Directing customers to digital support channels has long been a goal, but Agentic AI offers a way to make the experience more personable and seamless. For a long time, customers have come to expect personalisation of services and support – to a level that reflects an understanding of who they are, their communication preferences, the products and services they’ve purchased in the past, and what they might be interested in in the future. This expectation exists regardless of whether service is handled by humans or via more digital channels.
As customer interactions have relied more on technology, an ever-increasing proportion are being contained to digital channels. Yet, the experience often does not match-up to the expectation. As PwC Australia notes, “historically, self-service and digital have had negative connotations”.
Organisations running support through predominately digital channels have typically not met customer expectations , although limitations in the underlying technology often meant that little could be done in terms of meaningful improvements. Traditional chatbots required extensive training data, and the amount of refinement required to achieve incremental improvements could be disproportionate to the time, effort, and cost involved.
The arrival of artificial intelligence – at first Generative AI and now Agentic AI – promises to change that situation completely.
In a customer service context, Generative AI is helping to connect a lot of customers to answers faster: parsing large amounts of corporate knowledge to synthesize a succinct response that can either be sent to the customer directly or passed on to the customer via a human agent.
Agentic AI takes things a few steps further by understanding and then executing appropriate and commensurate responses to customer support requests, without human intervention.
As more and more Australian organisations adopt Agentic AI, it will permanently shift the needle on what customers expect and accept as a baseline for digitally delivered support.
Gartner predicts that Agentic AI will be autonomously resolving four in every five common customer service issues completely autonomously by 2029. The firm considers a broad range of services to be in scope: “AI agents will not only provide information but will also take action, such as navigating websites to cancel memberships or negotiating optimal shipping rates on behalf of business customers,” it says. “Beyond these delegated tasks, agentic AI holds the potential for proactive issue identification and resolution.”
The role of knowledge graphs in achieving wins
Agentic AI alone will not deliver these gains. Rather, the combination of Agentic AI with knowledge graphs promises to help organisations really benefit from the technology’s capabilities.
As Agentic AI drives decisions from data, the insights underpinning these decisions must be accurate, transparent, and explainable – characteristics that graph databases are uniquely optimised to deliver.
Gartner already identifies knowledge graphs as an essential capability for GenAI applications, as GraphRAG (Retrieval Augmented Generation), where the retrieval path includes a knowledge graph, can vastly improve the accuracy of outputs.
Knowledge graphs are structured around ‘nodes’ and ‘edges’. Nodes represent existing entities in a graph (like a person or place), and edges represent the relationship between those entities – i.e., how they connect to one another.
In this structure, the bigger and more complex the data, the more previously hidden insights can be revealed. These characteristics are invaluable in presenting the data in a way that makes it easier for AI agents to complete tasks in a more reliable and useful way.
What users have been finding with GraphRAG is that not only are the answers more accurate, but they are also richer, more complete, and consequently more useful.
For example, an AI agent addressing customer service queries could suggest a particular discounted package on a broadband offer based upon a complete understanding of the customer, as a result of using GraphRAG to connect disparate information about said customer. How long has the customer been with the company? What services are they currently using? Have they filed complaints before?
To answer these questions, nodes can be created to represent each customer and aspects of their experience with the company (including previous interactions, service usage, and location), and edges to reveal the cheapest or best service for them.
Without that specificity, the broadband provider is limited to escalating these calls to human agents and extending more generic offers to customers that may not consider all of the context about them, and their individual situation, into account. This is where the combination of Agentic AI and knowledge graph database technology can come into its own, boosting customer experience outcomes.
Harnessing the power of data relationships is essential to sustainable competitive advantage in an Agentic AI-enabled world. By ensuring their data is as rich, connected, and contextually aware as possible, and fully accessible and usable by intelligent agents, organisations in service-oriented industries can unlock the true value of Agentic AI.