The Diaper-Beer Model – Value-driven by Predictive Analytics
While data is growing exponentially, it’s becoming harder and harder to find valuable insights. This is often due to the data itself, but it’s also due to the complexity involved in looking for insights.
If you want your data to work for you, it’s important to make sure that you’re using it in the right way. While it’s becoming easier to collect data, the insights that you find are all about the model you are using them with.
This blog will look at how you can improve your store using data, using the diaper-beer model as an example
The Diaper-Beer Model
A legendary story from 30 years ago illustrates the value that data can bring to modern storefronts. Back in 1992, a group of consultants supporting a Midwest retailer found an interesting correlation between diaper and beer sales. The company cross-referenced sales data and store layout to generate unique insights, helping identify items that were often purchased together.
After digging deeper, the consulting team hypothesized that when men went shopping for diapers on weekend evenings, they often bought beer as well. This finding prompted the retailer to change the store layout so that the beer and diaper aisles were next to each other—leading to higher sales in both categories.
Though there are some skeptics of the diaper-beer model, the case study is a useful example of predictive analytics at work. The consulting team was able to sift through client data, uncover actionable insights and then make an informed decision that led to positive business outcomes.
The modern storefront is completely digital. It’s a company’s website and application. These new stores and digital aisles are generating an insane amount of data. To modernize the diaper-beer model, organizations can use a data lake to not only house all this unstructured data but also the tools necessary to generate insights on their data at an unprecedented scale. [source: Forbes]
It’s important to know when to look for patterns within your data, this way you can help guide consumers and encourage them to buy different products. It’s always a great idea to try and A/B test a few options and then compare them afterward. If you take a look at the famous diaper beer model, it’s good to make sure that your team is willing to constantly collect large amounts of data. This case story sparked conversations around predictive analytics.