AI slashes call center training time, boosts agent skills
NEW YORK, UNITED STATES — The world of call centers is often exposed to outdated training, which is expensive, yet the people who ultimately face customers are under-equipped.
A published article in McKinsey & Company by Devansh Sharma, Shikha Sharma, Subhrajyoti Mukhopadhyay, and Vinay Gupta, notes that generative AI-based simulations are transforming onboarding, making it nearly training-free and tripling performance indicators.
Traditional call center training is inefficient and costly
Current training methods in call centers are slow, expensive, and fail to prepare agents for real-world interactions. It takes four to six months for new hires to reach peak performance, which costs companies 5% to 10% of their total expenditure on agents.
Most of the training is generic; not tailored to individual skill gaps, which results in poor performance and stagnant metrics. Agents are often ill-equipped, and learning in the classroom usually fails to translate well to handling live calls.
Moreover, performance gaps are exacerbated by the fact that during recruitment, essential skills such as empathy or the ability to handle customers in real-time are rarely examined.
No formal evaluation, especially in a remote working environment, can prove difficult in terms of monitoring the progress made by new employees, which managers may struggle to assess.
AI-powered simulations deliver faster, smarter training
The AI platforms provide agents with an opportunity to train in a secure environment on various customer scenarios, offering immediate feedback and enabling agents to refine their work. This type of practical training may accelerate the learning process and establish trust before working with real calls.
Automated evaluations monitor primary indicators, such as first-call resolution (FCR), average handle time (AHT), and best practices violations.
One of its instances included a technical help desk that utilized AI simulations, where positive interactions increased by approximately 33% and interruptions decreased by approximately 28%.
Additionally, roughly 80% of coaching duties were automated. Such an approach, based on data, allows for constant improvement, whereas manual monitoring is minimized.
