|Home | Training | Software | Services | Testimonials | Contact | About Us|
|Demand Forecasting & Planning: FMCG Forecasting and Promotions|
FMCG Demand Forecasting - Promotional Forecasting
Forecasting for fast moving consumer packaged goods has all the usual challenges of demand forecasting, but is characterised most of all by the high impact of promotions carried out by companies and their competitors. Frequent promotional activity complicates the already difficult tasks of historical data cleansing and seasonal analysis Then there is the task of estimating and integrating the effect of future promotions 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 into the process, tends to weigh somewhat towards weekly forecasting. A slight downside to this is that, with weekly forecasting, the seasonal analysis becomes more difficult and may need special attention.
Forecast Solutions has successfully created forecasting solutions in Excel for FMCG, including a neat and convenient method for dealing with historical and forecast promotions. If you need a large scale and / or multi-level solution we can offer a very good choice of packaged software or help with a software selection process.
Short term forecasting for FMCG is most commonly approached using time series forecasting such as exponential smoothing. However, there may be benefits in using causal modelling 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. Click here to see some ideas on trend forecasting.
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 to the left.
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:
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.