How ATM predictive maintenance can increase customer satisfaction
In some ways, ATMs are like your children. If you walk into your kid’s bedroom and notice that it’s suddenly spotlessly clean, you can safely assume that something big is about to happen: a request for an allowance raise, permission to attend a concert, or maybe a request to borrow your car. Experience has taught you their behavior, so you prepare yourself for what is inevitably to follow. That’s predictive analytics on a human level. So how does this relate to ATMs? Just as with your children, predictive analytics can be applied to forecast future actions. New banking technology enables you to monitor ATM service performance to determine when a failure might occur.
One bank recently implemented this approach throughout its system, and the results have been astounding. Up to 50 percent of potential ATM failures are now predicted before they ever occur. What’s more, the process has resulted in a two percent increase in ATM availability. So a single ATM location might be online and serving customers for an additional 10,000 minutes per year.
Predictive maintenance prevents crisis management
When an ATM goes down, your customers are understandably upset. After all, they’ve come to depend on uninterrupted service so they can conveniently manage their financial activities. When they can’t complete activities as planned, they become frustrated and may contact their branch, or they’ll turn away and walk to the nearest competitor’s ATM. At that point, it’s not just the customer relationship that suffers — it’s also your bottom line.
Recent banking technology innovations
Over the past five years, there have been extraordinary advancements in machine learning, cognitive analytics, big data, IoT and sensor technology. These innovations enable the production and collection of relevant data (observations) combined with the application of historical data (outcomes) to predict a probable event.
Banks can apply this capability the ATM systems and monitor and analyze each unit’s functions. Patterns emerge that help you truly understand each unit and identify and resolve a problem before it results in operational failure.
For example, one ATM may process a specific number of transactions under normal conditions, but if it’s subjected to high humidity, you may learn over time that a malfunction is likely to occur sooner. Knowing this idiosyncrasy, and being able to recognize impending problems, you can schedule and perform preventative maintenance before you’re faced with a crisis.
It’s this granular level of data collection and analysis, regardless of machine type, manufacturer, or configuration, that helps prevent ATM service outages and ensures more customers have more successful interactions — and better customer satisfaction.
IBM professionals have decades of experience in the banking industry and expertise in the area of ATM analytics.
Predictive analytics are making a profound impact on the future of banking. I invite you to learn more about ATM predictive maintenance and the positive results it has brought to those who have already embraced it.
Watch this video to learn more.