In our recent webinar, “Cost per Case: The Economic Impact of Conversational AI and Automation“, Dan Flagella, Founder and Head of Research at Emerj Artificial Intelligence Research discussed the impact of conversational AI and automation on B2B tech support with Uniphore’s Director of Product Marketing, Kim West, and Sales Director, Peter Stevenson. Couldn’t attend the live event? We have you covered. Read on for some of the highlights.
Understanding the B2B tech industry and its support needs
The B2B tech industry has a unique audience with a highly specialized set of needs. Unlike B2C customers, who contact customer support with typical single-user queries, B2B clients often use products on a broader—sometimes enterprise-wide—scale and, consequently need support on a wider range of more complex issues.
“What’s happening in the tech sector is very different from any other place where you’re seeing conversations between a support person and a customer in the B2B tech space,” West explains. “It’s large organizations with multiple employees—and they all need support.”
However, for the complex—and often highly technical—nature of its issues, the B2B tech sector lags behind other industries in its adoption of high-tech support solutions. The irony isn’t lost on West, who’s seen firsthand how solutions like conversational AI and automation have transformed other “low tech” industries. “What we’re finding when we deal with our clients in the B2B tech support space, is that we’re finding a lot of them are not even thinking about the idea of AI and automation being a potential solution,” she says. “The mindset isn’t there yet.”
Overcoming that mindset requires a step back, says West. “One of the insights we know from our experience is: let’s make sure we’re actually looking at the full problem and understanding what’s going on. A big part of what’s going on is that, within the tech support space, it’s not just one call or one email—it’s multiple iterations.”
“Curing” the complexity with AI and automation
Multiple query iterations add up—and they add unnecessary complexity to the already time-consuming task of tech support. West sees this as a prime opportunity for AI and automation. By applying the technology to each support conversation, companies can identify which issues are occurring most frequently and can create a consistent response—whether via FAQ menu, self-service portal or guided live assistance.
“If we start using technology that's listening to the entire conversation, it will track all of those different emails or chat conversations that are happened as well, so, the next time somebody calls in with a common query, we're going to be that much quicker in resolving the issue.”
Kim West, Director of Product Marketing, Uniphore.
Stevenson agrees. He sees AI and automation as vital to identifying and ultimately solving recurring support issues. He also acknowledges that the root of the problem may be deeper than it initially appears.
“The challenge with identifying what’s the true issue is oftentimes a very complex area because your technology, as a B2B tech company, sits in an overall technology stack,” he says. He recommends that companies look closely at three key areas:
Case classification
Are queries being properly diagnosed, classified and linked to the appropriate resources (i.e. knowledge articles)?
Troubleshooting protocols
Are desktop tools, knowledge base and supporting documentation (call description, intent, case logs, etc.) being leveraged effectively?
Consistency and quality
Are previous, similar/identical cases being leveraged to create a consistent response moving forward?
Applying AI and automation to tech support operations
The key to leveraging AI and automation effectively is to take a holistic approach, Stevenson says. “I always think about AI and automation as it’s a journey, not an endpoint. It’s pragmatic to think about that end-to-end view and look at potential cases [and] impacts.”
He recommends focusing first on uses cases with the fastest rewards. “Where [are the areas with] the biggest impact and value equation that you can get started?” he asks. “How [does your] tech debt impact things like customer experience, friction points or in the knowledge AI space?”
In addition to thinking end-to-end, Stevenson urges tech companies considering AI and automation to plan for the long term. As business needs and even use cases themselves evolve, the technology that supports them must keep pace and allow for continuous improvement. Having an agile, unified platform and system of intelligence makes it exponentially easier to make evolutionary adjustments that add real value.
“Because it’s more of a journey, you need to look at solutions that protect your investment over time,” he says. “You need to have an investment approach that says, ‘whatever I do from a system of intelligence, I know it’s going to be extensible to other aspects of the overall process.'”
Breaking the “break-fix” mindset
Taking a journey-wide approach is just the first step, says Stevenson. To get the highest return on their AI and automation investments, tech companies must abandon the “break-fix” troubleshooting mindset and consider how individual cases can benefit the whole.
“You [need to] rethink brake-fix as a historical model, overall,” he says. “Look for opportunities to say, ‘how can I provide intelligence in the product to be able to fix itself, to anticipate issues, to notify the enterprise of a potential issue and to have some sort of resolution before it becomes an operational impact?'”
A unified AI and automation platform is the answer. In addition to identifying and resolving recurring case issues with consistency, AI and automation, when used efficiently, can help CX systems self-diagnose and even self-correct issues within the platform—essentially fixing issues before they become larger problems.
According to Stevenson, it’s high time tech companies got on board. “Most of the world is kind of moving towards that,” he notes. “Having the product fix itself.”
Optimizing the tech support experience for maximum value
West agrees. She sees self-correction as part of the bigger value optimization picture. By reducing the time and effort required to resolve complex issues, tech companies can optimize their support operations and maximize their value. While this includes self-service, it’s more applicable to cases requiring live agent assistance. “How can we move more of the incidents that are coming through to a place where instead of it being a couple of hours, it’s actually a couple of minutes?” she asks.
Stevenson explains how tech companies can measure the success—and ultimately the value—of their AI and automation initiatives. In addition to classic KPIs, like resolution or turnaround times and error reduction rates, he sees one benchmark in particular as indicative of overall success.
“One metric that describes the health of the franchise is cost per case, because that's a reflection of how well you're driving the digital capabilities; how well you're driving the cases that have to be handled by an agent across the different sectors of performance.”
Peter Stevenson, Sales Director, Uniphore.
While measuring cost per case is certainly important, Stevenson cautions companies not to confuse cost optimization with cost-cutting.
“Driving down costs is obviously, from a tech support standpoint, an important metric overall, but it’s also that correlation [with] what does this do from a business revenue standpoint?” he asks. “How does this technology support experience? How does it impact a company’s commitment from a loyalty standpoint and a company’s satisfaction level or NPS scores? How does this impact the growth of the business?”
Tech support providers can maximize the value they receive on their AI and automation investments—and continue to receive over time—by taking a forward-looking, big picture approach:
- Start with key use cases and apply the lessons learned to the broader user journey
- Create a cycle of optimization that not only solves specific problems but also strengthens operational efficiency
- Shorten and simplify journeys to improve end user experience and revenue-specific metrics (i.e. contract renewals and retention rate)
- Refine and build upon the process to grow customer lifetime value (CLV)
Stevenson likens this to a ladder. “A lot of times I’ll find tech support companies can handle the question about what their cost per case is, but they don’t know the ladder, which is what does this mean in terms of service experience overall? What does this mean in terms of the aggregate impact for a particular enterprise account as it relates to retention, revenue, growth, product expansion and so forth?” He asks. “That’s where I see tech support moving away from the traditional break-fix [approach], which was always [viewed through] a cost lens [toward asking], ‘how does that affect the business in the continued growth and expansion?'”