For General Motors, which sells about 10 million cars a year in dozens of countries, keeping a finger on the digital discourse isn’t always easy. The challenge is how to reach a global community of car owners across countless social media channels, call centers, email inboxes, and message boards with a comparatively small customer relations team.
GM found the answer in Crimson Hexagon‘s BrightView for Customer Care, an artificially intelligent platform that scrapes conversation data from social media, emails, chat logs, call transcripts, and almost any other source and automatically flags items of interest. GM uses BrightView alongside its in-house social media management software, Spredfast, and Opal.
“We saw a 3 times gain in efficiency,” said Rebecca Harris Burns, global head of GM’s Social Media Center of Expertise, in a phone interview. “It’s super important for us to be able to understand what’s going on around us. It allows us to put things together and make better business decisions.”
The BrightView platform helps customer service reps make sense of data deluges. A live dashboard lets them drill down into volume, sentiment, and content feeds. A handy search tool called HelioSight allows them to filter content by location, language, and social network, and to explore audiences (by demographics and interests), conversations (by top words, phrases, hashtags, and popular content), and images (by top logos). And data analytics tool ForSight can search through conversations and automatically categorize the sentiment (e.g., positive, negative, neutral), emotions (anger, fear, disgust, surprise, sadness), topics (themes), audience (interests, demographics, and psychographics), and images (logos, objects, actions, scenes, and facial characteristics).
But the two standouts of the suite are personalized insights and predictive alerts. The former highlights parts of conversations relevant to a customer representative’s department or specialization, while the latter looks for shifts and anomalies in volume, sentiment, and themes, proactively notifying the team when a trend — or a crisis, in some cases — emerges.
GM uses personalized insights to track conversations across different channels and facilitate interactions with individual customers, Burns explained. Thanks to a built-in prioritization feature, the company’s care team — which receives 33 requests every minute — can quickly determine which engagements are particularly time-sensitive. Customers asking for product assistance or inquiring about a safety feature, for example, take precedence in the system over a Twitter exchange about car commercials.
“We have a lot of detailed interactions in forums and within places like Reddit,” Burns said. “Without AI, it’d be difficult to get in there and understand what’s happening. We’re helping customers rather than sorting through information.”
BrightView data has even informed product design. One of GM’s full-sized trucks had a heated front driver seat with a vent that blew hot air into the back of the passenger roofing. The system flagged complaints on social media and in chat logs with customer reps early enough that the engineering team was able to redesign the air vent before new models hit the road.
Such AI would seem to be the harbinger of a completely automated customer service department, or at least one significantly reduced in staffing. But Burns is adamant that BrightView isn’t eliminating jobs. Instead, she said, it’s helping human operators focus on what’s important by cutting down on tedious, repetitive tasks.
“We have 72 models [of vehicles], and each model has three trim levels and complexity beyond what you can imagine,” she said. “You have to get into the details [and] understand what the issue is. That’s not going to be done by AI for quite some time.”
Crimson Hexagon, which derives its name from the boundless library described in Jorge Luis Borges’ The Library of Babel, was founded in 2007 by social scientist and Harvard University professor Dr. Gary King. In addition to customer service insights, it offers solutions that help companies monitor the conversation around product launches.
Samsung, another of Crimson Hexagon’s noteworthy clients, used its tools to perform competitive analysis ahead of a major smartphone launch. The South Korean company discovered that battery life was consistently one of the most buzzed-about topics, so when launch day rolled around, its public relations and promotions teams made sure to emphasize the new Samsung phone’s battery life above all else.
“Regardless of what industry you’re in … you’re working across time zones, languages, and [channels],” said chief marketing officer Lou Jordano. “It’s not an easy problem to solve. Organizations are drowning in a sea of data but starving for insights. They’re looking to process that information with a high degree of relevancy.”
The company’s online data library consists of over 1 trillion social media posts, blogs, forums, and news sites, and its customers include Anheuser-Busch InBev, Adidas, General Mills, General Motors, Paramount Pictures, and Pew Research.