Tuesday 21 July 2020

Understanding KPIs in Microsoft Power BI

Amongst the many different visuals you can select in Power BI, there is a KPI visual.

 

Whilst it is certainly not the only visual available for measuring important indicators, you should consider it when reporting on KPIs. 

So what is a KPI? A Key Performance Indicator, or KPI, is an important indicator, or value, that tells you whether or not you are on target to achieve something. Microsoft says that the KPI visual is designed to answer the question “Am I ahead or behind my target?” It is not a complex idea, although implementing KPIs so they are really effective takes some skill.

The name says it all:

Key = important.
Performance = the thing that needs to be happen.
Indicator = the numeric value to measure and compare.

The numeric value can be anything that you can measure, such as:

  • Money – sales, profit, etc.
  • Production output – number of widgets produced, number of articles published, etc.
  • Accidents – ideally none.
  • Patients – admitted, readmitted, recovered, died.
  • Training sessions attended.

KPIs are often lead indicators. That is to say, they are not lag indicators.

Lag indicators are the result of something happening, whereas lead indicators are the actions that lead to the result and the lag indicator. That’s complicated to explain, but easy with examples.

For instance, better trained people are likely to sell more, so if you want to increase sales, measure the number of training courses attended by your salesforce. If you want to lose weight, the amount you weigh is the lag indicator, but calories consumed or time spent exercising are lead indicators.

Lead indicators represent the things we can actually influence, which is why they make good KPIs.

Ideally, it must be something that you can measure at internals and compare against a target value.

Now let's look at the actual visual in Power BI. The KPI visual has three parts:

  1. Indicator – This is the actual value. You could think of this as the progress to date.
  2. Trend axis – this is a time period, eg day or month.
  3. Target goals - the indicator goal you are measuring against. The Indicator and Target goals must be the same, ie both must be currency, or both must be patients recovered. You cannot have a target goal expressed in monetary value and the indicator expressed as number of patients.  

This is an example of the KPI visual measuring daily output. Output could be anything from widgets to surgical operations, it is just designed to illustrate the point.



What does the visual show? Several things:

  • On the last day of data - 20th July 2020 - the output was 271.
  • On the last day of data - 20th July 2020 - output exceeded the goal by 8.4%. In this visual, the goal is not shown, but we could choose to display it. 
  • The KPI visual is in green, with a tick, showing that on the 20th July, we are on target.

A common problem is that your data doesn’t include a target. For example, you may be able to link to a database that stores data about output, but it does not include a goal. That’s not a problem for Power BI because you can enter goal data in an Excel spreadsheet and link the two sets of data together.

Is this the only way to show a KPI in Power BI?

No – and it isn’t always the best way depending on your requirements. The following illustration compares the KPI visual with a bar chart, a line chart, and a table with conditional formatting. All visuals are showing the same data.


If you were Production Manager, which way would you like to have this data presented? Which is the least informative?

If you would like to learn more about monitoring KPIs using Power BI in your business, do get in touch.