Skip to main content
How to create a custom KPI?
Updated over 2 years ago

INSIA is a No-code Business Intelligence and Analytics platform. We want business users to analyze data directly the way they want without depending on any technical support.

We all know that no matter how detailed your source dataset is, we will always need to create derived KPIs. But this could be daunting if you have to join columns, apply different conditions, and perform other mathematical calculations through SQL or coding.

To live up to our vision of simplifying analytics, INSIA comes with a No-Code formula builder.

Check out our series of three detailed video tutorials on how you can create complex formulae and KPIs for analysis without writing a single line of code.

Video 1 - Basic KPI Builder (1)

In this tutorial video, you will learn about building basic KPIs, which are -

  • Some aggregation (sum, count, average, etc.) of any column of the loaded data table.

  • Some aggregation (sum, count, average, etc.) of a new column derived from any two or more columns of the loaded data table.

Video 2 - Basic KPI Builder (2)

In this tutorial video, you will learn about building another set of Basic KPIs, which are -

  • Counting any column (mostly String/Text columns) of the loaded data table.

  • A ratio of aggregation of (sum, count, average, etc.) of two or more columns from the loaded data table.

Video 3 - Advanced KPI Builder

In this tutorial video, you will learn about building another set of Basic KPIs, which are -

  • A ratio of some aggregation of (sum, count, average, etc.) of two or more columns from the loaded data table where both numerator and denominator have different where clause condition

Elements of Formula Builder

This article will explain each component that makes INSIA's no-code formula builder.

Here are the different components of the KPI builder:

  • Table Name

  • Display Name

  • Basic Formula Builder

  • Time Dimension

  • Units

  • Time Conditions

  • Null/ Not Null Values

  • Priority

Table Name:

This represents the name of the table from which the KPI belongs. Select the table from which you want the KPI to be built.

Display Name:

Display Name is the KPI's name on the search bar. You can access data associated with this particular KPI or Dimension using this name on the search bar.
โ€‹

E.g., If the KPI you intend to build is related to 'Sales,' then you can name the display name 'Sales' or 'Revenue' - as you like.

Basic Formula Builder:

The basic formula builder section of the formula builder specifies how a KPI/metric or dimension is calculated.

This section is divided into two parts, as you can see above.

On the left side, you have a series of different formula types. To create any KPI, you will have to select from the list on the left. The entire structure will get populated on the right as soon as you choose one.

Once the structure is populated on the right, you must select different parameters depending on the KPI formula structure you have chosen.

Aggregate

In the aggregate drop-down, the user can select different aggregate functions. They can choose from the following -

  • AVG: Applies an average mathematical function on the column that follows

  • SUM: Sums values in the column that follows

  • COUNT: Counts the values in the column that follows

Add Column

Add column drop-down follows up Aggregate function generally. The drop-down will intelligently show the relevant column headers depending on your selected aggregate function.

For Example, if you selected SUM as the aggregate function, you would see only those column headers with FLOAT or INT as the data type.

Similarly, if you selected COUNT as the aggregate function, you would see column headers with STRING as the data type.

Math operator

The math operator drop-down has the following different math functions -

  • + Use this operator if you want to add two different column values

  • - Use this operator if you want to subtract two different column values

  • * Use this operator if you want to multiply two different column values

  • / Use this operator if you want to divide two different column values

Where condition

If building the KPI requires a where condition, the "where" button on the formula builder can be used.

Once clicked, the user needs to select the particular column on which he wants to put the condition from the Add column.

INSIA will intelligently show either dimension values or a form field depending on what type of "Add Column" the user has selected.

For Example, if the user selects a TEXT column such as Billing Type, it will show Dimensions drop-down with different dimension values of the selected column.

But if the user selects an INT/ FLOAT column such as Invoice Quantity, it will automatically show a form field where the user can enter any value.

Concat Function

Concat drop-down can be used to add multiple conditions on a 'where' clause. Users can select between OR and AND.

Constant

Once activated, the Constant function can help the user add, subtract, multiply and divide a constant within the formula.

Time Dimension

The time dimension drop-down gives all the different Date-Columns in the table. The date column selected would be used for any time-related analysis or applying any time filter on the built KPI. To elaborate, when the user selects the KPI on the search bar and applies the day cut or previous quarter filter, the time filer would apply on this Time Dimension.

Units

Users can select the Unit of Measure. (UOM) for the KPI getting built. This unit will show on the top left below the search bar and above the search container whenever this KPI is selected.

Null Values

Users can select any specific column from the drop-down to populate Null values if the KPI getting built requires a particular column to be NULL (have no values or is empty)

Not Null Values

Users can select any specific column from the drop-down to populate Null values if the KPI getting built requires a particular column to be NOT NULL (have some values)

Time Condition

The Time Condition function enables the KPI's Ageing / Early/ On-time and Late analysis.

The Current or Actual Date drop-down is to select the Actual Date Column for Early, On-time, and Late analysis.

The target Date drop-down is to select the Target Date column for Ageing, Due or Overdue analysis

Priority

Users can set the importance of the KPI

  • Show will enable the KPI to be visible on the search bar.

  • Hide will ensure the KPI is not visible on the search bar.

  • Love will ensure the KPI is given importance on multiple analyses while being visible on the search bar.

Read next:

Did this answer your question?