Forecasting Solutions - Demand Planning Services
At Forecast Solutions we offer a wide range of services and a particular strength is in building forecasting solutions in Excel. Excel comes into its own when the number of entities that need to be forecast is fairly small and if a blend of methods is required that may be hard to find in affordable specialist software.
Forecasting solutions in Excel can be based on time series forecasting methods such as exponential smoothing, including the new forecast.ets function. Or causal modelling can be used to estimate the effects of factors such as weather, prices or economic indices. We are expert in statistical analysis to quantify the effects of causal factors and the results can usually be implemented in Excel. A combination of time series and causal methods is often useful, a prime example being in many call centre forecasting systems.
An outline of some of our main services is given below and for further details please visit our key services page. Or feel free to contact us by email or telephone:
Tel: 01844 291942
Demand Planning and Forecasting Training
Demand planning and forecasting training is a major part of the Forecast Solutions offering. The aim is to provide demand planning courses that are practical, but sufficiently detailed to impart a good understanding of the basic principles and the most important methods. Our forecasting courses are geared to be relevant, regardless of any specialist software that the company may have, but Excel is used as a common language for the worked exercises.
The Introduction to Sales Forecasting course explains the three basic approaches to demand forecasting. It goes into detail on a few of the most common methods and covers forecasting for new products and promotions. There is an Advanced Forecasting Course that builds on the work done in the introductory day giving greater detail and additional methods. The 2 day Demand Planning and Forecasting Workshop gives a good amount of detail on forecasting methods plus some aspects of supply planning including stock policy and safety stocks.
Time Series Forecasting
Many sales forecasting systems rely on the simple forecasting methods of time series projection, thus analysing sales trends and seasonality in the history and projecting into the future. Both curve fitting and the commonly used exponential smoothing methods fall into this category.
In medium term forecasting, such as for sales and operations planning, it is important to reflect product life cycles and expected changes in the competitive and technological environment. On the other hand short-term forecasting is more to do with detailed tracking and management of short-term demand in order to minimise inventory and maximise customer service.
Monthly forecasting is good for many businesses and is the most straightforward platform for dealing with seasonality. Many consumer goods companies, however, find that weekly forecasting is a better basis for dealing with an active promotional programme. Sometimes daily forecasting is necessary for the forecasting of short shelf-life products and for call centre forecasting.
Causal Analysis including Pricing Models
Demand is often affected by causal factors such as price relationships, unseasonal weather or economic indicators such as GDP. It can be highly beneficial to carry out analysis to validate the importance of suspected factors and to quantify their effects. Such causal modelling can be an important element in the overall forecasting solution and can help define pricing and other business strategy.
Forecast Solutions is expert in statistical analysis to examine vital areas like price sensitivity and the effects of economic leading indicators such as disposable income and gross domestic product. Unseasonal weather can drive significant changes in demand, so a study to quantify the effect of weather on sales can be very helpful for sales forecasting.
Forecasting Process Improvement
We can help in all aspects of developing an excellent demand forecasting process. The first step is usually to understand current methods and assess how successful they are in meeting business objectives.
The aim then is to identify improvements that can be made involving the forecasting infrastructure, methods and processes, with a view to improved forecast accuracy going forward. If there are significant benefits to be achieved by consideration of new or improved forecasting software, Forecast Solutions is well placed to help.
FMCG Forecasting and Promotions
The presence of past promotional activity in the historical data that is to be used as the basis for demand forecasting can cause problems in FMCG forecasting. It affects seasonal analysis, choice of forecasting model and setting the correct level of forecast demand. Therefore the sales history often needs to be cleansed of promotions before it is submitted to statistical forecasting.
Then there is the question of how to approach promotional forecasting going forward. If promotions have occurred frequently in the past, a good option is to try to build promotional profiles through analysis of previous promotions. Otherwise a more manual approach may be needed. Sometimes it is helpful to keep the promotional forecast separate from the baseline.
Seasonality in Forecasting
Finding the best way of dealing with seasonality in forecasting is an essential element in sales forecasting for many businesses and can sometimes dominate the forecasting task. Seasonal analysis becomes harder as the data becomes more detailed, so it is often useful to use a GSI (group seasonal indices) approach.
Smaller time buckets also make the calculation of seasonal indices more difficult and the task is a lot more difficult with weekly as opposed to monthly forecasting. This sometimes contradicts an intention to improve forecast accuracy by forecasting in smaller time buckets.
Improved forecast accuracy is vital in short term sales forecasting in order to lower stock levels, reduced operational costs and provide better customer service. For medium term forecasting, such as that needed for sales and operations planning, an accurate forecast is essential for best business and financial planning.
Forecast Solutions can advise on methods for forecast accuracy measurement including MAPE and WMAPE and can help with the development of suitable accuracy reports.
Remote Forecasting Services
Forecasts can be created to provide an expert interpretation of difficult data, particularly if the calculation of seasonal patterns is problematical - we have a unique way of dealing with that. Alternatively, we can step in to provide a regular forecasting service if internal resource is over-stretched or not available.