Dmitry walks us through the various steps involved in Conversational Process Automation and discusses the factors that makes a company the right fit for CPA.
With the AI-driven “fourth industrial revolution†now at full speed, businesses around the world from all sorts of industries are scrambling to adjust to new AI-driven norms and status quos.
Though practical innovations in the AI space have undeniably provided companies with the ability to evolve their businesses in ways previously unimaginable, it’s also brought about increased end-customer expectations for quick, easy and accessible customer service through channels like email, chat, social media and more.
This is a good thing, yet it’s also a challenge. It puts larger, more established companies at risk of being left behind, as support structures put into place even ten years ago are quickly becoming outdated and insufficient in providing satisfactory customer care.
That’s where the newest breakthrough in practical, effective, AI-powered support comes into play. It’s called Conversational Process Automation (CPA). CPA solutions allow a business (and its contact center) to be more agile, more adaptive to business-specific demands, and able to increase both customer and employee satisfaction rapidly.
CPA is both, easy to understand and implement, and the nature of conversational AI that learns from every customer interaction means that a business will quite naturally align with its customers’ needs and demands, simply by virtue of having deployed it.
Conversational Process Automation: Definition
Let’s take a look at how Conversational Process Automation works. Behind any customer service interaction lie two main factors: processes and conversations. What CPA provides, unlike any technology before it, is the seamless integration of the two, which eventually allows for large percentages of customer service queries to be processed without any human involvement.
As contact center traffic grows over time, a well-implemented CPA system learns how to provide end-to-end resolutions on more and more types of queries, meaning that a huge percentage of a contact center’s repetitive and expensive inbound tickets can be resolved quickly and efficiently, and in a manner that significantly increases customer satisfaction.
Breaking Down the Process
To better illustrate the process behind Conversational Process Automation technology, let’s take the example of a customer who is interested in finding out whether her flight ticket makes her eligible for access to the lounge area before boarding.
Before Conversational Process Automation, the interaction likely involved these steps:
1. A question was asked to the airline through one of their support channels — email, web form, chat, or social messaging
2. After some delay, a human agent saw the newly-created ticket in the CRM system, and then tagged it, while attempting to understand the customer’s intent and level of urgency
3. The same human agent then opened a different system to retrieve the necessary information
4. The agent then returned to the CRM, typed in the response, and sent it to the customer
5. The agent then manually updated the ticket, then closed the interaction in the CRM
Using Conversational Process Automation, the interaction that takes place might instead look something like this:
1. Initial Contact & Intent Identification:
“Hi, do I have lounge access? My booking reference is GA12345. Regards, Helen.†As can be seen, the customer has contacted customer service to ask about lounge access. The CPA system can identify this intent, and then accordingly start a correct process for this query.
2. Triggering & Executing the Correct Business Process:
The CPA system will seamlessly access a business’ backend systems (instead of requiring API calls or an agent to look into it) and retrieve the necessary information; in this case, the order information was triggered by the mention of order ID GA12345.
3. Recording & Resolving Query:
The CPA system will then record (if necessary) the change/request that has taken place, and resolve the query by sending the requested information back to the customer; in this case, it would be in the form of a personalized template email.
This process and its multi-system resolution is entirely seamless and invisible to the customer, who only knows that her query was resolved quickly and effectively. Meanwhile, no human agent in the contact center was involved in any way in the transaction. Instead, it was completed entirely by API-driven Conversational Process Automation from one end of the transaction to the other, then back again to rapidly inform the customer that she was indeed eligible.
Instead of deflecting questions, CPA focuses on getting a customer from problem to resolution as quickly as possible.
Another problem that adds costs to a contact center is that tickets often require information and actions from multiple systems to be resolved. This usually includes a minimum of a business’ CRM (such as Zendesk or Salesforce Service Cloud) and at least one back-end system to do things such as process refunds, enable cancellations or validate accounts. Agents, therefore, are forced to “swivel chair†from their ticket management system to another back-end system to resolve the case. This adds time and cost to resolving each ticket.
Because the CPA engine is powered by APIs, it can seamlessly connect the intent of the customer to the back-end systems required for case resolution, without agent involvement. What the contact center manager sees in their CRM and the CPA system’s dashboard are automatically resolved tickets with corresponding CSAT scores, which are typically much higher than they were before.
Which Companies are Right for CPA?
Three key characteristics make a company fit for Conversational Process Automation:
1. They have common, repetitive, and expensive customer service interactions that take up a large percentage of an agents’ time — interactions, such as refunds, cancellations, user validation, etc.
2. These interactions require customer service agents to take actions in different backend systems (Billing, ERP, Account Management, etc.)
3. These systems have existing APIs to connect with an AI platform or new APIs that can easily be developed
Early results have shown that CPA provides significant impact to a contact center’s average handling time, its customer satisfaction scores, and its first-contact resolution metrics. It’s technology that’s the very definition of practical AI in action, and something we’re expecting to see disrupting more status quos and delighting an increasing number of consumers in the months to come.