We sat down with one of our data scientists, Matt McKinney, owner of our secret sauce, to breakdown the world of data into terms our whole team could relate to.
What is machine learning?
At its core, machine learning is pattern recognition. It is the practice of enabling machines to analyze large data sets and draw predictive insights, enabling the machine to become smarter over time.
How does machine learning work?
The more data you feed the model, the more it recognizes the patterns – it gets smarter. Our platform is getting smarter every day! Machine learning recognizes patterns in the same way that human beings recognize them (for survival, example: we learn as children not to touch something hot).
MakerScore is our proprietary metric that helps you identify which colors to carry in your basic T-shirt for next fall, or whether to put that logo on the pocket or the sleeve, or if your younger customers like a big tote or a satchel. At MakerSights, we help you reach out to your customers directly, and use their preferences coupled with more complex analytics to predict purchase intent (measured by the MakerScore).
What are the biggest benefits of data in retail?
By collecting data and using machine learning, it’s possible for brands and retailers to move away from instinctive purchasing (based on a hunch), and remove emotion from the decision making process. Using predictive purchase intent, they confirm their suspicions, and ultimately make more informed and confident decisions. It’s adding science to the art. Then, with these clear product preferences tied to each unique respondent, brands can follow up personally with each of their customers and engage them in a much more meaningful way.
How is this data actionable?
We strategize a lot of different tests to support brands’ hypotheses, but Matt’s favorite is the line review test, which projects top, middle, and bottom performers within an assortment, making it easy to identify SKUs to invest in more aggressively, and others to scale back.