Innovation without disruption: virtual agents for hyper-personalized customer experience (CX)

Chatbots and IVRs are contact center staples, but most still provide automated service for basic, repeatable tasks. What about when a customer needs to be intelligently routed to a better resource or has a question that’s not so cut-and-dry cut. Virtual Agent, or VA, is the next natural step for significantly better customer and business outcomes. 

What is a VA?

VAs make use of automation and a host of AI technologies like machine learning (ML), natural language processing (NLP), sentiment analysis, language translation, speech-to-text, intent recognition, and robotic process automation (RPA). Together, these technologies provide much-needed speed and efficiency – cutting down on the time needed for customers to interact to solve problems and have questions answered – while improving Customer Experience (CX) with data-driven hyper-personalization. VAs can take the form of a chatbot as well as integrated IVR systems that use AI to power conversational experiences (unlike standard IVRs, which limit users to saying a set of specific keywords). 

How do VAs achieve hyper-personalization? 

VAs help “hold the fort” on routine calls so live agents can focus more on complicated interactions, but they’re smart enough to handle certain complexities on their own. They can effortlessly navigate topics, handle a wide range of questions, and seamlessly operate across multiple channels. 

The technology also grows in intelligence with use, allowing VAs to act with greater – comparably humanlike – awareness. For example, you might present a customer with a choice of channels for engagement such as chat, phone, and social media. After communicating with the customer, your VA can default to that person’s preferred channel for future conversations. 

Another great example is intent recognition. A customer may say, “I want to know when my order will arrive” when their intent is “I want to track my order.” If the VA is programmed skillfully to ask good clarifying questions, it will come to learn the customer’s intent, store this knowledge, and apply it to better navigate future conversations. For situations in which VAs directly interact with external customers, IBM reports the average intent recognition is 70%.

VAs can hyper-personalize even routine interactions. Let’s say a customer initiates a chat session with a VA for resetting a forgotten password. The VA can ask the customer if they would like to switch to text messaging for a more effective multimedia experience. If the customer accepts, the chat session will end and the VA will seamlessly switch to SMS. In this case, the VA may send links to related resources or a “how-to” video that’s directly embedded in the message, so all the customer has to do is click “play.”

This simple enhancement is a major service differentiator, and large enterprises especially benefit from the ability to digitally deflect high incoming call volume with a promising containment rate (64% on average, according to IBM). Let’s look a bit closer at the metrics surrounding VA technology.

VAs by the Numbers 

VA technology isn’t an end-all solution but, when done right, has a positive aspect on nearly every measure including customer satisfaction, agent satisfaction, and revenue. The aforementioned IBM report leaves no stone left unturned:

Proven cost savings: 99% of organizations polled by IBM reported a reduction in cost per contact as a result of using VA technology. For a large organization, it’s estimated VAs can save $5.50 per contained conversation. Respondents also reported a 9% increase in revenue directly attributed to their use of VAs.

Faster handle time: Respondents, on average, reported a 12% decrease in human agent handle time and a 20 percentage point increase in first contact resolution as a result of using VAs. For VA leaders (organizations that implemented early, integrated VA tech with backend systems, and trained the tech on high volumes of contacts), handle time dropped by 15%.

Greater customer and agent satisfaction: Some organizations fear that VAs will adversely affect customer and employee satisfaction; however, IBM found average improvements of 8 and 7 percentage points in customer and agent satisfaction, respectively (VA leaders saw 12 and 9 percentage point increases). 

Overall, 94% of VA leaders have already achieved or exceeded their business case and 96% of all respondents have exceeded, achieved, or expect to achieve their anticipated return on investment for VA technology. 

Can on-premises enterprises with highly complex environments deploy VAs?

Absolutely. As contact center transformation explodes all around, hybrid cloud charts a clear path toward innovation without the disruption of throwing away what works. Well-established enterprises can overlay innovation by tapping into cloud-powered tech like AI, automation, and ML (required for virtual agents and endless other competitive solutions) while safeguarding the stability of existing operations. 

Enterprises can achieve innovation without disruption even faster with a ready-to-deploy, turnkey, configurable VA service. These AI-powered VAs do it all – recognize intent, communicate in dozens of languages, and handle a range of complex inquiries via both chat and voice calls – and can be launched and scaled within minutes. Many organizations start immediately leveraging their VAs to help manage incoming call volume and then incrementally introduce new value in line with hyper-personalization.  

If you’re thinking of using VAs externally for interaction with customers, there’s no better time and no easier way to currently do so. Speak with one of our experts about Avaya’s no-code, ready-to-deploy VA solution. 

If you want to learn more about hybrid cloud for innovation without disruption, Avaya just published its latest report in partnership with Ventana Research on hybrid cloud migration for contact center innovation. It’s a crucial asset for enterprises at every stage of the hybrid cloud journey, from consideration to execution. You can check out the free report here.