What’s special about actuals, budgets and forecasts?


It seems strange that the management accounting profession has yet to invent a universally recognised collective noun for the different types of data like actuals, budgets and forecasts. These data categories comprise one of the fundamental dimensions of corporate data, so it would be helpful to have a universally understood word to describe them.

Some software companies have created their own name for the dimension, such as “scenario”, “comparison”, “category” or “version”, and at Metapraxis we call this the “mode” dimension.

There are two main characteristics about the mode dimension that are important to remember when visualizing data, as follows:

1. Whether the data pertains only to the past (such as “actuals”) or both the past and the future (such as “budget” or “forecast”)

Implications of this distinction for visualization:

  • It should not be possible to display actuals for periods that have not yet happened, even if our spreadsheet or database contains values in the future
  • On trend charts, the current period up to which we have actual data should be clearly marked
  • We should be able to see budgets and forecasts both in the past and the future, and compare different versions of them (eg Quarter 3 Forecast against Budget)
  • The budget for a given year should never join back to the budget for the previous year, but should instead join back to the previous year’s actuals.  That is because the budget for a year is set in the knowledge of those prior year actuals, and bears almost no relation to the previous year’s budget

These key principles are illustrated in the trend line chart below, which plots a moving annual average of Gross Margin for a particular business.


MAA trend



2. Whether the data values for the mode pertain directly to the time period in which they are stored (e.g. actuals for the particular day, week or month), or to the end of a longer time interval (e.g. the latest forecast for the current quarter or the next year end)

We call the latter type of mode a “period end” mode.

Implication of this distinction for visualization:

  • We can plot the trend of a period end mode, but the values always pertain to the end of the given period.

This principle is illustrated in the chart below, which plots a year to date trend of Net Sales for a particular subsidiary.  The History of Year End Forecast line plots the time trend of the period end mode called “Year End Forecast”.




These key features of the mode dimension are easy to overlook when we create charts unless they are designed into the underlying source of the data, providing consistency in the treatment of different types of mode in our analyses.


  1. Gerald Davies

    You are right to draw attention to the importance of values pertaining to the same period end. A common problem when comparing so called like for like data.


Leave a Reply