Injecting Real-Time Personalization at Scale with Machine Learning

Posted on: July 19, 2018 | Posted by: Sona Sharma

Built to provide amenities with speed and ease, convenience stores are well-aligned with the on-the-go lifestyle of today’s consumers. But while they have fared better than their fellow brick-and-mortar retailers in recent years, these types of stores are not completely immune to the challenges of engaging customers in our hyper-connected world: they face competition from e-commerce sites that offer same-day delivery; stores that allow customers to order online and pick up in-store; and quick service restaurants that boast new, healthier menus.

So what do convenience stores have to do to stay competitive and relevant to their customers?

Most convenience stores are aware that today’s consumers are looking for them to offer fast, economical, and increasingly healthy food options. But restocking their shelves with fresh sandwiches, salads, and wholesome snacks is only part of the answer. In order to truly meet evolving customer expectations, they should also be seeking out new technologies that offer highly-personalized shopping experiences built to encourage customers to visit their stores, and keep them coming back for more – because studies show that engaging customers with personalized and relevant communications can build loyalty and drive revenue growth of 10% to 30%.

Let’s take the example of convenience stores that are attached to a gas station. These stations may have some customers who pump gas on a regular basis. Some of their customers may even have the store’s native app on their mobile phone and/or some form of a loyalty program. The challenge for this setup is to convert gas purchasers to in-store customers. They need a way to capitalize on the few minutes customers spend at the pump right outside the store with meaningful offers that give them a reason to enter the door. Let’s consider this scenario:

Richard uses the gas station close to his office on a regular basis. The gas station has a convenience store attached to it that offers snack items, coffee, donuts, salads and sandwiches. At times, when Richard pumps gas in the mornings, he picks up orange juice and his favorite chocolate glazed donuts. He even has the mobile app of the store on his phone and is eligible for a free drink. However, because he is usually in a hurry, it has been a two visits and he has not gone to store to get his free drink. On his third such visit, while he is waiting to fill gas he gets the following message on his mobile phone:
“Good morning, Richard. Your free orange awaits you in the store. And don’t miss out on our fresh batch of chocolate glazed donuts while you are there.”

Chances are that after reading this message, Richard will take a few minutes out of his busy schedule to quickly grab some his juice and his favorite donuts.

The technology that supports this level of influential personalization is a new stack; a new generation of systems that provides enterprises continuous intelligence about every customer, based on both current and past activities. This visibility into a customer’s complete context enables enterprises to touch every customer in a hyper-personalized manner.

Machine Learning (ML) is at the heart of such personalized engagement. ML ensures that as the system ingests more and more customer data, it is able to learn and act on customer affinity and intent. These insights then trigger the most appropriate and relevant action—a personalized message, offer, or notification for the customer while the customer is still in or near the store.

To read more about how ZineOne’s Customer Engagement Hub enables convenience stores and quick service restaurants to incorporate personalized experiences at scale, download the case study on Intelligence Personalization at Scale: How Machine Learning Is Helping Convenience Stores Drive Foot Traffic.