Customer service is emerging as the front line in the battle for operational efficiency and enhanced customer engagement. Business leaders face complex challenges related to data quality, privacy, and the optimal integration of AI across their organizations. Frank Fillmann, ANZ General Manager of Salesforce and a leader at the forefront of this change, believes that the most transformative impacts of AI are likely to emerge on the front lines of business.

In the future, approximately 75% of the value of generative AI will come from customer-facing functions such as customer operations, marketing, and sales. "The biggest AI disruption will occur on the front line—where your customer and your employee engage with one another," says Fillmann. Despite this, many Australian C-suite executives remain focused on back-office functions like IT and operations, rather than investing in customer-facing applications.

While back-office investments are essential for boosting productivity and streamlining internal workflows, Fillmann underscores the potential downside of an overly narrow approach. "To achieve the greatest returns, companies must also focus on front-line applications," he explains.

One of the biggest challenges business leaders face is achieving the right balance between cost-cutting and enhancing customer experience. "Many companies are asking: how do we boost productivity without compromising customer experience?" says Fillmann. "Reducing costs, saving time, and giving customers the best possible experience all contribute to the bottom line."

Recent advances in AI have largely solved this challenge by enabling productivity and customer experience to go hand in hand. "With the introduction of AI agents, productivity and customer experience no longer have to be trade-offs," he says. Salesforce's Agentforce allows companies to deploy autonomous AI agents that support employees, effectively acting as an extension of the team. "AI agents help reduce the cognitive load on employees, allowing them to focus on the more nuanced and human aspects of their work."

However, deploying autonomous AI agents in customer-facing roles comes with potential risks and challenges. There are concerns about data reliability, accuracy, and maintaining customer trust. "If AI agents provide incorrect information or fail to address customer needs effectively, it can harm customer trust," warns Fillmann.

To mitigate these risks, companies should implement robust governance structures and ensure careful oversight of AI deployments. "It's critical to have humans in the loop to monitor AI outputs and intervene when necessary," he advises. Additionally, organizations must prioritize data quality and privacy. "Without high-quality, well-managed data, AI cannot deliver reliable results," Fillmann emphasizes.

AI adoption can be a daunting task, and executives are feeling pressure from both internal and external stakeholders to move quickly. "There's pressure from the board to hurry up and deliver value, and pressure from customers to bring the 'magic' of public AI models to their experience with companies," says Fillmann. Part of this pressure falls on technical leaders, who often delegate AI decisions upward, seeking guidance from the CEO. He warns against overcomplicating AI implementation by focusing on the technology itself rather than the business outcomes it can support.

"Companies are spending too much time deciding which user interface or model to choose, or even trying to build their own large language models," he says. "The fastest, safest, and most sustainable path forward is to abandon do-it-yourself efforts that start with the AI, and instead identify the most urgent, high-impact use cases that will deliver business value quickly."

For AI to succeed, Fillmann stresses that companies need robust, well-managed data infrastructures that address quality and privacy concerns. "Almost every organization tells us their data is fragmented, trapped in disconnected systems, and not ready to ensure AI investments are effective," he says. As a result, many organizations are spending considerably more on data infrastructure than on AI itself. "Our global study found CIOs are spending four times more on data infrastructure and management than they are on AI," Fillmann notes.

Establishing best practices for building a strong data foundation is crucial. This includes consolidating data sources, ensuring data is clean and properly labeled, and implementing strict data privacy measures. Salesforce's Data Cloud aims to solve this issue by creating a secure, unified data foundation. "Data Cloud has been designed to tackle these very issues. It's the foundation for every AI transformation and the core of Agentforce," Fillmann says.

For business leaders, the challenge lies not only in managing data quality, privacy, and training, but also in cultivating a holistic AI strategy that aligns with their strategic goals. Fisher & Paykel, a household appliance leader, is leveraging AI on the front line to enhance its technology-driven customer service and maintain a competitive edge. The company is exploring the integration of generative AI to streamline support processes and enhance user experience.

Rudi Khoury, Chief Digital Officer, sees customer service as an ideal entry point for generative AI applications due to its high volume of manual, repetitive tasks. "These AI tools can help boost efficiencies while enabling customers to resolve issues quickly and independently," he says. Fisher & Paykel is using AI bots to triage customer needs, directing them to appropriate resources and self-service options via tools like Salesforce's Einstein Bot.

This innovation has already yielded positive feedback, with around a third of users choosing to self-serve after interacting with the bot, reducing wait times and improving response rates. However, Khoury stresses that these tools don't replace the human touch entirely, particularly when it comes to quality control and delivering nuanced support.

Khoury acknowledges that while AI holds significant promise, it also presents challenges, particularly in terms of reliability and data accuracy. "If you're putting generative AI in a customer-facing role, it has to be flawless," he says. "Any AI-induced errors could harm customer trust, which is paramount in our approach to support. We don't want any savings to come at the expense of customer experience."

To ensure that their AI implementations are flawless and maintain customer trust, Fisher & Paykel is implementing governance structures and careful oversight. The company employs human oversight to review AI-generated information before it reaches customers. "This controlled deployment provides an additional layer of quality assurance," Khoury explains. Internally, the AI tools offer productivity boosts, enabling staff to complete tasks more efficiently and focus on higher-level problem-solving.

As for data, Khoury describes it as the "oil that fuels AI," stressing the importance of structured, labeled, and consolidated data systems. Fisher & Paykel has invested heavily in data platforms to streamline this infrastructure, understanding that cohesive data sources are fundamental to AI's effectiveness. "Building a robust, well-managed data infrastructure that addresses quality and privacy concerns is crucial for effective AI adoption," he says.

Rather than feeling pressured to deploy AI in every facet of the business, Khoury says the focus remains on aligning AI solutions with core business goals. For Fisher & Paykel, that means prioritizing tools that enhance, rather than replace, the human-centered service approach they are known for.

In conclusion, as C-suite leaders race to onboard generative AI, they must navigate the complexities of risk management, data quality, and strategic alignment. By focusing on high-impact use cases, investing in robust data infrastructures, and maintaining a balance between technology and human touch, companies can unlock the full potential of AI in customer service while maintaining customer trust and achieving business objectives.