Webinar: Consumer connected, the next frontier in forecasting and demand management
More and more companies have to deal with uncertainties when making forecasts. The increase of the number of SKUs, the influence of the promotion of goods and the shorter lifespan of the products make it more difficult to make good forecasts, says dr. Freek Aertsen, academic director of the Executive Master of Operations and Supply Chain Excellence of TIAS School for Business and Society.
Image: © Nationale Beeldbank
What is difficult about making good forecasts?
“There is more and more uncertainty when forecasting the demand of goods. Promotion of goods are becoming more important. The amount of beer sales depends on the amount of promotion. When a beer manufacturer advertises, the competition sells less beer. Also the lifespan of products is shorter than it used to be. Therefore there is less historic data to depend on. And like professor Marshal Fisher stated in the Harvard Business Review in 1994: ‘Thanks to global competition, faster product development, and increasingly flexible manufacturing systems , an unprecedented number and variety of products are competing in markets ranging from apparel and toys to power tools and computers.’ Mobile phone manufactures introduced 900 more varieties of handsets in 2009 than they did in 2000. This all does not help to improve forest accuracy.”
Also look at promotions and new products
How do companies forecast now?
“Companies look at the history and use this data to predict the future demand. They don’t look at what determines demand variability, like promotions, new products or phased out products I think companies need to use more differentiated data to make a better forecast. A good forecast is important; the quality of the estimation of the demand directly influences the quality of the chain in terms of customer service levels, inventory levels and cost levels.”
Information from other sources
What could go better?
“Over the last year more consumer related information becomes available from sources like social media, internet searches, groups and the point of sales data from the shopping outlet. This provides a wealth of additional knowledge that can be used to increase demand forecast ability and to actively shape demand. But what information can you use as a company? And what information do you need?
And although we are able to generate a near to perfect statistical forcast humans are able to ruin the quality while enriching the forecast. The role of the planner, decision making and financial target setting from the budget influence the quality of this forecast. How do you make sure humans don’t ‘empoor’ the forecast?”