In the latest episode of You Should Know, hosted by William Tincup, Assaf Bar-Moshe, Chief of Research and Development Officer at HiringBranch, presented new research that challenges the conventional practice of scoring soft skills in isolation. The study, which focuses on frontline hiring, suggests that evaluating empathy, acknowledgment, active listening, and reassurance as separate metrics leads to weaker predictions of job performance.
Bar-Moshe, a trained linguist, explained that HiringBranch's assessment method is grounded in sociopragmatic analysis—examining the actual words candidates use in scenario-based, open-ended responses. The research found that single-skill scoring shows only moderate correlation with human annotators, while a combined proprietary model demonstrates much stronger predictive validity. This finding is critical as employers increasingly seek to measure soft skills in candidates for customer service, sales, and retail positions.
The episode highlighted the four pillars of customer service that HiringBranch measures: acknowledgment, reassurance through positive language, empathy, and active listening. Bar-Moshe emphasized that these skills are interdependent. "If a candidate can express empathy, but is unable to solve the issue correctly or to comprehend the issue correctly or to reassure the customer, then this empathy is nice, but it's actually useless," he said. This perspective underscores the need for a holistic evaluation that mirrors real-world customer interactions.
The discussion also covered how HiringBranch translates job descriptions into conversation flows and scenario-based assessments, calibrated per client, region, and role. Regional variations are significant; for the same role, scoring weights differ across markets like Vancouver, Toronto, and Montreal. This localization ensures that assessments reflect cultural and linguistic nuances, which are crucial for accurate predictions.
HiringBranch's approach is validated by tracking on-the-job performance months after hire. The company's team of IO psychologists and linguists uses years of textual data to build machine learning models that predict soft skills, then validates these predictions against actual job outcomes. This longitudinal validation strengthens the credibility of their findings.
Bar-Moshe also previewed a self-serve capability in development, which would allow hiring managers to build assessments from a library of conversation flows and skills. This feature aims to reduce reliance on weak or generic job descriptions, enabling more tailored and accurate evaluations. The full study will be available under the AI research tab on the HiringBranch website.
The implications of this research are significant for HR professionals and talent acquisition leaders. By moving away from isolated soft-skill scores, organizations can improve the precision of their hiring decisions, leading to better hires and reduced turnover. As frontline roles become increasingly customer-centric, the ability to assess combined skills in context is not just a competitive advantage but a necessity.
The episode, part of the WRKdefined Podcast Network, is available now on major podcast platforms and at the You Should Know Podcast page. For those interested in the intersection of linguistics, AI, and hiring, this research offers a fresh perspective on optimizing frontline talent acquisition.
