AI-Driven Customer Service Delivers the Best Problem Resolution—Not Having a Problem at All

What if a business could realize that its product was about to present a challenge and could intervene proactively, before the customer experienced any frustration at all?

Much of the buzz surrounding artificial intelligence (AI) in B2B customer service focuses on problem resolution. Emerging AI technologies now offer immense capabilities to understand customers when they communicate in their own words, identify the needs they’re trying to express, and plumb lakes of information to return the precise response each customer was seeking.

Often faster and more accurate than humans, AI bots resembling IBM’s famous Watson or Amazon’s Alexa are becoming a great resource. Sometimes they take the place of customer service specialists, such as in online chats. Other times, they work behind the scenes as tools for telephone representatives, field technicians, and other support staff, making these individuals more effective in meeting customers’ expectations.

Such solutions are rapidly changing the customer service experience for the better, but they exhibit one devastating shortcoming—they only come into play after a customer has a problem or question in the first place.

What if a business could realize that its product was about to present a challenge and could intervene proactively, before the customer experienced any frustration at all?

Eliminating the Negative to Improve the Customer Experience

This we know, customers frequently contact the service department when they’re having a negative experience. An incorrect bill was issued, they can’t navigate a menu option, or a product is malfunctioning. Representatives sweep in to repair the relationship.

Some companies are attempting to become proactive, striving to shape customer satisfaction with positive engagement. But now AI is opening the final frontier, predictive service.

The idea is to remove friction altogether, by discovering and repairing problems before they happen. It sounds like the stuff of sci-fi movies, but it’s here now and being applied to an incredibly complicated use case—corporate information technology infrastructure.

Nearly all companies today rely on a complex web of data systems to do everything from serving office applications to employee desktops to keeping their mobile apps running. Eventually components in their huge server farms and along their intricate networks will fail, or the “brains” in the software that runs them will act up. The role of Park Place Technologies is to maximize IT uptime for companies large and small—keeping them online and in business—and for us, predictive service has been a transformative innovation.

The Value of Fixing Problems Before They Start

Here’s how predictive service works in our operations. The concept applies machine learning, a form of AI, to understand patterns in a customer’s data center and identify when certain events within vast streams of performance blips and status updates might portend a serious issue-in-the-making. When the potential for outage is spotted, a trouble ticket is automatically created. Our team receives an alert with all the information they need to initiate a solution.

This model has numerous benefits over the old school means of problem identification and resolution:

  • Ensuring nothing falls through the cracks. It’s easy for a human to overlook an error flag amidst the flood of information in today’s management dashboards. Keeping AI on the job means even the smallest piece of data generates the appropriate response.
  • Taking customers out of problem reporting. In the past, customers took on the responsibility of communicating with support and even conducting the initial troubleshooting. Now we have complete visibility and can diagnose issues at a distance, without their involvement.
  • Personalizing the interaction immediately. Leveraging complete data center inventories and account information, we skip that annoying information-collection phase that usually launches a customer service contact.
  • Improving response times and outcomes. Technicians hit the ground running. For example, engineers can arrive with the right spare parts in hand, ready to install. This accelerates problem resolution and boosts first-time fix rates.

Most importantly, all of this is typically happening before the customer is even aware of the issue. Many times, the first a data center manager is hearing of an event is from us, asking for permission to make a remote change to an operating system or alerting them that we’re on the way to replace a component.

This approach initiates a radically different and more positive customer service interaction than many companies enjoy. We get to pair a solution-in-progress with the initial notice of a problem. Customers barely have time for dismay before they’re back to 100%.

Implications for Customer Service

Last year, our company, Park Place Technologies, assessed the status of machine learning and decided to take the plunge. We deployed a machine learning-based predictive maintenance capability for our clients and called it ParkView. It was an industry first, and there’s always risk in getting out in front—but less than a year later, the results are impressive.

For any B2B company, the potential to avoid the fraught and often angry problem-reporting phase and jump right into making the customer happy—it’s nothing short of a golden ticket to customer satisfaction. We’re finding that this capability is shifting our conversations with clients, from handling immediate crises to designing longer term optimization and planning strategies. The predictive mode is infectious and empowering, and it can turn a standard, transactional customer relationship into a true partnership—and yes, revenue-generation opportunities ensue.

The applications for predictive maintenance certainly aren’t limited to data centers. The world is currently adapting to the Internet of Things (IoT), an age where almost every product imaginable will be collecting data and connecting to the Web. New services of all types will be layered on top of these devices to leverage the wealth of information in seamless ways for customers. We’re already seeing it in our refrigerators, doorbells, and other home automation equipment, and with industrial IoT, smart technologies have hit factory floors, oil pipelines, and remote facilities around the globe.

In other words, every facet of our lives and businesses is coming to resemble the complex IT systems to which Park Place Technologies has applied AI. Other B2B leaders will be soon applying the predictive service model to their offerings, so they too can make their customers’ problems disappear before they happen.