The solution WATR developed was an AI-powered traffic light system.
“We created this machine learning model that is fed or trained with sources, different sources,” Eduardo says. “For instance, the CRM, the support desk… use Zendesk. So Zendesk has all the information of interactions between support, tickets, agents, problem solution, and customers.”
The system analyzes multiple data points to predict renewal likelihood. Eduardo explains, “If this member is in this tier of membership, they’re paying, let’s say, $100 per year. They only have one barcode. They have not attended any events. They haven’t logged into the platform. And they last signed into GS1 in 2020… maybe it will be red, and they won’t renew, you know, because they’re not taking value out of the membership.”
The traffic light system doesn’t just identify at—risk members—it provides specific recommendations for action.
“You have a customer agent… that has all this information… in the computer and has the insights and has the recommendations… and knows that this client is at risk of actually not renewing, then you know you can drive the conversation there,” Eduardo explains.
The system is particularly effective at identifying three key member segments:
– For red-flagged accounts, Eduardo notes: “They’re not going to renew. All indicators are telling us this guy is going. He’s leaving us.”
– Yellow-status members require attention but aren’t immediate flight risks.
– Green-status members present opportunities for growth: “The green ones also have a strategy, because you can work with the green ones, because they’re your customer success stories,” Eduardo says. They’re the ones who will recommend your company because they’re the ones who actually see value, who understand how to leverage your business and how to use it for their own growth.
The platform continues to evolve. “We’re updating the rules and the algorithms,” Eduardo says. “Like if a client enters the platform 10 times a week, maybe it’s a good sign, but maybe it’s because he’s having problems and he’s trying to solve them and he’s not solving them.” WATR is evolving to make sure they can differentiate between two situations like that using data.
Beyond basic predictive analytics, the system includes a recommendation engine. These recommendations enable proactive engagement.
If the customer loves GS1 events and has attended three of them in the last four months, the system can prompt the customer service agent to recommend an upgrade to the next membership tier that gives the customer access to six additional access passes to GS1 events.
“You can analyze the data based on the historical, then you can project and try to foresee a little bit what’s going to happen and take a course of action based on the traffic light, depending on the probability of renewal,” Eduardo says.
The platform includes what Eduardo calls “prospective analytics.”
“The forecasting and the prospective [are] where you can play with different scenarios: What happens if we increase the price of the membership? What happens if we lose 5,000 members? And what happens if we increase the price of this platform?” Eduardo explains.
This approach represents a significant evolution from GS1’s previous marketing efforts. As Eduardo explains: “For me, understanding these marketing efforts with technology, it’s not just automating campaigns, right?… It’s not just using AI to enhance your Google Ads campaigns… Everyone can do that.”
The success of GS1’s AI-powered retention system demonstrates how machine learning can transform member engagement when properly implemented. As Eduardo concludes, “You have the 360 view. You can analyze the data based on the historical, then you can project and try to foresee a little bit what’s going to happen and take a course of action based on the traffic light, depending on the probability of renewal.”
For organizations struggling with member retention, GS1’s experience shows that AI can do more than just automate existing processes—it can fundamentally transform how organizations understand and engage with their members. The key is combining sophisticated backend analytics with simple, actionable frontend interfaces that empower customer service teams to take the right action at the right time.