Leading Retailer Makes Use of Weather Data
Category : Case studies
A leading retailer with a diverse product range – electrical appliances, home and garden furniture, toys and more – needed to manage its stock levels and distribution network better.It was eager to identify factors that would predict sales trends in order to match supply with demand, improve its operational performance, and respond effectively to customer requirements.Its seasonal product lines included a range of items that each had a limited sales window, which was often related to the local weather. The retailer also had access to highly detailed weather forecasts for all areas of the country.If these projections could be connected to customer behaviour, rapid-response buying and stock allocation decisions could be made in order to get the right products, in the right regions, on sale at the right time.
Station10 were engaged to help the company make use of this weather data. By merging the dataset with that of regional populations and boundaries, along with the organisation’s customer order records, Station10 built a structured and fully comprehensive database, which aligned customer purchasing behaviour with temperature and location.This enabled the retailer to pinpoint specific temperatures that effected the sale of particular products and how they varied across the country. Station10 also used econometric modelling, on a long timeframe dataset, to separate the influence of temperature from general product seasonality. Revealing, for example, that certain lines sold well when the temperature was low, rather than because it was Christmas.Identifying peak sales temperatures and climate based tipping points for a variety of product categories, Station10 uncovered some unexpected results. For example, it was discovered that hosepipe sales increased as it became warmer, but only to a certain point. Beyond that the impending threat of hosepipe bans hinders further orders.
Drawing correlations between customer purchasing behaviour and local weather patterns provided the planning and merchandising teams with the intelligence needed to predict sales trends and manage processes accordingly.This valuable insight generated long term business benefits.• Stock levels could be managed to directly reflect customer purchasing behaviour. This would in turn;
• Reduce waste
• Increase revenue
• Enable rapid response buying
• Enhance customer satisfaction
• Optimise operational processes