Forecast Solutions

FMCG Forecasting - Forecasting for Promotions

Forecasting for fast moving consumer goods has all the usual challenges of demand forecasting, but is particularly influenced by the high impact of promotions.  Frequent promotional activity complicates the already difficult tasks of historical data cleansing and seasonal analysis, then there is the task of estimating the effect of future promotions and integrating them into the forecast.

As in any company a choice has to be made regarding the time bucket to be used in forecasting.  The dominance of promotional forecasting in FMCG, together with the likelihood of needing to integrate weekly customer forecasts, tends to tip the balance somewhat in favour of weekly rather than monthly forecasting.  A slight downside to this is that the important matter of seasonal analysis becomes more difficult and may need special attention.

We have successfully created forecasting solutions in Excel for some small FMCG businesses, including a neat and convenient method for dealing with historical and forecast promotions.  When the scale and complexity of the forecasting task for consumer products suggests that specialised forecasting software may be necessary, Forecast Solutions can recommend a particular packaged solution or help in a software selection process.

Click here for more information on forecasting software.


retail shopping mall

Forecasting Methods

Short term forecasting for FMCG is most commonly approached using time series forecasting such as exponential smoothing.  However, there may be benefits in using a causal modelling approach to take account of drive factors in the business such as pricing, weather or economic indices.  Forecast Solutions can help in terms of a price modelling study or weather sensitivity analysis.

If causal analysis is not applicable or is not possible due to complexity or lack of data, then the family of methods called time series forecasting is usually used.  This involves the analysis of the seasonality and trend shown in historical demand, with the intention of projecting those patterns into the future.  For short term forecasting it is usually sufficient to deal with trend as a straight line or to ignore it altogether.  In medium term forecasting such as in a sales and operations planning process, there may be a need for alternative methods and processes, particularly for the longer horizon where a detailed promotional plan may not yet exist.

What the FMCG forecaster often benefits from is a wide range of information including EPOS data and continuous market research information, so a further challenge is to make the best use of all the rich data available.

Forecast Solutions can help with all of these considerations.  For further details of our consulting and training services please use the links above left or click here.

Historical Data Cleansing

Before any form of statistical forecasting for FMCG can be carried out it is essential to cleanse the sales history of promotional activity and other abnormalities. This is particularly true if there is pronounced seasonality as promotional history can easily confuse the seasonal analysis.

There are a number of ways of dealing with promotions in the sales history, although the forecaster may be somewhat dependent on tools existing in software to hand. Some of the options that can be considered are:

  • try using the automatic data cleansing that may be offered in your forecasting software

  • adjust the historical data on the basis of what is thought would have occurred had the promotion or other event not taken place

Once the sales history has been cleansed it can then be submitted to the  statistical procedures chosen for the preparation of the baseline forecast i.e. the forecast excluding promotions.

Estimating the Effect of Promotions

In order to estimate promotional effect, either for data cleansing or for forecast preparation, it is sometimes necessary just to use judgement guided by market intelligence. This may be particularly true for companies where promotions are uncommon or when the particular type of promotion has no precedent. For most established FMCG companies, however, there is likely to be a good amount of history on various types of promotions.

Given that, it is often possible to analyse previous promotions, consolidating those of a particular type, in order to arrive at a set of promotional profiles. These can be based on volume or on percentage uplift. Profiles will be useful to help with data cleansing and to create the promotional forecast going forward. Success in building suitable profiles is helped dramatically by having good information on the mechanic, display features and support involved in prior promotions. Maintenance of such records on an ongoing basis is essential for an effective FMCG forecasting process.

Forecasting Process for FMCG

If any statistical forecasting is to be carried out the first step is to determine what historical data is to be used. This might be invoiced sales, or it may be decided that historical orders provide a better measure of true demand. Many FMCG companies will also receive EPOS data from some major retailers, giving the further possibility of forecasting based on EPOS sales or on Rate of Sale combined with a distribution forecast.

Historical data needs to be cleansed, then the chosen statistical process can be run in order to create a baseline forecast excluding promotions. For new products there will inevitably be the need for subjective and market intelligence inputs. Then the promotional forecast needs to be built and added or otherwise integrated with the baseline to product the overall forecast. The promotional forecast is likely to need a subjective input and is sometimes the responsibility of a different department (e.g. sales) as compared to the baseline (e.g. demand planning). As in all forecasting, forecast accuracy needs to be continually monitored with the aim of maintaining a path of continual improvement.