Soapbox
AI to Bolster CSM Services

In the era of artificial intelligence and automation, it makes little sense to expend valuable human labor to complete repetitive manual tasks. For example, customer care services get bombarded with hundreds of calls every day, most of which would typically be trivial issues such as email troubleshooting and password reset requests. Investing quality labor and resources to accomplish such mundane tasks, to say explicitly, does not add-up to profitability. To counter such unreliable practices, Soapbox, an Australian-based customer service management company, combines natural language understanding and machine learning to create intelligent service automation systems. “Increasing the quality of services has always been in conflict with budget constraints, and the only way to resolve this is by adopting automation,” Dion Williams, CEO of Soapbox, says.

Soapbox tackles the challenges inherent in a human resources dependent service function like customer service management with Sofi, an intelligent assistant for existing service management platforms (like ServiceNow and Salesforce) by leveraging machine intelligence and cognitive search to assist service agents and customers to fulfil requests, provide answers and automate tasks.

Sofi has been engineered to master natural language understanding by leveraging machine-learning algorithms in the company’s computational model, which enables her to enhance efficiency with every iteration of information processing. Aggregating historical data obtained from similarities in consumer-client interactions, Sofi addresses repetitive queries instantly and designates complex tasks to authorized personnel.

Soapbox has designed its assistant to adapt and improvise according to different operational models in the deployed ecosystem, to cater to various CSM requirements.

Soapbox’s Intelligent Assistant addresses both the customer and service agent side of the customer service process: its Service Agent Assistant caters to incoming calls, emails, and other means of communication while its self-service (virtual) agent provides an automated means for customers to serve themselves. The Service Agent Assistant interacts with the agent as they capture calls from customers, providing recommendations designed to provide answers to customers and greatly reduce call handling time. One example of this is the automatic classification and assignment of calls based on the description of the problem, ensuring that calls get assigned to the correct team faster. Sofi also works with emails sent to the Service Desk, bypassing the agent interaction altogether.

The Sofi Virtual Agent offers employees and customers a conversational channel of engagement to automate organizational processes such as request and issue management. For example, by understanding an employee is requesting leave and extracting relevant information from the request, such as the number of leave days, and integrating with enterprise systems to understand how much leave is available, an employee can engage with Sofi to apply for a leave without going through multiple processes and systems while Sofi directs the leave application straight to the appropriate person for approval.

Sofi simplifies repetitive and routine customer service processes and streamlines operations to obtain maximum efficiency of labor. One such implementation done by Soapbox is for an Australian media group that received an average of 10,000 calls monthly, requiring an equivalent human resource to triage and resolve them. Utilizing the Sofi Intelligent Assistant, Soapbox was able to automate 30 percent of the communication interactions under various categories; Sofi assigned templates for each incoming query by classifying them according to the description of the query. Soapbox utilized Sofi’s machine learning algorithms and the customer’s historical data to acknowledge and respond to repetitive calls, while the newer and complicated queries were classified and redirected to the appropriate resolver groups for action. With a single phone call costing A$10 to 12 previously, the media giant ended up saving A$40,000 in terms of human resources within a month of Sofi’s implementation.

Soapbox is currently engaged in bringing Sofi to the mobile platform with support for Google Home and Alexa; to extend Sofi’s usability, the company aims at implementing voice functionality in the future iterations. Already named in the Top 50 Sydney Startups to Watch, Dion envisions the company to set benchmark standards for enterprise systems and take automation to a new level by leveraging artificial intelligence.