Search

ASAPP Launches AutoSummary Service API to Enable Automatic Call Center Functions

Automated Conversation Summaries and Analytics-Ready Data Input Can Help Agents Be More Productive

Call center automation

Beyond interacting and servicing customers, call center agents are typically required to handle the additional tasks of documenting each interaction they have, usually by typing up conversation summary notes, or inputting information via a drop-down menu interface. To provide the most value to the organization, in terms of better serving customers, identifying points of friction, and improving agent training, these summaries should include the reason for the call, the resolution steps that were taken, and the result or next action.

However, due to high volume of calls and the limited time to complete these call summaries before they need to answer a new customer inquiry, agent-completed contact summaries often are inconsistent in the terminology used, inaccurate, or simply too poorly written to quantify what actually occurred.

That’s why ASAPP, Inc. today launched AutoSummary, its new AI Service API that is designed to provide contact centers with automatic conversation summaries and analytics-ready data. Although the machine learning (ML) technology used to generate the summaries is not new, it is the first time ASAPP has offered the service via an application programming interface (API) that can be integrated within any existing technology stack and contact center/customer relationship management (CRM) application or platform.

While many organizations have tried to reduce the load on live agents by deploying more advanced AI bots, smarter interactive voice response (IVR) systems, and other deflection technologies, most companies have had to add agents to handle increased inquiry volume, according to Macario Namie, chief strategy officer with ASAPP.

“The labor that [organizations] spend on agents today equates to typically 80% or more of their budget,” Namie says. Instead of simply using technology to reduce costs, “we believe that we can make the same economic benefit occur by making agents twice as productive.”

Typically, agents are asked to type up summaries either during an interaction, or directly after it occurs, which can eat up anywhere from 1 to 2 minutes of time, per call. Worse, due to the inconsistency with agent-written summaries, Namie says that “very rarely have we seen organizations actually use the data that comes out of these summaries.”

ASAPP’s AutoSummary service allows agents to focus on serving customers, instead of writing summaries. ASAPP’s ML and natural language processing (NLP) models are used to ingest interaction transcripts, and create two different summaries, which are designed to meet the needs of agents, managers, and CX executives.

  • Readable Summary. AutoSummary automatically generates a detailed, paragraph-style summary from the conversation transcript without agents having to type a single word. The free-text summary, typically stored in a case management or CRM system, provides agents the level of detail they need to quickly get up to speed on previous interactions with a customer. 
  • Analytics-Ready Summary. For managers and executives, AutoSummary produces high quality, structured data about the content and outcome of every call, that is architected for Business Intelligence and Analytics systems. This structured data can provide rich insights into how to improve operations, create new revenue generation opportunities, and enhance products.

ASAPP works with customers in advance to gather historical transcripts of interactions, and develops company-specific ontology and taxonomy notes that will be used to analyze and classify calls. The ML model is trained on this data, ensuring that key insights and actions that can be used to improve CX are pulled out in each summary, so it can be used by a variety of end users. As more live transcripts are generated (either via the company’s AutoTranscribe transcription service, or via third-party services), the model will continue to improve.

“We don’t go to market with a generic, one-size-fits-all text-summarization model, because it frankly doesn’t work well,” Namie says. ASAPP typically asks for around 500 to 1,000 transcripts to get initial ML model training completed, though “as is the case with any machine learning, more data is better,” Namie says. He says that as part of the initial work with their customers, ASAPP will monitor and tweak the AI model so that each summary is an accurate reflection of what Namie terms the three “Rs”: the reason for the call, the resolution steps that were taken, and the result or next action to be taken. Says Namie: “That’s how we measure quality.”

In terms of results, Namie says that customers can expect a 9% to 10% reduction in average handle time (AHT), which results in more time for agents to interact with customers, as well as better summaries that can be used to positively impact CX.

Namie adds that while the concept of using NLP and ML to create text summaries of transcripts is not new, ASAPP’s service is unique in that it is specifically designed around the domain of the contact centers, and the broader CX mindset, and is designed to incorporate company-specific ontologies and taxonomies to create data outputs that are useful and actionable.

“The technology needs to be able to summarize what could be a 10, 15, or 20 minute call into three or five sentences that can be read by an agent, a manager, or a supervisor in 10 seconds,” Namie says. “Understanding that ontology or taxonomy for the structured data is vitally important, and needs to be unique and specific to that company, or it doesn’t have any value.”

Author Information

Keith has over 25 years of experience in research, marketing, and consulting-based fields.

He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.

In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek, CNBC.com, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.

He is a member of the Association of Independent Information Professionals (AIIP).

Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.

SHARE:

Latest Insights:

The Six Five team discusses Oracle Q4FY24 earnings.
The Six Five team discusses enterprise SaaS reset or pause
The Six Five team discusses Six Five Summit 2024 wrap.

Latest Research:

In our latest Research Brief, Fortifying Operational Technology (OT) Systems Against Cyberattacks–done in partnership with Honeywell– we examine the benefits of a comprehensive strategy for protecting OT assets against cyberattacks requiring asset discovery, ongoing risk assessment, and compliance management.
The Futurum Group’s Research Brief, Unlocking AI Potential: How HPE Private Cloud AI Accelerates AI Deployment and Innovation, completed in partnership with HPE and NVIDIA, delves into the complexities of AI deployment and the solutions offered by HPE's Private Cloud AI.
In our latest research brief, Intel AI Everywhere: Ready to Transform the AI Ecosystem, we analyze why Intel is perfectly suited to pace the AI Everywhere proposition. We look at why Intel is fundamentally committed to the core proposition of bringing AI everywhere, through key offerings such as Intel Xeon processors, Gaudi accelerators, and Intel Core Ultra Processors, which are aimed at ushering in the age of AI PC and securely distributing AI workloads in data center, cloud, and edge environments.