Analytics Move to the Hiring Process

Agent turnover at the typical contact center ranges from 30 percent to 45 percent annually, with some sectors reaching as high as 60 percent, and agent tenure averages just 13 to 15 months. Both of these realities make life difficult for contact center managers and the people responsible for contact center staffing.

And even though the U.S. Bureau of Labor Statistics projects that employment opportunities for customer service representatives will decline 5 percent between 2024 and 2034, hundreds of thousands of job openings are still expected each year.

In fact, the bureau expects to see about 341,700 openings for customer service representatives each year over that time frame, with companies still needing to replace contact center agents who transfer to other occupations, retire, or exit the labor force.

The Bureau of Labor Statistics maintains that much of the expected job loss in the contact center industry will be driven by increased automation.

Ironically, that same automation is increasingly being used to fill the available positions that remain. Any one of the hundreds of thousands of job openings in contact centers could have thousands of applicants hoping to fill it, and artificial intelligence is taking a bigger role in sharply narrowing down the field to the best potential fits.

Human resources departments alone simply don’t have the people or time to comb through all the resumes, interviews, references, etc., to ensure that contact centers under their purview have enough agents and the right mix of agents with the right skills for the job at hand.

“People analytics has been around for a while,” says Rebecca Wettemann, founder, CEO, and principal analyst of Valoir, who notes that in the HR domain, analytics has largely been restricted to cost-containment metrics like adhesion and average call handling time.

“We’re starting to see contact center managers look for new things that reflect the changing nature of contact center metrics being driven by AI,” Wettemann observes.

The growing use of people analytics for contact center hiring comes as no surprise to many across the industry either.

“Just to cycle through the sheer volume of hires that we have to do requires a tool,” says Trudy Cannon, senior director of go-to-market strategy for workforce engagement at Verint. “If you don’t [use analytics], and use a manual process, you can only screen so many potential employees a day, depending on the number of recruiters that you have.”

Verint in early 2024 introduced the Interviewing Bot, a solution that automates the initial stages of candidate screening and assessment. It uses natural language processing to analyze candidates’ responses to prerecorded video or text-based interview questions, extracting key information and assessing communication skills. It also detects the emotional tone of a candidate’s responses and uses data and machine learning algorithms to identify candidates with the highest probability of success in the specific role.

Candidates complete prerecorded video or text-based interviews. These interviews typically focus on job-specific skills, experiences, and behavioral questions to help determine energy, engagement, empathy, language proficiency, and problem-solving skills.

The Verint Interviewing Bot analyzes the responses and then generates a score for each candidate and more detailed reports summarizing each candidate’s strengths, weaknesses, and overall fit for the role and which candidates are most likely to be high performers, stay longer, and better achieve service-level targets.

“If that initial assessment helps me uncover the right attributes, like communication skills, language proficiency, etc., then we can drill down from there,” Cannon says.

Another widely touted AI tool for agent hiring is Vervoe’s Customer Service Simulation, which Rohit Agarwal, cofounder of Zenius.co, a staffing company specializing in remote hires, calls “a must-have tool for our selection process for customer service candidates.”

“The Vervoe algorithm automatically ranks the performance of the candidate in real-life situations and makes it easier for us to assess their overall performance,” Agarwal says.

Vervoe as early as 2021 introduced the Customer Service Simulator, an AI-based recruiting tool purpose-built to test empathy, learning agility, attention to detail, and problem solving. Candidates click a link and are immersed in an assessment environment that mimics real customer support scenarios, interactive tickets, extensive knowledge bases, and rich customer profiles.

Simulations across e-commerce, subscriptions, services, and technical support are all available out of the box, catering for a broad range of use cases.

The Vervoe tool enables Zenius.co to assess employees for specific queries that they will regularly have to face as customer service professionals, according to Agarwal. “In assessments while hiring for an e-commerce client, we’re able to deploy simulations specifically about orders lost in transit or delayed deliveries,” he says.

Agarwal endorses sellmethispen.ai as well, which he says is good for testing candidates with simulations of customer calls. This helps assess how they deal with high-stress situations, which are common in call center environments.

Rajeev Butani, CEO of MediaMint, a provider of revenue operations technologies, is also a big supporter of AI-driven simulation tools. “These tools allow associates to practice and role-play customer interactions in a simulated environment. This not only helps assess associates more efficiently but also reveals how effectively they will work alongside AI copilots. The insights gained from these simulations can inform both hiring decisions and ongoing training, ensuring a seamless and productive human-plus-AI partnership.

“Companies need to assess traits like learning agility, digital fluency, and problem solving, which are critical for an associate to successfully partner with and work alongside customer service AI agents,” Butani says further.

AI-enabled hiring tools, he adds, “are the foundation for ensuring our talent can bring the oversight and empathy required in a human-plus-AI delivery model.”

Other required skills in the modern contact center include good communication, empathy, active listening, product knowledge, and, as noted, problem solving. Since most of these are soft skills, AI-enabled hiring tools are important to turn them into measurable criteria, experts agree.

