The emergence of autonomous AI agents, which are capable of automating complex tasks, ‘thinking’ their way around unforeseen scenarios, being grounded in your company’s data with appropriate guardrails and increasing revenue, will bring profound change to both B2B and B2C businesses.
A new type of digital workforce will emerge, enabling firms to reimagine their business operations as technology and human workers combine to unlock more productivity, innovation, transformed customer experiences and growth.
To harness agentic AI’s full potential, organisations must carefully consider how best to strategically embed the technology within their operating models and integrate it with human talent.
To that end, we reached out to our CIO Experts Network – CEOs, CIOs, IT and business leaders, technologists and influencers – to ask this question: What key steps must businesses take to successfully leverage agentic AI technology? Here are their insights and recommendations.
Unlocking opportunities with agentic AI
Isaac Sacolick, President of StarCIO and former global CIO in the media, travel, construction, market research and financial services industries, was keen to point out the benefits of agentic AI, stressing how the technology has the potential to revolutionise complex, personalised decision-making.
He says: “Agentic AI will radically transform consumer experiences, much like mobile technologies expanded how consumers interacted and developed loyalty with leading brands.
“A key step is recognising where consumers have challenges making big, personalised decisions, such as researching expensive purchases, deciding on insurance options, managing wealth portfolios, or optimising their healthcare.”
Sacolick also highlights the limitations of current technology ‘vis-à-vis’ autonomous agents.
He goes on: “These personal decisions require factoring in consumer preferences, financial constraints and other contexts where faceted search capabilities are limiting and agentic AI provides a significantly more consultative experience.
“B2C companies must be early adopters in these AI capabilities, select experiences to develop learning experiments, and centralise their customer data to accelerate the development of agentic AI capabilities.”
Identifying high-impact use cases
Kieran Gilmurray, Chief AI Innovator, Technology Transformation Group, agrees that to maximise return on investment, early adopters must pinpoint those areas where agentic AI will generate the greatest value.
He says, “To effectively leverage agentic AI technology, businesses must identify crucial areas for impact, such as automating tasks and enhancing decision-making processes. Strengthening data infrastructure is non-negotiable to ensure high-quality, accessible data for AI systems.”
Laying a solid foundation: quality data and robust governance
The experts we interviewed reinforced Gilmurray’s point about data, agreeing that successful agentic AI implementation hinges primarily on high-quality data, adding that strong governance frameworks and business alignment were also imperative.
“To leverage agentic AI technology, businesses must first ensure high-quality, structured data and robust governance frameworks to maintain accuracy and compliance. Next, they should integrate AI models that can autonomously analyse, learn, and make decisions while aligning with business objectives,” says Tom Allen Founder and CEO of the AI Journal.
Furthermore, Allen advises that implementing human-AI collaboration mechanisms is crucial to oversee AI-driven actions and optimise outcomes.
He concludes, “It’s also a must for businesses to invest in scalable AI infrastructure as this can enable much easier AI agent deployments for different business units that exist in enterprises. Finally, continuous monitoring and iterative improvement through feedback loops will ensure AI agents remain effective, ethical, and aligned with the businesses blueprint and vision.”
How integrating data drives revenue
Peter Nichol, Data & Analytics Leader for North America at Nestlé Health Science, lays out the data challenges with the following hypothetical.
He explains, “AI agents are only as smart as your data. Imagine an agentic AI assistant which is designed to help manage your calendar. If you forget to provide your PTO schedule, it may be smart enough to not double book you, but you may end up working eight-hour days during your week-long vacation.”
Leveraging integration platforms that combine discrete and disparate data are the building blocks that ensure the intelligence of agentic agents, Nichol adds.
He continues, “Agents act on data intentionally, requiring integrated data. It doesn’t matter if you’re using a lakehouse approach, API-led connectivity, or automated data discovery capabilities. They all have a place. Intelligent agentic agents start with smarter data.”
The key point is that power of data comes when discrete and disparate data are combined, Nichol states.
Turning to growth, Nichol also offers insights into how agentic AI could help B2C businesses in fast-moving markets generate millions of dollars of additional revenue annually.
IoT sensors could be used to monitor product usage patterns, generating data that agents use to make recommendations that increase sales (such as creating ‘you’re running low, re-order now’ alerts).
Estimating a hypothetical revenue opportunity, Nichol outlines that – for a base of 15 million customers – if only 5% change their behaviour, an AI agent could generate an additional $2.5 million annually in top-level growth.
Ensuring scalability, adaptability and sound governance
Scalability enables seamless growth of agentic AI infrastructure while agility ensures capability can evolve at speed. Together, scalability and adaptability ensure agents stay smart and aligned with business needs, no matter how fast things change.
Javier Campos, CIO and AI Strategist at Fenestra, says a modular approach can help organisations achieve scalability and agility.
He says, “I advise organisations to build modular architectures that focus on fundamental components –profile, memory, planning, and action – rather than becoming overly dependent on specific platforms that rapidly evolve. This approach ensures scalability and adaptability as the technology landscape changes.”
By following these actionable insights and embracing a strategic approach to agentic AI adoption, businesses will be better placed to effectively leverage this transformative technology, driving innovation, enhancing efficiency, and achieving sustainable growth.
As Campos states, “I emphasise that successful implementation requires establishing clear operational boundaries, robust feedback mechanisms, and seamless human-AI collaboration protocols. Too many organisations rush into adoption without proper governance frameworks, which inevitably leads to disappointment or, worse, operational disruption.”
Conclusion: The road to agentic AI success
As agentic AI continues to mature, its potential to reshape industries, redefine customer experiences and increase revenue is clear, but realising this potential will depend on thoughtful, strategic implementation.
From establishing robust data foundations and governance structures to fostering human/AI collaboration and embedding flexibility at the architectural level, the message from our expert network is unanimous: success with agentic AI requires intent, investment and ongoing oversight.
Businesses that approach this shift with clarity and purpose will not only avoid the pitfalls of rushed adoption, but will position themselves as leaders in the coming era of intelligent automation.