Forecast Solutions

Sales Forecasting Service

At Forecast Solutions we can provide a remote demand forecasting service for any size of project ranging from a forecast for a single product to the creation and operation of a major forecasting support system for many thousands of products or listings.  It may be just an ad hoc analysis carried out in order to give guidance or a second opinion, or a detailed ongoing sales forecasting service to replace, on a temporary or permanent basis, the company's own specialist software and human resources

Ad Hoc Forecasting Services

For the creation of a one-off demand forecast for a small number of product lines we will receive the historical sales data on an Excel file, then carry out expert analysis to construct the sales forecast. 

The historical data is tested for the presence of a seasonal pattern and, if the seasonality is statistically valid, a set of seasonal indices is calculated for use in forecasting.  If the sales history is weekly, as opposed to monthly, it may be beneficial to smooth the indices to reduce the jagged characteristic that weekly indices often display due to the greater volatility at that level of detail.

The parameters will be optimised for a number of alternative forecasting models and the best performing model selected based on simulated historical forecast accuracy.  A unique feature in this process is that the parameters can be optimised based on forecast accuracy over the supply lead-time for the product in question.

The result is that a sales forecast is supplied.  Optionally we can provide an Excel spreadsheet for updates to the historical data so that the forecasts can be refreshed.

Outsourced Forecasting Service

For an outsourced service we will build a forecasting database using Excel or selected specialist software.  Extensive analysis will be carried out to set up the forecasting system in the most advantageous way.  It is often beneficial to generate seasonal patterns at an aggregated level and deploy them at the detailed level, so this possibility will be explored.  A range of forecasting models will be deployed and the most successful selected on the basis of simulated forecast accuracy over the sales history.

A set of forecasts are produced at the outset and at regular intervals thereafter, usually but not necessarily on a monthly or weekly basis.  There will be a regular review of the system configuration to ensure best possible forecast accuracy over time.

Depending on the type of software that is chosen it may be possible to provide remote access to the hosted system for the purpose of intervention with forecast overrides regarding promotions and other exceptions.  It may also be possible to automate data transfer operations for the receipt of sales history and distribution of forecasts.

Incorporating causal factors in the forecasting process

As well as using time series forecasting methods such as exponential smoothing in the most effective way, Forecast Solutions is also skilled in demand analysis and econometric modelling using regression analysis to quantify the effect on sales or market share of drive factors such as pricing, economic indicators or  unseasonal weather.  It may be possible to incorporate the results of a causal analysis into the forecasting process.