Skip to Content
Author's profile photo Former Member

Best Practices in Demand Forecasting

It is not uncommon to hear that a company’s management is worried about their forecasts.  “Our forecast numbers were not accurate,” is probably the most common statement made in management meetings. It even becomes the single reason for many business problems. It is needless to say that there will be some amount of risk when we try to predict the future. However, this risk can be reduced if proper tools and processes are used for forecasting.  Some of the key points to be kept in mind while designing the forecasting process are:

  1. Data – Wherever possible, historical data should be the key input to the forecasting process
  2. Collaborate – Different datasets provide different insights to forecasters. Various organizations have put in place different business processes for information sharing e.g. VMI, S&OP, CPFR
  3. Granularity – It is important to understand what kind of data is more important with respect to forecast accuracy. Is it the external data like competitor sales, POS data, sales team forecast or the internal data like stock-outs, shipments, orders, etc? Apart from this, it is also important to determine which time buckets are most suitable for forecasting. For example, whether to use monthly time buckets or weekly time buckets for planning.
  4. Analysis – Perform a detailed analysis of seasonality, trend and business cycles. Descriptive statistics can be used to analyse the boundaries. For example, the expected peak a product can reach, expected growth cycle time, etc. Based on this analysis, a model can be appropriately selected.
  5. Adjustments – In the field of demand forecasting, it is very important to understand that no single model will be suitable every time. Every forecasting cycle provides guidance on the directional changes affecting the market demand. Thus setting up a good feedback model is crucial.  Companies can think of comparing forecasts and actuals every quarter, and then tweaking the models based on the extent of deviations.

A good forecast can go a long way in helping companies operate efficiently and effectively. It is the most important lever for ensuring a win in the market. Demand Forecasting has been labelled by many as a dark-art and an inexact science, where importance of judgement cannot be overruled. However, the forecasting process helps companies ensure satisfactory levels of product availability in the market, thus laying the foundation stone for achieving good delivery service to the customers.

To read more blogs written by me, please visit :

Assigned Tags

      Be the first to leave a comment
      You must be Logged on to comment or reply to a post.