“Empathy, communication, critical thinking, and stress management skills have all become more important as agents are handling more complex and sometimes contentious interactions,” Wettemann states. “The traditional assessments that nailed down the old skills requirements don’t work very well for many of the new ones.”

Contact center hiring automation also needs to consider the roles that the potential agents will be asked to fill. “There are many different roles in the contact center, whether it’s supporting different industries, different customer types, different call types,” Verint’s Cannon says. “When we think about hiring, we have to be very specific about the skill set that’s needed for a particular role.”

Beyond Hiring

Once job applicants are brought on board, AI can help HR departments during their employment, starting with the onboarding and training process.

“We have to have tools that help them manage different interactions in real time. This helps reduce the stress of taking calls over and over, and it also allows me to be able to enable them right away for high-level situations,” Cannon acknowledges.

AI is also automating other HR-related tasks, such as compliance monitoring, performance management, and benefits administration, as well as crafting employee handbooks, employment contracts, and offer letters.

Additionally, contact centers are using AI assistants that can help agents get to the right answers, communicate those answers effectively in the right language and tone, and provide a quality customer experience. These are rapidly becoming must-haves for contact centers, according to Wettemann.

To further ensure that new hires meet expectations once brought on board, managers can also leverage AI quality control technology that enables them to assess every call at scale and in real time to evaluate relative quality and compliance and intervene on the agent’s or customer’s behalf when needed. Sentiment and empathy analysis capabilities can track when individual agents are having a difficult time or when more training is needed, Cannon says.

Beyond all that, when the time comes to part ways, AI can make the process seamless by drafting termination letters and other documents, all while ensuring compliance with local laws.

Protecting Against Fraud

Though probably not a main reason for the technology’s development, an unintended benefit of AI-based screening tools is the ability to filter out fake or fraudulent job applications. The problem, which is becoming a widespread issue across global job markets, and contact centers are no exception, involves legitimate job seekers who misrepresent themselves during the hiring process or lie on resumes.

More troubling, though, some recruiters have reported that up to 40 percent of all applications for certain vacancies are fake, often generated by bots or coordinated fraudsters overseas. The FBI has warned that criminals are using AI deepfakes and stolen personal data to apply for remote jobs in the United States, attempting to pass background checks by impersonating real identities.

These job candidates aren’t who they say they are and use AI to fabricate photo IDs, generate employment histories, and provide answers during interviews. Once hired, they can install malware to demand ransom from the company or steal customer data, trade secrets, or funds.

“The challenge for all HR pros and hiring managers right now is the sheer volume of applications they receive and the potential for fraud,” Wettemann says. “Ironically, some of the nice-to-have tools we’re seeing emerge in the space are AI interviewers that can both engage with a high volume of candidates that opt into engaging with them and screen out a lot of potentially fraudulent or inappropriate candidates.”

AI can be used to combat job application fraud by employing complex algorithms, facial recognition, and voice, video, and text analytics to detect suspicious patterns, inconsistencies, and AI-generated content, including deepfakes and AI-generated text, in job applications. These tools can flag synthetic content in real time, analyze anomalies in behavior or appearance, and adapt to new fraud tactics as they emerge.

One such tool is Greenhouse’s Real Talent, an AI-powered solution to help companies address the increase in spam, fraud, and cheating in the hiring process. The application, which Greenhouse launched earlier this year, helps employers with three main tasks:

• Filter for fit.The technology compares applications to job criteria to highlight qualified candidates and filter out irrelevant ones.
• Combat spam and fraud.It analyzes applications for signs of bots, mass applications, and fraudulent activity.
• Integrate identity verification.In partnership with CLEAR, the platform allows for secure identity verification by comparing a government-issued ID with a selfie to confirm candidates’ identities.

“With Real Talent, we are moving fast to give recruiters a competitive edge in identifying best-fit candidates while enhancing security against new and growing threats,” said Daniel Chait, Greenhouse’s CEO, in a statement at the time of the release.

Balancing Tools with Human Recruiters

While assessment, interviewing, and related AI tools are increasingly necessary just to sift through the sheer number of applicants for any positions, humans are still an essential part of helping to ensure that contact center hires have the right attributes for success, Cannon says.

Once they’ve used automated tools to winnow down the number of applications to a manageable few, recruiters can interview candidates to determine how well they can think on their feet, sell and upsell (if sales is part of the job), be empathetic with callers, etc.

“Automation will get us through that first pass,” Cannon says. “But we still need leaders who are versed in the work that needs to be done to be able to do those interviews.”

“It’s important to remember that automated analytics and AI—like any technology—are accelerators for the business management practices and processes already put in place,” Wetteman adds. “Managers are going to have to rethink their recruiting, onboarding, and incentivizing policies and practices to reflect the need for agents with a different skill set and rethink their roles as coaches, not supervisors.” 

Phillip Britt is a freelance writer based in the Chicago area. He can be reached at spenterprises1@comcast.net.