Overview
It is often useful to explain variances of sales between two or more files by splitting total variances into the volume, price, exchange and new product contributions. This topic includes the following:
- how to execute this type of analysis
- highlights some of the practical issues involved in preparing data for the analysis
- discusses some of the practical interpretation issues of results produced
Calculation Logic
Formulae
When IFP calculates variances between values in File A against those in File B, the following basic calculations are carried out:
Volume = (File A units – File B units) * File B price
Price = File A units * (File A price - File B price)
Exchange = File A value @ File A rates - File A value @ File B rates
Calculation Periods
By default, the above calculations are carried out as follows:
- calculate variances for each month within the selected analysis period
- sum the results to give the total for each variance component
Technically this method is the correct one to use, however some organisations prefer to use year to date totals for this analysis. For this reason, an advanced option, Use consolidated period totals for variances, is available to perform analysis as follows:
- sum monthly data over the selected period to create consolidated period totals
- calculate variances directly using this consolidated data
A simple example is shown below:
|
Jan |
Feb |
Mar |
Total |
Actual Units |
100 |
60 |
40 |
200 |
Plan Units |
150 |
0 |
80 |
230 |
Actual Value |
500 |
300 |
240 |
1040 |
Plan Value |
600 |
0 |
320 |
920 |
Actual ASP |
5.0 |
5.0 |
6.0 |
5.2 |
Plan ASP |
4.0 |
0.0 |
4.0 |
4.0 |
Total Value Variance = Actual Value - Plan Value = 1040 - 920 = 120
Monthly analysis
Price Variance = 100*(5.0-4.0) + 40*(6.0-4.0) = 180 (Feb has zero Price Variance since Plan ASP = 0)
Volume Variance = (100-150)*4.0 + 300 + (40-80)*4 = -60 (All Feb variance is Volume, since Plan ASP = 0)
Consolidated period total analysis
Price Variance = 200*(5.2-4.0) = 240
Volume Variance = (200-230)*4 = -120
Sequence of Files
The order of selected files is important here. For example if 2013 Actuals are being compared with 2013 Planned Values, then File 1 must be 2013 Actual and File 2 must be 2013 Plan.
Option Selections
The following analysis is created using File Summary Analysis, making the Custom File Selections as shown:
Price Variance Special Item Types
For items assigned to Item Type 09, Price Variance Special, the total variance (excluding exchange) is assigned to price for all months in the selected analysis period. This item type is used to store special discounts given to customers that are not related to individual products.
New Product Variances
For any month within one year of the item’s launch date, the Volume and Price Variances are set to zero and the Total Variance (excluding exchange) for the month is assigned to the special variance category New Product.
For example, suppose an item has a launch date of July 2012 and the analysis period is January to August 2013. In this case, the Total Variance for each of the months January to June 2013 would be assigned to New Product, whereas a normal set of Volume/Price variance calculations would be used for July to August inclusive. Note that an item may have a mixture of New Product, Volume and Price Variances during a particular selected analysis period.
Launch dates are set via Item Codes Maintenance. The classification variable Item status is not used in Volume/Price Variance analysis.
Special Rules for Zero Values
Where the Units and/or Value for either file are zero for a given month, no Price Variance can be calculated. Hence, the Total Variance (excluding exchange) is assigned to Volume. This is the case for February in the example shown above.
Levels of Detail
All calculations are carried out at the lowest Level of Detail for the database concerned and then summed to give required sub-totals.
For example, if the lowest level of detail is Item Code, then all calculations are carried out for each Item Code and then summed to give Product Group totals.
Comments
1 comment
Excellent article Damon,
First time I find one addressing potential pitfalls in level of detail, zero values, etc. In my experience, it is critical to have these concepts clear as they can really affect the results and interpretation of your variance analytics. I guess the key is to make sure the results of the analysis portray the reality of the drivers affecting the business.
Best,
Enrique
Please sign in to leave a comment.