click the cell in the fifth column to set the upper bound comparison operator.
    • Enter the upper boundary value for the state. By default, this will also be a static boundary. You can enter a numeric value directly or use the token menu on the right to set the upper boundary value to +Infinity.
    • Click the cell in the Description column and enter a description for the state.
  2. Repeat the previous step to add more states to the state group as needed.
  3. If you want to delete a selected state, click the Delete State button.
  4. Click Finish. The State Group Definition wizard closes and the state-related grids in the State Groups tab are updated with your changes.

State Group Definition wizard - Step 2.

State Group Definition wizard - Step 2.


You can now preview your KPI or dataset to see the state evaluations displayed as an extra column:

KPI Preview showing states.

KPI Preview showing states.


Editing a state group

To edit a state group:

  1. Locate the row in the State Groups grid that corresponds to the state group you want to modify.
  2. Click the Edit button in the Action grid column. The State Group Definition wizard is displayed.
    Edit state group.
  3. Follow the steps as shown in the section, Adding a state group.

Removing a state group

To remove a state group:

  1. Locate the row in the State Groups grid that corresponds to the state group you want to remove.
  2. Click the Remove button in the Action grid column. When prompted to confirm the operation, click Yes. The state group is removed from the grid.

Ordering states

The order of the states that you define in the second step of the State Group Definition wizard is important. States that you consider to be the “worst” should be positioned at the top of the order and states that you consider to be the “best” should come last in the order. Certain DV controls and mobile apps will use this ordering by default when displaying states.

To adjust the order of your states, use the four reordering buttons in the top-right corner of the wizard:

Reordering a selected state.

Reordering a selected state.


Dynamic states

Dynamic state boundaries

In the previous example, the lower and upper boundaries of each KPI state were defined as static numeric values (⪚ 2000), or set to a special value like +Infinity. But in Dundas Dashboard version 4 or later, you have the option of defining a dynamic state boundary that is based on a metric, contextual metric, or calculated value column from your KPI/dataset.

As an example, consider a KPI that is defined as follows:

  • KPI name - Sales By Date
  • Measure - Sales
  • Contextual metric - Target Sales
  • Instant dimension - Date

Target Sales is a contextual metric which is based on historical performance. Its value varies throughout the year. A chart displaying the KPI and its contextual metric is shown below:

A chart showing a KPI with a contextual metric (Target Sales).

A chart showing a KPI with a contextual metric (Target Sales).


For this example, instead of setting up KPI states with static boundaries, the goal is to use the Target Sales contextual metric to define the boundaries dynamically. Specifically, two states are to be defined:

  • Bad state - This is when the Sales value is less than the Target Sales value.
  • Good state - This is when the Sales value is greater than or equal to the Target Sales value.

To set up the state group for this example:

  1. In the first step of the State Group Definition wizard, set the Measure/metric column to Sales, and then click Next.
    Set the Measure/metric column to Sales.
  2. In the second step of the wizard, add a new state. Set the Label of this state to Bad. Click the right Constant value boundary button in the State Condition column. This button works like a toggle; when you click it, the upper boundary control turns into a dropdown list. Set the dropdown list to Target Sales.
    Add a Bad state.
  3. Add a second state. Set the Label of this state to Good. Click the left Constant value boundary button in the State Condition column. The lower boundary control turns into a dropdown list. Set the dropdown list to Target Sales.
    Add a Good state.
  4. Save, and then preview the resulting KPI. Observe that the state evaluations match the chart shown previously.

Preview of KPI with dynamic state boundaries.

Preview of KPI with dynamic state boundaries.


Displaying dynamic states

Chart

When a KPI that uses static state boundaries is added to a chart, the chart automatically creates striplines to delineate the different states (as shown in the Overview section).

However, if the KPI uses dynamic state boundaries, no striplines are created. Instead, the data points of the chart series may be given a different appearance depending on the corresponding state evaluation. This is achieved via the State Styles collection property of a chart series. One state style is created automatically for each possible state in a state group.

Displaying a KPI with dynamic state boundaries on a chart.

Displaying a KPI with dynamic state boundaries on a chart.


The following steps show how to set up the above chart.

  1. Create a new dashboard.
  2. Add the Sales By Date KPI from the previous example to the canvas. The KPI turns into a bar chart with two series (&ie; the Sales metric and the Target Sales contextual metric).
  3. Edit the series corresponding to the Target Sales contextual metric:
    • Change its Chart Type to Line.
    • Set its Fill color to orange.
  4. Add a legend to the canvas and set the bar chart's Legend property to Legend1.
  5. Edit the series corresponding to the Sales metric and set its advanced property, Type of Items Displayed, to None.
  6. Edit each of the state styles for the series corresponding to the Sales metric and set the Displayed in Legend property to True.
  7. Save, and then preview the dashboard.

