How AI Will Make Customer Service Reps Better At Their Jobs
This article originally appeared on behalf of the Forbes Technology Council, a community for world-class CIOs, CTOs, and technology executives. Read the original post here.
Good customer service differentiates companies and directly ties to top-line revenue. We may only think about these roles when we have a truly terrible or utterly amazing experience, but they're critical in helping us navigate the products and services we use every day to live, work and play.
What isn’t immediately obvious is how important they are to our wider economy. There are nearly 2.8 million of these jobs in America, making customer service one of the top 10 jobs in the country.
Historically considered nothing more than a cost center, customer experience departments are starting to be seen as critical ties to creating the positive, enduring relationships companies want to have with their customers. Emphasis is starting to be placed on customer service teams because let’s face it: They have the most touch points with the customer, lasting long after the initial deal is closed. They play a massive role in customer loyalty and propensity to renew or upgrade services.
Increasing customer retention rates by 5% can increase profits anywhere from 25% to 95%. A recent report showed that if a $1 billion revenue company has just a moderate increase in customer experience, it can generate an increase of $823 million over three years.
Despite this trend, the vast majority of technology companies today that are building offerings for customer service teams are focused on creating technology that distances them from their customer. Artificial intelligence (AI) is being hailed as a way to either deflect the customer away from engaging the customer service professional or is used to create bots programmed to simulate conversations between companies and their customers.
It wasn’t that long ago that the customer service industry made this same mistake. In the early 1990s, there was a wave of companies that thought they could outsource support overseas to be handled by teams that had substantially lower cost structures. This was a huge failure; any short-term cost savings realized were massively outdone by frustrating customers who were driven to competitors. Years later, the outsourcing industry is alive and well, but the whole process from top to bottom was revamped to make it work.
In the AI era, the same result will happen. While it seems like a fun technical challenge, in reality, AI is far from understanding human emotions like empathy. A customer may already be frustrated by the time they reach out to your support. Greet them with a cute bot that generically asks them how their day is going, and you can expect a bad situation to get worse.
More importantly, a conversation, even in a messaging paradigm, is a very common back-and-forth that we are used to having all day with our friends and family. The accuracy of AI is not there for many of these use cases, resulting in strange and unnatural interactions between the bot and the customer. Even when the conversation is right from a technical perspective, it can still miss the point.
I recently had an incident with a ride-sharing company where the car looked like it was at the pickup point on the app, but was nowhere to be seen. I wrote to support to let them know they may have a bug, but the bot cut me short after my first reply with “Sorry about this! You won’t be charged for this ride!” before I could explain the reason I was writing in.
Putting algorithms in between you and your customer sounds appealing from an academic and technical perspective. But the reality is that this technology lowers CSAT scores, frustrating customers instead of actually solving their problems.
Despite all this gloom and doom, I am actually a huge believer in AI. Much like the last massive technology shift (cloud computing), AI will drive a fundamental transformation in enterprise software. When applied in the right areas, it can make humans much more impactful at their jobs.
There’s a much bigger and better opportunity to leverage AI to help customer service professionals be better at their jobs versus replacing them. AI can be extremely useful at automating many low-level tasks that customer service teams do all day, like categorizing tickets by topic or severity for future analysis, proactively coaching service agents with knowledge that they need to reply to specific support issues or alerting the agent who is an expert on a given topic to assist in escalating more complex support issues.
These are all things that slow down agents today and lengthen the time it takes to solve a problem for a customer. The value of good service, both literally and figuratively, is exponential: Memorable customer experiences facilitate upsells and renewals and drive revenue. This is driven by humans, not machines. So let’s have the AI empower the humans so they can spend more time doing this work and less time doing redundant, low-value tasks.
A perfect example of this is playing out in the real world right now. T-Mobile is differentiating itself from other carriers with its focus on customer service. They talk about “the hated phone menu and call center runaround” we’ve all experienced, and putting customers first with “No bots. No bouncing. No BS.”
We’ve all felt the pain of getting the runaround from a machine when trying to deal with a support issue, and customer satisfaction drops significantly when bots get involved (no real surprise there). This approach is misguided, even for simple questions like "How do I reset my password?" because every customer touch point is an opportunity. Companies like T-Mobile recognize this and are converting customer experiences done right into loyalty and revenue.
It’s not just morally important, but in the interest of all businesses to look at how AI tools can help humans be better at their jobs. AI can play a critical role, not by automating jobs away, but in amplifying humans to perform better.