Empower your Virtual Assistant to learn from your Knowledge Base

Virtual Assistant to learn from your Knowledge Base

The criticality of customer service has multiplied manifold over the last few years. What has always been a cost line item in enterprise balance sheets, is now being looked at strategically. And not just good customer service, consumers today expect great self-service. They want to get things done on their own and getting help from a human is often not the first thing on their mind. Rightfully so, self-service has emerged as a key pillar in the CX strategy of many enterprises.

Implementing such a strategy, however, requires careful orchestration. Most enterprises stitch together multiple best-of-breed solutions on top of larger suites. For instance, an enterprise might use Salesforce as the CRM solution but will leverage Haptik as a consumer-facing virtual assistant. Hence, such strategies require enterprises to work with a multitude of CX vendors while at the same time, ensure that all the individual investments amplify each other’s impact. An interesting challenge emerges when multiple best-of-breed solutions are implemented by enterprises in isolation and these solutions don’t talk to each other. The interoperability of SAAS solutions, thus, is key in ensuring that the potential of the individual solutions is realized to its fullest. 

Connect your Virtual Assistant with a Knowledge Base - Make the most out of your self-service investments

Keeping self-service as a key component of the CX strategy, enterprises spend a good amount of time and resources in setting up their consumer-facing knowledge bases. These knowledge bases (e.g Zendesk Guide) act as a single source of truth for all the level 0 queries for end consumers and enable self-service. Taking a step further from there to provide self-service on level 1 escalations, enterprises are increasingly adopting virtual assistants.

However, an interesting problem emerges here for many enterprises. Consumers develop a habit of doing repeat interactions with virtual assistants owing to a simpler experience. Since consumers begin querying the self-service requests on the virtual assistant itself, this puts the knowledge base investments of enterprises at the risk of being cannibalized by the virtual assistant. At the same time, enterprises end up spending a larger effort on ensuring that both customer service solutions - Knowledge Base and virtual assistants - stay up to date with the latest business information.

Read More: 5 Ways Intelligent Virtual Assistants Have an Edge Over Chatbots

The Solution - Knowledge Base Integration

To solve this conundrum, Haptik is launching a new platform capability - Knowledge Base Integration. Haptik’s intelligent virtual assistant can seamlessly connect with an enterprise’s knowledge base - be it in the house or third party knowledge base (e.g Zendesk Guide) - and accurately deliver the most relevant knowledge base articles for the consumer queries. This is a classic instance of Haptik leveraging its best-in-class natural language capabilities to deliver superior CX and business value.

HubSpot Video

 

Benefits of Knowledge Base Integration

The interoperability between the virtual assistant and the knowledge base supercharges the virtual assistant to answer a wide range of queries, which otherwise would have required a long time for the virtual assistant to learn. To add, this increases the capability of the virtual assistant to contain a larger proportion of consumer queries and prevent handover to human agents.

At the same time, enterprises can achieve a great degree of agility amidst fast-changing business scenarios. With this integration, any piece of knowledge or content that’s updated in the knowledge base is automatically incorporated into the training algorithms for the virtual assistant.

But how does the whole solution really come to life technically?

It’s quite simple! Haptik’s virtual assistants connect with the Knowledge base content via the REST APIs. The underlying Natural Language Understanding (NLU) models of the virtual assistant are trained on the knowledge fetched from the knowledge base. Later, when a consumer sends a query on the virtual assistant, the NLU models search for the most relevant article in real-time and surface them to the consumer. The consumer can read the detailed content in the article and continue their conversational journey.

As a result, the consumers end up with a better CX, enterprises see lesser human escalations and overall, higher ROI from the virtual assistants. Thus, a seamless amalgamation of the knowledge base and virtual assistants is a classic example of a best-in-breed platform like Haptik working in tandem with another CX solution to compound each other’s capabilities.


Try Now

Related Articles

View All