In this post, I will share some thoughts on why I think predictive analytics, which is also sometimes called prescriptive analytics, have become one of the most sought after solutions within the retail industry. Why? Because predicting what shoppers are likely to do is critical to understanding how best to complete the dialogue with them.
So what exactly is predictive analytics? In our experience, predictive models find relationships between historic data and subsequent outcomes so that near-term and long-term customer behavior can be predicted. But sometimes these relationships are so complex that only machine learning techniques will find them.
Enter Hivery, which has started to use predictive analytics to fuel growth in Coca-Cola’s vending business. Hivery, which is based in Australia, is one of our portfolio companies and two things caught my attention during their initial pilot phase:
1. Their algorithm was able to generate 15% additional revenue and 17% reduction in costs compared to the existing and global standard plan that Coke has been using for the past few years. These percentages equate to roughly $500 additional operating income per vending machine. When you consider that Coca-Cola has more than 13 million vending/cooler machines worldwide in the market, you can see the enormous potential strategic value Coca-Cola can reap from insights like that.
2. Hivery’s solution can quickly be applied to other parts of our business as well, such as retail shelf and value optimization. Traditionally, we along with our retail customers, and where legally permissible, have collected data on each of our customers to be able to augment our marketing and sales strategies. In order to make sense of that large amount of data, the information has generally been clustered. Reports or standard templates were then developed to share with our commercial and marketing teams.
But when Hivery came along, they changed the game. While those reports we had been using did an effective job of targeting the general behaviors of consumers, they do not provide the useful insights we needed to determine how individual customers or outlets are likely to behave because general behavior tendencies are simply too broad. Hivery gave us the ability to give us more real-time and individual data to base our decisions on.
In another example, Hivery is now working with one of our largest retail customers to help optimize stocking their shelves and increase sales by improving demand forecasting. The Hivery solution will be able to tell us three key insights:
1) Which shoppers buy which products.
2) How much they will spend.
3) Their frequency of spend.
Getting insightful information like this allows us to target our marketing more effectively and provide our customers exactly what they are looking for.
Hivery is driving the trend where retailers will be able to have a much richer and more personalized dialogue with their customers and be able to deliver on things like individual shopper or outlet preferences. This is why we believe that predictive analytics will be truly transformational for the fast-moving consumer goods sector and, in time, the entire retail industry.