The omnichannel agent desktop

Hi, my name is David Fulton, and I am part of the Customer Service Product Management team at Pega. Thanks for visiting today. I want to spend a few minutes talking to you about chatbots, virtual assistants, and agent desktop.

Sara has sadly been furloughed from her job and is now in the unfortunate situation of making some really hard choices on household expenditures. She's logged into the U+ Communications website to check in on exactly how much she's paying for mobile, TV, and internet.

Right after Sara Logs in, she's greeted with a proactive notification delivered as part of the chatbot experience. And did that offer get selected, I hear you ask? Well, Pega AI has decided that this is the most relevant offer for Sara based on what we know about her payment and service consumption history and other data points from other U+ communications customers.

Replying "Yes" to that offer, Sara is guided to a menu of options on what to do next. Could Pega instead understand any request if Sara preferred type in the question herself? Absolutely. We simply made the decision on what to do next a little easier for Sara, who might need a little guidance on what this experience allows her to do.

The Pause Service option sounds intriguing to Sara, so she clicks on it to find out more and sees the set of options for services she can suspend without losing some of the perks of a triple-play package. Sara is watching more Netflix and less TV these days. And her internet package is something she simply can't live without. Sara decides to pause her TV subscription. The steps that Sara is going through are identical to the steps that would have been followed by agents if Sara had decided to call or chat. How? Well, this Pause Service process is built once in Pega and then specialized for different personas and points of engagement, helping U+ go from experiences in the past that were siloed, channel-led experiences to experiences that are channel-less.

OK, Sara still has a nagging question as to whether there are costs incurred in pausing a service. She neither wants nor can afford a nasty surprise on the next bill. Sara sees an icon that looks helpful. And look, there are a couple of articles referenced that answer common pause service questions. How does that work? Well, in Pega, you can build relationships between steps in a service process like Pause Service and knowledge articles. Again, just once for all potential consumers of that information, from customers to employees. As we know, Sara is counting every penny, and while a lot of questions have been answered, there is one question that the knowledge team haven't yet anticipated. And that's the value, at least to Sara, of spelling out those fees explicitly.

Needing to get an answer to that question before going further, Sara asked whether someone is available that can help her. At this point, Sara will be prompted for additional information before we request an agent. Doing data collection like this in advance of the request being routed, it's a big time saver for your chat contact center.

Tony joins the chat with Sara and immediately gets a sense of Sara's context, what she was doing through the chatbot, including the articles that Sara viewed prior to requesting additional help. Tony now has everything that he needs to answer that question quickly.

What about that pause service that Sara started earlier? Well, Tony doesn't have to repeat questions already asked of Sara. Prompted through the Next Best Action widget, Tony just continues from where Sara left off on a Pause Service request. That saves time, but Tony also has another powerful productivity tool in his arsenal. Augmented Agent Assistant is an agent productivity feature you can use to pass some of the data collection tasks back to Sara. While Sara is filling in the dates, she's requesting a pause TV service for Tony can focus on other work. There are additional chat requests at the moment, so Tony provides some feedback to his knowledge team, which, if actioned, is likely to reduce unnecessary escalations in the future.

Tony is then alerted by the assistant when Sara is finished updating a pause window. When he looks back at the conversation, Tony doesn't need to copy and paste Sara's answers from the chat thread into the work area. The assistant has just taken the information Sara provided and put it straight into the appropriate fields, even highlighting the fields that Sara has populated. That's a huge time saver for Tony, removing the error-prone cutting and pasting he had to do on the older agent desktop solution U+ have replaced with Pega. Released from low-value but high-frequency data collection tasks, agents like Tony could serve more customers on digital channels and do so extremely efficiently. Sara is all set, and now Tony can thank Sara for a chat. Maybe suggest the payment plan, as it looks like Sara is somewhat in arrears on her bills.

Let's look at another use of knowledge in answering customer questions through a chatbot or virtual assistant experience. Let's say, in our earlier demonstration, Sara wasn't looking for account assistance but simply wanted to take care of a technical issue and also to update her mailing address. Saying no to the proactive notification, Sara opens up the chatbot experience. In addition to delivering knowledge during the service task flows we saw earlier, you can also deliver knowledge right to the beginning, at the beginning of the chatbot interaction. Sara's simple technical question can be quickly addressed, allowing her to move on and to update her address on file. At any point, she can return to that knowledge article list.

That's just a very short summary of how chatbots, virtual assistants, and agent desktops come together in Pega Customer Service. We showed how we blurred the lines between simple chatbots and the more sophisticated and guided virtual assistant experience. And we showed that when agents are required, they have every tool at their disposal to understand the customer need and to quickly address it.

You have reached the end of this video.