Budgets Vs Actuals – part I

This is the first post of a (probably) three part series discussing the techniques required to present Budgets Vs Actuals visualisations.  I would gamble this is one of the most commonly requested report types, it can however be ‘tricky’, to the point where many charts are presented without budget figures. 

“How’s your husband?” asked a colleague to a data analyst, “Compared to what?” came the reply.

This post discusses the available visual styles for presenting budget Vs actuals – subsequent posts will deal with handling the data required, and how some of the more complex visual styles can be built. 

Budgets Vs Actuals using bar type charts

Bars should generally be used for showing data which is split into categories – here I present five different approaches for displaying bar type charts which show both the actual and the target on the same chart.  I have chosen to present charts which I believe communicate the message to the audience well, and have deliberately left out chart types which do not communicate as well – such as the side by side bar for example.

Bar #1 – Using a simple line to represent the target

---

In this example, the actual figures are shown by the length of the bar, and the target by the black line.  The red/green colour of the bars is used to show which of the region/segment combinations have met their targets.  This is intended to draw the eye to ‘below target’ areas.

This is a very ‘basic’ visualisation – it shows the ‘black and white’ situation of ‘was target met’ extremely well, and would communicate this to any audience.

It doesn’t however present the variance between budget and actual very well (you can obviously see the variance, just not very clearly).

Bar #2 – Using a bullet chart

---

This is the classic Stephen Few designed bullet chart – here with a ‘stop light’ colour scheme showing where 60% and 100% of the target value appear.  There is a lot of information encoded in a relatively small area on this chart.

As with Bar #1, the bar showing the actual value could be encoded with a colour to show some other information such as the profit or transaction could within the region.

Bar #3 – Bar within bar

---

A personal favourite.  Perhaps it is just the high contrast colours chosen for this visualisation which make the areas which have achieved target so easy to see, or maybe the fact that there are less data points than on bars 1 & 2.  (8 Vs 16 data points).

Bar #4  - Super bar in bar

This bar in bar chart is a extension of chart #3.  The actual bar now encodes the profit ratio in the colour of the bar, and the sales value in the length of the bar.

Increasing the signal to noise ratio of the chart works in this example – although it is easy to over do this – I could have chosen to use the width of the actual line for another measure, but surely that would be too much!

Bar #5 – Bars to show variance

This chart is the only one of the bars to attempt to encode any time series.  The % variance between budget and actual is shown as the height (or fall) or the bar.  The same value is also encoded into the colour of the bar to make the good and the bad periods ‘pop’ from the data set.

This chart has a more specialist use than the others, as it only presents the variance without any consideration given to absolute values.  Presenting this alongside a chart showing absolute values gives the best of both worlds.

Budgets Vs Actuals using line type charts

If the budget Vs actual chart needs to include a time element, in order to see the trend, then line charts are a good choice.  The following two examples show alternatives based on line charts.

Line #1 – Running total towards a single value target

---

In this example, each of the small charts shows the running total of sales, towards a target line. 

This is a very clear representation of performance against target,but only really works when the story being told is that of achieving a single goal.

“We will sell the company when we have 1000 customers” – for example.

Showing the target as a running total instead of a single line makes more sense in most business scenarios.

Line #2 – Aligned target and actual lines

---

In this final example, both the target and the actual lines vary over time.  This is a commonly used chart type, but one that suffers from a number of significant problems:

• The chart is ‘just ok’ for revealing variance from target within a specific period, but not exceptional
• There is no way of seeing the trend of variance from budget – as there would be in a running total chart
• It is hard to compare differences between regions and segments

In summary…

As I hope I have demonstrated above, there are many ways to solve this particular charting problem.  As with most situations, the best solution will be the one which communicates the message appropriately to your audience.  Are you trying to show progress towards a single goal?  Or do you plan to beat target every month?  Perhaps the target is so important, that the number of days on which the target is missed per month is the most important measure. 

Good luck, I hope you reach your target.

The Information Lab

COMING UP IN PART II – Using ‘data blending’ in Tableau 6.0 to combine budgets from excel with actuals from a database