Adding derived metrics to an analysis

You can create a new metric while you are viewing a Visual Insight analysis. For example, you can subtract the values of one metric from the values of another metric, such as Revenue Forecast - Revenue. You can assign a number to each value of a metric, to rank its value from least to greatest, or display metric values as percentages of an accumulated total.

Derived metrics are metrics that you can create based on existing metrics in the analysis. A derived metric performs a calculation on-the-fly with the data available on an analysis, without re-executing the analysis against the data source. Derived metrics are metrics that are saved and displayed only on the specific analysis on which they are created.

You can only use metrics already displayed in the analysis to create a derived metric. For background information on derived metrics, see About derived metrics.

Prerequisite

Steps are below to create each type of derived metric in an analysis.

Task

Steps

Create a new metric by changing the aggregation function used to calculate values in the metric

  1. From the Drop Zones panel, hover the cursor over the name of the metric you want to use to create the derived metric, then click the arrow icon on the right. A list of options is displayed.

  2. Point to Aggregation, then select the aggregation function you want to use to calculate metric values, as follows:

    • To create the metric by calculating the sum, select Sum.

    • To create the metric by performing an average calculation, select Average.

    • To create the metric by performing a maximum calculation, select Maximum.

    • To create the metric by performing a minimum calculation, select Minimum.

    • To create the metric by performing a count calculation, select Count.

    • To create the metric by performing a geometric calculation, select Geometric Mean.

    • To create the metric by performing a median calculation, select Median.

    • To create the metric by performing a mode calculation, select Mode.

    • To create the metric by performing a product calculation, select Product.

    • To create the metric by performing a standard deviation calculation, select Standard deviation.

    • To create the metric by performing a variance calculation, select Variance.

Create a new metric by assigning a numeric rank to each value in a metric

You can choose to break the rankings by a specific attribute. For example, a Grid visualization displays revenue data for each Call Center in a Region.

If you choose to break the data by Region, the metric values for each Call Center will be assigned a number based on each value’s rank within its region. If you choose to break the data by Country, the metric values for each Call Center will be assigned a number based on its rank within its country.

 

  1. From the Drop Zones panel, hover the cursor over the name of the metric you want to use to create the derived metric, then click the arrow icon on the right. A list of options is displayed.

  2. Point to Rank, then select one of the following:

    • To display the metric values by rank in ascending order, select Ascending.

    • To display the metric values by rank in descending order, select Descending.

  3. Under Break By, select the attribute you want to use to break the ranking data.

  4. Click OK.

 

Create a new metric by displaying metric values as percentages of a cumulative total

  1. Hover the cursor over the name of the metric you want to use to create the derived metric, then click the arrow icon on the right. A list of options is displayed.

  2. Point to Percent to Total, then select one of the following:

    • To display the metric values as percentages of an accumulated row total, select Over Rows.

    • To display the metric values as percentages of a grand total, select Grand Total.

    • To display the metric values as percentages of a cumulative total for each element of an attribute, select the name of the attribute.

  3. Click OK.

Create a new metric by combining the values of two or more metrics.

  1. Press CTRL and click the names of each metric you want to combine. Hover the cursor over the name of the last metric you selected, then click the arrow icon on the right. A list of options is displayed.

  2. Point to Insert Metric, then select the function you want to use to combine the metrics. The options are:

    Note
    : Some of the options below are only available if you have selected two metrics.

    • To add the metrics together, select Add.

    • To calculate the average of the selected metrics, select Average.

    • To calculate the maximum across each of the selected metrics, select Greatest.

    • To calculate the minimum across each of the selected metrics, select Least.

    • To calculate the product of the selected metrics, select Multiply.

    • To divide the first metric you selected by the second, select Metric1/Metric2.

    • To divide the second metric you selected by the first, select Metric2/Metric1.

    • To subtract the second metric you selected from the first, select Metric1-Metric2.

    • To subtract the first metric you selected from the second, select Metric2-Metric1.

    • To calculate the difference between the metrics as a percentage of the second metric you selected, select (Metric1-Metric2)/Metric2.

    • To calculate the difference between the metrics as a percentage of the first metric you selected, select (Metric2-Metric1)/Metric1.

Create a new metric based on an attribute

  1. Hover the cursor over the name of the attribute you want to use to create the derived metric, then click the arrow icon on the right. A list of options is displayed.

  2. Point to Insert Metric, then select one of the following:

    • To perform a count calculation on each element in the attribute, select Count.

    • To calculate the maximum for each element in the attribute, select Max.

    • To calculate the minimum for each element in the attribute, select Min.

    • To calculate the sum for each element in the attribute, select Sum.

    • To calculate the average for each element in the attribute, select Avg.

Related topic