Dashboard Tips: Spotting Overlooked KPI Trends

Last Updated: Wednesday, January 23, 2013 by Johnathan Briggs

Imagine that your company has reams of data from its important KPIs. Management reporting software can take that data, display it in a way which managers can use to find trends, and allow them to take meaningful action. Although performance measurement often depends on finding trends, some of which are as easy as charting a measure with its target, managers can find it hard to pinpoint meaningful, actionable trends in their data, which results in opportunities being missed, or actions being taken when it’s too late.

With a little bit of creativity when using your dashboard software, it is easy to get around this problem. KPI dashboard software often includes ways to quickly manipulate your data, from options within specific chart choices to combining several charts into one. In this post, I will share with you 4 tips for making your KPIs trends far easier to understand.

1. Combining multiple KPIs

As I discussed in a previous post on how to best visualise measures and trends , individual trends can be easily seen using line charts, area charts and other chart types where the shape of the upper line is easily seen. However, charting only a single KPI on a chart might not give you all the information relevant to that KPI, if you see where I’m going.

The performances of various KPIs are often interlinked. Therefore, combined charts often yield more meaningful trends than single-KPI charts. For example, suppose that the manager of a publishing company wants to gauge the progress of a new software magazine. To do this, he would most likely measure the number of subscribers each month, as below.

https://www.targetdashboard.com/nlimages/magsalesSub.png

Looks like everything’s going well, right? Furthermore, if you were to look at this chart for the same magazine’s raw sales (i.e. number of copies sold in shops), you can see that they also look pretty healthy.

https://www.targetdashboard.com/nlimages/magsalesRaw.png

However, by combining both into one chart, you can see that raw sales of magazines are rising a lot slower every month than subscriptions. Management reporting software such as Target Dashboard usually allows this information to be combined easily, thus equipping management to be aware of, and make good decisions based on, the whole picture.

https://www.targetdashboard.com/nlimages/magsales.png

Though neither of these KPIs indicates any danger to the company, the manager might decide to use some of the resources used to market subscriptions, and turn them towards raw sales. If these were displayed on different charts, this trend might have gone unnoticed.

This information can also be displayed in different ways, such as the double bar graph.

https://www.targetdashboard.com/nlimages/magsales_Barchart.png

What you must remember though is that sometimes data can become hard to read if too many measures are added to a single chart, so it’s up to you to judge the balance. For more info, please refer to this post on the advantages of two charts over one.

2. Changing the Y-Axis Scale

Though this may seem simple, you’d be surprised how often it works! Sometimes, if there are only slight variances between monthly figures, trends might be hard to spot, as in this example of customer satisfaction.

  https://www.targetdashboard.com/nlimages/satisfaction_line.png

Since all of these figures are well into the 400’s, setting the minimum figure on the Y-axis shows a more varied line, which allows trends to be more easily seen.

https://www.targetdashboard.com/nlimages/magsales_yaxischange.png

A note on changing the Y-Axis – in order for this to give a meaningful trend, the figures can’t be too varied as otherwise you’ll end up with an exaggerated trend that can’t really be used.

3. Changing the Time Scale

As most KPIs are recorded at regular time intervals, the X-axis for KPI dashboard charts is always a unit of time, usually in weeks or months since those figures are most meaningful for finding trends. As we changed the Y-axis to compensate for slight variances, we can manipulate the time scale for figures that display too much variation.

Take this example of a company whose main revenue comes from a few major sales each month, rather than frequent small sales. Due to the nature of such a business model, the company might make a large number of sales one month, and a barely any the next. This would show up on a line chart as an incomprehensible squiggle.

https://www.targetdashboard.com/nlimages/sales_muddled.png

As you can see, this chart is not very useful for spotting a trend. In a situation like this, what you can do is have the chart display the total figures for each quarter. With this, the chart becomes not only easier to follow, but it is very easy to see the overall trend of the KPI too.

4. Creating a Rolling Average

As an alternative to changing the timescale of a chart, you can set a graph to display a ‘rolling average’ (also called a ‘moving average’). For every value, an average of a pre-determined number of previous values (e.g. the values from six months before) is taken.

https://www.targetdashboard.com/nlimages/sales_rollingavg.png

When this is done, a trend line with erratic jumps is ‘smoothed’ out, becoming visually similar to a chart with a high Y-axis value (as I mentioned above). While this may seem counter-intuitive, it gives a pattern to an otherwise unreadable graph.

While the chart above shows a six-month rolling average, you can make the line more exaggerated by decreasing the number of months counted, such as switching to a three month average.

While you’d ordinarily need to enter a mathematical formula into a program to achieve this result in a program like Excel, KPI dashboard software will do this at the click of a button, allowing managers to get the data they need exactly when they need it.

By experimenting with your data, you can uncover useful trends

In this article, I have discussed some methods of analysing data which can lead to finding trends which would have never been uncovered otherwise. Though some of these methods are quite simple, their effectiveness should not be underestimated.

Because we know how hard it can be to understand the amounts of data measured, we have purposefully designed Target Dashboard to be the easiest KPI dashboard software for managers to use. With just a few clicks, you can use Target Dashboard to do all of the tips mentioned in this article!

One thing I would urge you to take away from this post is to try experimenting with your data. By doing this, you may uncover some tricks of your own which will be useful in the future. For more ideas on how to make your dashboards and management reports as effective as possible, have a look at our Best Practice Guide.