How Zendesk AI will work in a real world scenario session #ZendeskRelate2023 #IntelligentCX #OpenAI




Hemanth of Alliance Department here. In this blog, I will try to explain a session of Zendesk Relate 2023 which is a real world scenario of how Zendesk AI will work.

Speaker for the Session

Zendesk AI

This is a new intelligence layer which powers the customer experience, enhances every customer interaction to make it more efficient and more helpful to customer. It is Zendesk at its core fast and user friendly. Knowledge search makes it faster more powerful which is accurate and all this works behind the scenes. All the models in Zendesk AI are pre-trained there is no model building on customers hand.

Zendesk AI in Action

Considering a scenario that you have ordered a pair of sunglasses, you are waiting for it's arrival and you get the following message in channel you like the most. You reply to it with a context here it's about fishing Now AI does it's magic and gives a suggestion for upgradation of the sunglasses based on context here it is about best glasses to use while fishing From that you pick one as polarized lens are best for fishing So how does the above suggestion come up, it's because of Intent, Sentiment and finally the language. Personalised product recommendation is done here is with the combination of AI and shopify integration which enables skill inventory and location data straight into conversation. Right now the feature is available in beta version Now it upscaled with a discount and created a coupon based on past purchases and preferences Finally with delivery notification, the product is delivered Now that is Intelligent CX Viewing a scenario with agents involved that is when a bulk order is placed Bot hand's it over to the best agent who has access to unified workspace, such as customer status, order history, past interactions and many such from agent side. Now with the help of intelligence panel and Summarize, the agent can get a summary of the conversation, suggestion of macros to get back to customers quickly, in this situation it recommends customer to be a corporate buyer Now after customer agrees the agent writes few words for response Zendesk AI does it's magic by making use of toneshift, it expands the message and makes it similar to a friendly reply.


The Evolution of how agents works in the future is where agents will be not be manually triaging tickets, searching for macros or modifying responses. They will be offering solutions and deciding what's appropriate to solve an issue for a customer. Hope with this you have an understanding of how Zendesk AI will work in a real world scenario.