Retailers and consumer products companies have done a lot of work to apply data science to key challenges around demand forecasting, store replenishment, and optimisation of product assortment. Many have used historical sales and product data to identify patterns and glean insights. Some have used proprietary software products. Others have created data lakes and performed their own analysis. But an increasingly important tool is data from the wider world.
Many retailers and consumer products companies already engage with their partners up and down the supply chain. They ingest and analyse elements of their partners’ data but, as we all know, the external world is increasingly full of data too. Some of this data can have a major influence on customer activity, particularly at a local level. After all, no two stores are exactly the same and what happens within them differs for a variety of reasons.
Influences such as demographics, social media, local news, and the weather will affect each location and store differently. Bringing these external influences into the mix is important and it’s one of the primary focuses of IBM’s MetroPulse solution.
IBM MetroPulse collects data from internal and external systems and sources across the supply chain and the wider world. Critically, it combines and synchronises these data sets and applies deep learning algorithms powered by IBM’s Watson technology to them. The data is visualised to make it usable and to bring it to life. Insights, forecasts and actionable recommendations are created to enable professionals in retail and consumer products companies to positively improve their key performance indicators, reduce costs and increase customer loyalty, sales and profit.
How does this work? I’ll give you an example. IBM worked with a global consumer products company to use MetroPulse to ingest huge volumes of data for selected major cities where the company was targeting increased sales for a set of their products. The insights gained were used to optimise product and unit placement within specific stores. This resulted in approximately a 3% increase in revenue.
For retailers, this technology results in improved store replenishment. As IBM MetroPulse provides a hyper-local understanding of demand across the store estate, by combining internal sales history data with key external demand signals such as event and weather data, it can be used to optimise demand forecasting by “hyper-localising” replenishment down to a very local store level. Benefits include reductions in out of stock conditions, decreased shelf cycle times and increased sales and revenue.
This might not be the most glamorous end of a retail or consumer products company. But getting these things right is critical for success of the business.
Check out my previous blog post about the use of cognitive AI-based technologies to enable merchandising and logistics professionals to identify, predict and act upon issues in the supply chain. Read more about how IBM is innovating in the supply chain here.