Data Grid

The following figure shows how the same KPI is displayed in a data grid (by defaulting all options in the data grid wizard):

Displaying a KPI with dynamic state boundaries using a data grid.

Displaying a KPI with dynamic state boundaries using a data grid.


Map

The following example shows how to use dynamic states to compare population changes on a map.

First, prepare a virtual table that contains some population change data. Call it StatePopulationVT.

Preview of StatePopulationVT virtual table.

Preview of StatePopulationVT virtual table.


Next, create a new dataset, StatePopulation, from the virtual table and define a state group for it as follows:

  1. In the first step of the State Group Definition wizard, set the Measure/metric column to Pop2010. Click Next.
    State group definition wizard - Step 1.
  2. In the second step of the wizard, add a new state with these settings:
    • Label: Pop Decrease
    • Upper boundary: Pop2000 (dynamic)
  3. Add a second state with these settings:
    • Label: Pop Increase
    • Lower boundary: Pop2000 (dynamic)
      State group definition wizard - Step 2.
  4. Click Finish.
  5. Save, and then preview the dataset to see the dynamic state evaluations.
    Preview dataset with dynamic states.

Finally, create a new dashboard and add a map control to it:

  1. Select the prepopulated map resource, United States (Mainland Only).
    Prepopulated map resource.
  2. Add the StatePopulation dataset to the map control.
  3. In Step 1 of the the Map Data Binding Setup wizard, select “Change the appearance of Shapes, Paths or Symbols”.
  4. In Step 2 of the wizard, select State as the column containing the names of map shapes. Select Pop2010 as the column containing metric values for the map shapes.
    Step 2 of map wizard.
  5. In Step 3 of the wizard, select Shapes for binding, and select Color as the property that will be adjusted.
  6. In Step 4 of the wizard, select State Group 1 in the KPI state groups dropdown list.
    Step 4 of map wizard.

You can now save and preview the dashboard to see the map colored to show the dynamic state evaluations.

Map showing areas of population increase and decrease.

Map showing areas of population increase and decrease.


Note: There is a limitation with using dynamic states on map. For dynamic states where two columns are compared, the map uses the state index as the metric value and defines the map states as 0-0, 1-1, 2-2, &etc; This means that the Y values will not be usable as tooltips because they won't be the actual Y values.

Linking state boundaries to hierarchy levels

If your KPI uses a rendered/visible dimension that has a hierarchy (⪚ a time dimension), while defining states for a state group, you will have the option of specifying different state boundary values depending on the dimension grain (&ie; hierarchy level). For example:

  • When the grain is Calendar Month, use state boundaries based on a monthly sales target of 5 million dollars.
  • When the grain is Calendar Year, use state boundaries based on a yearly sales target of 60 million dollars.

Linking state boundaries to hierarchy level.

Linking state boundaries to hierarchy level.


To link to hierarchy levels:

  1. Select the Link to specific hierarchy levels check box. The wizard expands to show an area with three buttons on the right.
  2. Define the default boundary values for your states as shown previously. You can use static boundary values or a dynamic value such as a contextual metric. These default boundary values will apply to all hierarchy levels unless customized (in a later step). Note that if you choose a dynamic boundary value, you will not be able to customize that boundary value for each level.
  3. Click the Cycle dimension hierarchy grains button (&ie; the middle one on the right) to display a specific hierarchy level, such as Calendar Year.
    Cycle through dimension grains.
  4. Customize the state boundary values for this grain as desired. If you want to reset the boundary values back to default, click the Revert to default boundary values button (&ie; the third button on the right). In this case, each boundary value will be reset to the value, Default.
  5. Cycle through the remaining hierarchy levels (using the middle button) and customize the boundary values for each as desired.
  6. Click Finish.

Next, add this KPI to your dashboard in the form of a chart, and then add a dashboard grain parameter. Observe how the state boundary automatically changes whenever the grain/level is changed.

When grain is Calendar Month, state boundary is 5 million dollars.

When grain is Calendar Month, state boundary is 5 million dollars.


When grain is Calendar Year, state boundary changes automatically to 60 million dollars.

When grain is Calendar Year, state boundary changes automatically to 60 million dollars.


Related topics


Click to return to: Documentation | Defining KPIs

About Dundas | Contact Us Follow us on Twitter! | Privacy Statement | Report Site Issues

Copyright © 2009-2014 Dundas Data Visualization, Inc.