Dynamic Predictive Segmentation Will Upgrade Retailers Relevancy with Shoppers
Retailers Emails are Misfires for Many Holiday Shoppers – Dynamic Predictive Segmentation Will Help
It’s well known that the “Amazon effect” is wreaking havoc on retailers’ ability to lure and retain shoppers because of theirinability to predict what motivates customers to buy and to be relevant in the moment of interaction. A WSJ article from November 27, 2017, Retailers Emails are Misfires for Many Holiday Shoppers, puts an exclamation point on how much retailers are struggling with relevancy in their messaging to current and potential customers.
The article states “Nearly 90% of organizations say they are focused on personalizing customer experiences, yet only 40% of shoppers say the information they get from retailers is relevant.”
The article goes on further to provide specific examples of brand challenges.
- “Gap Inc., with a score of 40, sent emails featuring women’s clothes to one of Sailthru’s researchers, even though he had created an online profile indicating he was a man and that he was most interested in items for men and babies.” A Gap spokeswoman declined to comment.
- Kohl’s Corp., which scored a 45, segments its emails based on gender, income, geography and other metrics, which is different from personalizing messages for a single shopper.
- Segmentation, while effective, can miss nuances of consumer behavior, said Jason Grunberg, Sailthru’s vice president of marketing. “Ozzy Osbourne and Prince Charles are both British men in their late 60s, but they aren’t necessarily interested in buying the same things,” he said.
Dynamic Predictive Segmentation is a new application enabled by explainable machine learning algorithms that helps solve this problem, and it’s time for the retail industry to take notice and take action. It’s still early days for the adoption of machine learning for marketing, but it absolutely can make a difference now.
And there is no better area for quick wins than aligning email campaign messages to match the primary drivers of why a customer is predicted to buy a certain product.
Dynamic Predictive Segmentation (DPS) is an approach that segments customers based on propensities to take a specific action grouped by the most important shared characteristics revealed for each segment. Segments and their associated characteristics change dynamically in real-time based on new data from customer interactions. Therefore, this type of segmentation is dynamic in multiple ways because not only do customers’ segment assignments change as their behaviors change, but their propensities also update in real-time as well. Advances in “explainable AI” have made the factors leading to these segments transparent, enabling a new level of contextual relevancy in customer interactions.
How does DPS provide greater contextual relevancy for marketers? Imagine that instead of the traditional descriptive characteristics that that define a group of consumers homogeneously, you have predicted buyers defined by the weighted factors that are the drivers of their predicted behavior. These factors could range from shopping frequency, timing or channels used, to specific product features favored, or recent events that have occurred. Most machine learning algorithms currently used identify prospects with a high likelihood to purchase. While this technique identifies the right customer to target, it does nothing to reveal the best message for the customer. For example one predicted Apple iPhone accessory buyer may be interested in protective cases for damage prevention, while another likes colorful designs. Dynamic predictive segmentation reveals these customers’ differing preferences, so retailers can exploit them by delivering the right content to each.
Retailers should begin testing Dynamic Predictive Segmentation in email campaigns to evaluate the significant value that comes from aligning the right message to a buyer’s motivations. Just the improved precision from using machine learning predictions will elevate performance by 20%-70% or more over statistical models. Add relevant messaging and retailers will reap big gains and happier customers.