Generative AI is reshaping industries at a remarkable pace, with one of its most profound impacts being the empowerment of frontline agents. A study by the National Bureau of Economic Research (NBER) revealed that customer support agents using AI tools experienced a productivity boost of nearly 14%.
By integrating AI technologies like generative AI, organisations can amplify agent productivity by up to 35%, allowing them to handle complex queries more efficiently.
[24]7.ai is emerging as a leader in harnessing generative AI to redefine customer service and agent experiences. With deep operational knowledge in contact centres and extensive experience with large language and generative models, [24]7.ai is pioneering AI-driven training methodologies tailored to diverse industry needs.
The Journey of [24]7.ai
In an exclusive conversation with AIM, Animesh Jain, chief operating officer of [24]7.ai, shared insights into the company’s evolution and technological advancements.
“We started as a pure-play company in 2001 as one of the pioneers in the industry when the sector was taking shape in cities like Gurgaon, Bengaluru, and Mumbai. Back then, we were known as ‘24-7 Customer’.
“Our initial decade was focused on BPO services, outbound telemarketing, email support, inbound tech support, and customer service. As the industry evolved, we became early adopters of digital transformation, helping retail clients transition from voice-based interactions to digital engagements around 2007-08,” he said.
By the early 2010s, [24]7.ai had begun making strategic acquisitions to enhance its technology stack. Acquiring Voxify and Telme from Microsoft allowed them to advance voice automation capabilities and build AI-powered applications for customer interactions.
“While LLMs became mainstream only recently, we had been working with them as early as 2011-13, leveraging AI for IVR automation. We automated 40-70% of calls for major clients using voice bots,” Jain explained.
Over the past decade, [24]7.ai has continued to push the boundaries of AI by integrating automation across both voice and digital channels. More recently, the company has focused on GenAI, exploring its potential in customer interactions and agent training.
Agent Training with GenAI
Traditional training methods often follow a one-size-fits-all approach, making it difficult to cater to individual learning speeds. “With GenAI, training becomes personalised. Trainees can learn at their own pace, significantly improving comprehension and retention,” Jain mentioned.
GenAI enhances training by enabling personalised learning, offering early exposure to customer environments, and bridging the experience gap. Early exposure to customer environments is facilitated through GenAI simulations, where agents practice interacting with a diverse range of personas, from an 18-year-old online shopper to a 70-year-old retiree.
These interactions help trainees familiarise themselves with different customer behaviours and cultural nuances before engaging in real conversations. Bridging the experience gap is another crucial advantage.
“The initial 4-10 weeks on the job can be overwhelming, as agents transitioning from training to live customer interactions often struggle with real-world challenges. GenAI reduces this gap by simulating various scenarios, including handling frustrated customers, allowing trainees to develop effective response strategies before facing actual clients,” Jain explained.
By leveraging GenAI, [24]7.ai has observed significant improvements in learning curves, emotional intelligence, and customer sentiment analysis. Agents trained with AI experience reduced average handling time (AHT), leading to higher efficiency and better customer satisfaction.
Challenges in Implementing GenAI
While GenAI brings massive advancements, it is not without challenges. One major challenge is to keep the models updated. AI effectiveness depends on data quality and recency. In fast-moving industries like e-commerce, frequent updates to offers and product details are crucial.
If GenAI relies on outdated data, agent training suffers. Continuous feedback loops are essential to keep AI models relevant and ensure that the information used for training remains accurate and up to date.
Another challenge is technology adaptation. While younger employees, particularly Gen Z, adapt quickly to AI tools, many other agents still struggle with AI-driven training. As future generations grow up with AI-integrated learning environments, adoption will become more seamless.
However, organisations need to provide additional support and guidance to help employees effectively transition into AI-enhanced training models for now.
What’s Next?
Fraud and data misuse have always been significant concerns in the customer service industry. Companies developed their own systems to address these challenges. Historically, these systems were highly labour-intensive, requiring manual transaction monitoring to detect irregularities.
Over time, they built their own analytics models to analyse data, identify patterns, and flag potential issues. For instance, [24]7.ai’s system can detect anomalies like repeated refunds to the same customer, which clearly signals a problem.
GenAI is set to accelerate this process significantly. While advanced analytics models are already used for fraud detection and compliance, these models take time to develop.
Industries have been advocating for an AI+HI (human intelligence) approach for the past seven to eight years. The vision has always been clear: automate simple, repetitive tasks while reserving complex interactions for human expertise.
For example, a straightforward customer query like “Where is my order?” can be easily handled by automation, as it mostly involves checking and reporting the order status.
However, when the scenario becomes more complex, such as a customer who has placed multiple orders and received conflicting updates, grows increasingly frustrated. Since the situation changed, now, both logical complexity and emotional intelligence come into play.
At this stage, while AI can provide information, a human touch is essential to understand the customer’s emotions, reassure them, and resolve the issue effectively. A human agent can explain delays, empathise with the customer, and ensure they feel valued.
This is how AI and human intelligence must work together—AI assists by providing insights and streamlining processes, while human agents handle complex, emotionally charged interactions.
