How to use chatbots, the Pega way

Customer service conversations don’t have to be difficult. Pega’s chatbot gets smarter the more solutions it finds, without the need for complicated coding.

Customers need help day and night, for problems big and small. Beginning the resolution process instantly via chat tools like Facebook Messenger puts their minds at ease as the system works hard to find the swiftest, most satisfactory solutions.

The power of Pega. The convenience of a chatbot.
Using AI and NLP (natural language processing), Pega’s chatbot attempts to resolve customer concerns before passing them along to an agent. The chatbot shares information with agents to solve more complex issues so customers don’t experience the frustration of relaunching their complaint.

Every interaction is an opportunity
Using the Pega Customer Decision Hub™, agents are empowered with relevant offers to present to the customer, bringing value and a chance for increased revenue with every interaction. This valuable function is available, again, without the need for cumbersome coding.

Agent feedback for an automated future
The system learns from agent feedback, so similar inquiries can be solved in the future without the need for valuable agent resources. The more customers use it, the better, smarter, and more powerful your chatbot gets.

Learn more about the Pega’s chatbot in the video above.

Transcript:

In this demo, we'll show you how easy it is to use a chatbot to help your customers quickly self-service on basic questions, and seamlessly transition to an agent if more service is needed. This demo was created using the Pega 7 platform with our Pega Customer Service Application using simple configurations. No code was written. Say, for example, you are a shipping company and your customer, Mr. Chekov, who paid for expedited shipping, noticed that his shipment did not arrive yet. So Mr. Chekov decides to use Facebook Messenger and is first greeted by a bot that tells him what it can help him with. So Mr. Chekov informs the bot that he thinks his package is lost. Using natural language processing, the bot recognizes Mr. Chekov is looking for information on the status of his package and performs a quick query of the tracking number. Based on tracking number, we now see that Mr. Chekov is a premium customer and his preferred method of contact is chat.

The bot suggests if he would like to speak with an agent. This triggers a screen pop on the screen of an agent. What you see here is Pega's Customer Service Application. Because both the chatbot and the customer service portal are using Pega technology, we can have an immediate seamless transition from an automated to a personal interaction. Mr. Chekov's information and package information is preloaded into the agent's portal based on the tracking number that was entered in the bot. Mr. Chekov explains the issue while continuing to use the Facebook Messenger app. The issue comes down to, well, missed expectations for when he should have received his package. The agent can quickly identify the problem and clarify the issue with the customer.

In addition to resolving Mr. Chekov's issues, we also wanna identify why this conversation needed to leave an automated channel in the first place. Once the agent has resolved the issue, he will use the information from the conversation to create a feedback item. This ensures that we address any functionality that may be missing from the automated system. The agent will also check for any offers that may be available to this specific customer, thereby using this personal interaction as a chance to offer something of value. The agent fills out a feedback item regarding the cause of the interaction.

This feedback item will be routed to the appropriate group to identify gaps in the current system and provide greater visibility into the customer experience. The agent then checks for any offers that may be available for Mr. Chekov and is able to offer a coupon. The coupon offers one of many possible follow on actions, but it was determined to be the most relevant by Pega's Customer Decision Hub. This smart offer could also have been made via the chatbot without having to involve an agent. The intelligence behind this is not channel specific and is easily configured in Pega without having to write any code. After confirming that the offer has been sent, the agent begins to wrap up the interaction. Mr. Chekov has quickly reached a solution and a follow-up satisfaction survey email can be sent if desired. The agent specifies the reason for the interaction and is given the option to email a log of the chat to the customer. This is how you deliver a seamless bot to agent experience without writing any code.