Predictive Analytics: Data and Retail Expertise - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Data Management // Big Data Analytics
Commentary
7/27/2015
11:30 AM
Ariella Brown
Ariella Brown
Commentary
50%
50%

Predictive Analytics: Data and Retail Expertise

Predictive analytics have been proving their worth in the retail sector, with examples showing showing the sector how predictive analytics can blend with industry experience in decision making.

Here on All Analytics, we’re generally sold on the value of predictive analytics. The question is: Are retailers, particularly those managed by people who believe in their gut intuition, sold on it? Even they are starting to appreciate what analytics can do for their business.

Dean Abbott
Dean Abbott

According to Dean Abbott, co-founder and chief data scientist at SmarterHQ and author of Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst, it is ushering a cultural change for retail.

I recently chatted with Abbott about what the application of predictive analytics means to the retail space. He said that it has emerged only in the last few years. At present, there is still “so much low-hanging fruit,” that it is “really not hard to improve” results. Though it’s something new for retailers to wrap their heads around,” they do recognize that predictive analytics bring better results than “gut” decision. Seeing the success others in the industry have is effecting “a cultural change” that embraces data analytics.

But data and analytics alone will not bring success. Abbott stresses that a successful data-centric organization depends on a “three-legged stool,” which is made up of good data, advanced analytics, and domain expertise. It’s on the last leg that the “gut feeling” or intuition that is built on experience comes into play. It’s “just as important as the math behind the algorithm,” Abbott says.

He explains, the people involved in the actual business are the ones “who know the customers” and “which metrics are significant” for the data modelers to incorporate. Without that “deep domain expertise, you don’t know what you’re looking for.” That’s why you need all three working together to have “the data accessible, collected, and contextualized in the right way.”

Getting the context right is a critical part of making accurate predictions based on the data. Abbott explains that when a customer makes a single purchase, you don’t get enough context to understand that customer’s purchasing behavior. One the other hand, a “timeline of repeat purchases of a particular product, say a wallet every six months,” two or three times can be identified by predictive analytics to establish a purchasing pattern.

On the basis of that historical data, you can predict when the customer will be in the market for a replacement and market to him accordingly. Historical data can also reveal what types of offers resonate with individual customers; whether or not they are receptive to upselling, and how they prefer to be contacted, whether it is via email, snail mail, or text.

That’s why loyal customers are so valuable, not just for the sales themselves but for the trail they leave about their purchasing habits. Once you have their pattern, you know when to expect a purchase and can send a message that has gotten a positive response before. More historical data leads to greater understanding, which can direct more effective customer communication.

Timing is everything, particularly in marketing, as you have to catch the right person at the right time to seal the deal. Analytics “can determine the most effective sequencing and most effective duration between contacts to drive sales.” It will also identify which forms of communication are not effective for particular customers, say mailing catalogs, and suppress the ones that don’t work.

Because customers can behave in one way online and in a different way in physical stores, the optimal tracking would include information on both. That would reveal if the customer is the type to browse in store but buy online or the opposite. Abbott admits, though, that tracking in store behavior has to strike the delicate balance between picking up information and coming across as creepy.

Ultimately, business decisions don’t have to be reduced to intuition versus data. Instead, the expertise of business people can inform the metrics applied to the data collected. And with all three, you’d have a truly data-driven marketing strategy.

 

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Commentary
Enterprise Guide to Edge Computing
Cathleen Gagne, Managing Editor, InformationWeek,  10/15/2019
News
Rethinking IT: Tech Investments that Drive Business Growth
Jessica Davis, Senior Editor, Enterprise Apps,  10/3/2019
Slideshows
IT Careers: 12 Job Skills in Demand for 2020
Cynthia Harvey, Freelance Journalist, InformationWeek,  10/1/2019
White Papers
Register for InformationWeek Newsletters
Video
Current Issue
Getting Started With Emerging Technologies
Looking to help your enterprise IT team ease the stress of putting new/emerging technologies such as AI, machine learning and IoT to work for their organizations? There are a few ways to get off on the right foot. In this report we share some expert advice on how to approach some of these seemingly daunting tech challenges.
Slideshows
Flash Poll