Correlations between your Performance Indicator Results

A good set of KPIs contains a mix of:

  • leading and lagging indicators.
  • indicators across the four perspectives of the Balance Scorecard.

There are implied relationships between these KPIs.

  • Leading and lagging: a change in a leading indicator should be followed by a corresponding change in a related lagging indicator.

    For example, an increase in marketing activity and/or sales enquiries (both leading indicators), should be followed by an increase in sales (a lagging indicator).

  • Balanced Scorecard: an improvement in a learning and growth indicator should lead to an improvement in one or more business process indicators, which should lead to an improvement in one or more customer indicators, which should lead to an improvement in one or more financial indicators.

    For example, an increase in machine learning skills (a learning and growth indicator) should lead to faster/more accurate credit decisions (both business process indicators) which should lead to more and more satisfied customers (both customer indicators) which should lead to an increase in revenue (a financial indicator).

A good strategy is based on these cause-and-effect relationships between different goals, objectives and indicators.

If you're not seeing those relationships in your performance indicator results, then your strategy is not working. You'll want to understand why and fix it.

But how do you see if they are working:

  1. You can do a visual scan on your Results Scorecard (click Results on the Main Menu).

    By default (you can change it) the summary charts for the quantitative Performance Indicators on your Scorecard all have the same time frame on the x-axis.

    That make it easy to visually scan up and down the scorecard to see if indicators are moving with each other.

    But this visual scan can be inconclusive and unreliable, so...

  2. can calculate the mathematical correlation between each performance indicator and each other performance indicator (click the Correlation button on the top right-hand corner of the Scorecard). measures correlation in terms of the R2 of the two data sets.

    "In statistics, the coefficient of determination, denoted R2 or r2 and pronounced "R squared", is the proportion of the variance in the dependent variable that is predictable from the independent variable(s)." (Source: Wikipedia)

    Not only will it calculate the R2 of each Performance Indicators results with each other Performance Indicators results, it will also calculate their R2 if we assume a 1, 2 or 3 month delay between the movement of one performance indicator and the other.

    The best fit for each pair of performance indicators is then presented in table format. Higher R2s, representing greater strengths of correlation, are shaded in darker shades of green. You can hover your mouse over any R2 to see a popup which also shows the number of months of delay (if any) assumed in the calculation.

    From this table, you should be able to uncover patterns of correlation that either support or don't support your assumed relationships between the Performance Indicators.

NB: Correlation does not prove causation. But it is a factor you might consider in your analysis.

If any part of this text is not clear to you, please contact our support team for assistance.