Operationalizing Contact Center efficiency with AI

Ruth Yakubu

Challenges of delivering customer satisfaction

Excellent customer experience leads to loyal customers. When it comes to working with contact centers, there are multiple customer service factors that lead to unhappy and frustrated customers. These include the wait time; being bounced around dept agents; repeating issues; providing sensitive data on the phone; or getting inconsistent information. We live in a world of instant gratification, when a customer has a problem, they want a response right away. The longer the response time, the more the customer gets frustrated. According to Forrester’s research, “63% of customers will leave a company after just one poor experience, and almost two-thirds will no longer wait more than 2 minutes for assistance”. For businesses, simply hiring more employees to reduce wait time is very costly. On the other hand, some make the mistake of reducing cost by cutting the workforce and focusing only on prerecorded phone messages. However, customers end up spending most of their time pressing multiple numbers on the phone that often leads to a dead-end with a broad answer that leaves the user unsatisfied. Businesses needs to find a way to balance reducing cost when improving customer support more efficiently.

Modernizing call center communications

In this episode of Tech Exceptions, I had an exceptional conversation with Anand Janefalkar, CEO of Ujet, who have raised over $114M in funding for their AI-Driven real-time communication platform that makes things easy for customers and businesses to instantly connect so they can resolve problems faster. Ujet’s Virtual Agent supports intelligent conversational AI for a more natural human-like conversation across voice and messaging channels. In addition, their platform brilliantly bridges the gap of virtual and human agents by integrating with any organization’s CRM system. In this episode, we discussed how the company is going beyond the traditional chatbot by providing call center support using CRMs and natural language processing (NLP) to handle cyber security, chatbot state management and customer emotions for multimodal agent support to provide exceptional customer experiences.

Leveraging NLP for customers to feel heard

Using AI and machine language, virtual agents provide an opportunity for customers to get their questions answered in real-time through chatbot conversations. Ujet uses natural language processing (NLP) to understand, interpret and be able to respond to customers. By providing the virtual agent intelligence, Ujet can understand a user’s intent in a conversation even when the customer misspells words or speaks another language. This allows the solution to accommodate more users. The solution empowers contact centers to be inclusive and scale to a wider audience that supports multiple languages. This could be a demographic that traditionally call centers do not have the costs hire employees that are able speak multiple languages. Having a virtual agent provides an immediate response to issues or a ticket to be created for a specialist to address which helps reassure customers that their issue has been acknowledged. Thus, this automatically enables customers to feel heard.

Image NLP icons

Leveraging NLP and CRM for intelligent virtual to human routing

Ujet uses NLP’s sentiment analysis to be able to detect a customer’s tone and emotion in text or voice. Ujet’s virtual agents don’t just understand phrases but can identify various cues like urgency or frustration and be able to adjust their responses based on a customer’s emotion. In a situation like this the virtual agents quickly identifies a problem and automatically routes a customer to a human agent for the fastest resolution.

Ujet’s intelligent platform integrates with CRM systems like Microsoft Dynamics, Zendesk, or Salesforce to manage all customer communications. This helps call centers have a centralized system that unifies data from all channels that interact with their customers. In cases where the solution escalates an issue by routing a virtual agent to the appropriate human agent, the platform’s contextual routing pairs real-time customer requests with the prior conversation information and communications to a CRM.

AI’s bot state management is important to enable chatbots to have more meaningful conversations by remembering information that the customer shares. Bots need this session information to be saved, so they don’t have to ask the customer the data again. Ujet’s platform uses CRMs as a data source to store conversations from the virtual agents. This is essential because it gives human agents transparency to the customers communication history and it also helps avoid having the customer repeat their issue over and over again.

Agent routing CRM

Leveraging NLP for Security & privacy

When dealing with call centers, customers oftentimes must provide sensitive data such as date of birth, address, social security number or credit information. This exposes customers to potential hackers. NLP text classification models are used to identify personally identifiable information (PII). Ujet uses NLP to recognize and mask PII information from text. The solution enables human agents to use Virtual Agents as their own virtual assistants to protect customer privacy. For instance, if an agent needs to gather a customer’s credit card information, instead of asking the customer to give their credit card number to the human agent, which presents an array of security concerns, the human agent could transfer the customer to the Virtual Agent just to give their credit card number. Once the Virtual Agent has the necessary payment information, the customer can return to their call with the human agent with the peace of mind that their personal data is secure.

The Solution

Watch and learn how startup Ujet is modernizing digital and in-app experiences by unifying the enterprise brand experience across industries and eliminating customer service frustrations from multiple channels using CRMs and AI’s natural language processing (NLP) to handle cyber security, chatbot state management and customer emotions for multimodal agent support to provide exceptional customer experiences.

Coming up Next…

Mark your calendars! I’ll be joined by Rajib Saha, CEO of Parabole.ai, on how the startup is a game-changer for digital twin implementations, with their AI platform to optimize enterprise’s data-driven decision-making with simulations to uncover data mapping, data lineage & data catalogs. In this episode, we will discuss how the company is using neural networks to create ontology from their datasets and reinforcement learning to generate advanced knowledge graphs to drastically accelerate time to market deployments. Watch Live stream on Learn TV March 24th 2:00-2:30pm PT!


Discussion is closed.

Feedback usabilla icon