Agentic artificial intelligence (AI) is an advanced yet practical technology that is starting to be applied in sales, service, and marketing organizations to increase automation and enhance the customer and employee experience (CX and EX), improve productivity and operational efficiency, and reduce operating costs. This form of AI has near-human cognition capabilities, which enable it to think and act autonomously, probabilistically, and proactively.
While this means companies do not have to provide it with detailed rules or instructions for it to know what to do, agentic AI does need access to a large language model (LLM) and possibly other data sources to find the patterns on which to base its decisions and actions. Agentic AI can establish its own goals, but it must have the right inputs, contextual framework, and ability to measure its effectiveness, particularly when getting started. Once those goals are established, agentic AI can determine what needs to be done to achieve them and execute the necessary actions. A significant component of agentic AI's value for organizations is its ability to learn from previous experiences and to adapt in real time to improve effectiveness.
Agentic AI has many functional and high-value uses in contact centers, including being leveraged alongside generative AI to enhance the performance, effectiveness, and outcomes of conversational AI systems. For customer-facing self-service, it can speed the end-to-end resolution process by completing the actions identified by the conversational AI solution. This improves self-service experience quality, enabling consumers to get issues resolved or interactions completed more easily while reducing the need for live employees to be involved.
Agentic AI is equally valuable when it comes to helping employees via agent assist/augmentation applications like real-time guidance or automated post-interaction summarization. It can also apply its intelligence and automation capabilities to address activities that would previously have required manual follow-up or intervention. But these are just the earliest and most obvious uses of this technology.
Agentic AI is being embedded into CRM and contact center solutions to enhance many aspects of these systems and to help create a more personalized and effective experience for all customer-facing interactions. It can be used to automate tasks and work cases that previously required input from or a hand-off to a human employee. It enables a CRM solution to use company data, knowledgebases, and LLMs to identify real-time insights to improve the relevance of its recommendations and immediately take the right actions to satisfy customers.
Agentic AI can proactively identify potential service and operational issues, notify the appropriate managers, and prevent a small issue from becoming a large one.
When it comes to sales, its data-driven analytics abilities can positively influence the outcome and success of activities, reducing operating costs while increasing revenue.
And marketing organizations can apply agentic AI to develop personalized content, altering and greatly improving the reach and impact of their campaigns.
Contact center infrastructure (omnichannel routing and queuing) and CRM solutions have used intelligent automation technology to improve interaction handling and output of CX operations for decades. The challenge is that each work case or workflow had to be individually built, and it only applied when an inquiry or interaction explicitly complied with its intent. Agentic AI changes this dynamic for CX organizations and many other departments since its application is limited only by the imagination of business managers and technologists and organizations' risk profiles.
An important factor driving agentic AI's benefits is its ability to break down and address tasks at a detailed level. When this is combined with its planning and decisioning-making skills, it's able to handle more complex and interrelated functions than other AI technologies. The potential for agentic AI to automate a growing percentage of activities that currently require the reasoning and soft-skills of live agents, customer service representatives, marketers, and salespeople is massive, but adoption depends on the willingness of businesses to deploy it and consumers to use it, no matter how good it is.
As is the case with all AI technologies and models, including genAI, it takes effort and oversight for agentic AI to deliver the desired benefits. It is not a case of create it and let it make decisions and operate on its own. As with any AI-enabled application, such as conversational AI solutions, which are using agentic AI, it's a best practice to continually keep a human in the loop. Agentic AI has amazing potential, but must be implemented with the necessary automated and manual controls and supervision to make sure it performs as expected and doesn't expose organizations to inappropriate risks.
Donna Fluss, president of DMG Consulting, is an expert on contact centers, analytics, and back-office technology. She has 30 years of experience helping organizations build contact centers and back-office operating environments and assisting vendors to deliver competitive solutions. She can be reached at Donna.Fluss@dmgconsult.com.