Driving revenue growth with Agentic AI: strategic opportunities in APAC's key verticals

In the second blog of this three-part series, Lian Jye Su explores how enterprises across APAC are moving beyond experimentation – using agentic AI not just for automation but as a strategic lever for revenue growth.

If you missed the first blog "From automation to autonomy: the APAC Agentic AI journey", click here to read it.

APAC's Agentic AI market opportunities

Agentic artificial intelligence (AI), namely AI systems that can plan, decide, and act autonomously, is emerging as a defining shift in enterprise automation strategy. While generative AI (GenAI) paved the way, the next phase of value creation will be led by agentic AI.

According to Omdia’s AI Software Revenue Forecast in Asia and Oceania, generative AI (GenAI) in APAC is projected to grow from $19 billion in 2025 to $71.8 billion by 2029, a compound annual growth rate of 39%. As enterprises mature in their GenAI adoption, the majority of growth will be driven by agentic AI.

As Ainkaran Krishnarajah, Exponent Venture Group notes: “Agentic AI isn’t a tech play — it’s a growth play. The real opportunity lies in turning automation into new revenue engines.” 

Where adoption is accelerating

The industries furthest along in agentic AI are those with established data and AI foundations, such as banking and financial services, retail, and healthcare. Standout examples include:

• Singapore's DBS Bank: using agentic AI systems to autonomously manage customer service escalations and identify cross-selling opportunities, resulting in a 23% increase in product conversion rates

• Indonesia's Bukalapak: implementing agentic AI to dynamically optimize marketplace recommendations and merchandising strategies based on real-time consumer behavior

• South Korea's Lotte Group: deploying autonomous pricing agents to adjust margins across 14,000 SKUs, while maintaining brand positioning 

• Japan's COEDO Brewery: partnering with NEC to uses Agentic AI for consumer analytics and product development based on local tastes and aromas.

“Enterprises won’t fail because they lack AI tools — they will fail because they treat GenAI as a side project. Agentic AI needs to live where revenue lives,” adds Krishnarajah.

Implementation framework: dual path to value creation

Enterprise deployments of agentic AI tend to follow two strategic directions:

• Operational excellence: automation, optimization, and predictive maintenance

• Revenue growth: Personalization, dynamic product engagement and intelligent customer service

To scale successfully, five foundational factors are key:

• Talent and capability: Hire AI architects and prompt engineers, while building internal muscle through upskilling and centres of excellence.

• Data foundations: Go beyond structured data. Agentic AI needs access to diverse sources—social media, live chats, reviews—and infrastructure to manage them responsibly.

• Legacy integration: Seamless connections to legacy systems remain a challenge. Open protocols like MCP may ease the path forward.

• Testing and tuning: Fine-tuning foundation models is essential, especially for open-ended, creative tasks like campaign design.

• Governance: Implement guardrails and output checks to manage hallucinations, risk, and accountability.

Investment considerations

Enterprises face a pivotal decision: whether to purchase pre-built agentic AI solutions or develop custom capabilities.

Buy Advantages
• Speed to market: proven solutions accelerate deployment
• Lower barrier to entry: less upfront cost and technical lift
• Stable support: predictable costs and vendor assistance

Build Advantages
• Custom fit: tailored to unique business needs
• IP ownership: long-term strategic control
• Deeper integration: closer alignment with proprietary systems

Most APAC enterprises will benefit from a hybrid approach — buying foundational tools while building differentiated layers to protect their competitive edge. Crucially, cost-benefit analysis for agentic AI in APAC markets must consider regional variables such as labour costs, technology infrastructure, and competitive dynamics. 

Enterprises must establish clear attribution and evaluation methodologies to isolate the revenue, cost, and risk impact of their agentic AI systems from other initiatives, providing executives with transparent ROI visibility.

Agentic AI is no longer a future play — it’s already reshaping how leading enterprises grow. But the winners won’t be the ones who experiment — they’ll be the ones who commercialize.

As Krishnarajah puts it: “The winners won’t be the ones with the most pilots — they’ll be the ones who ship revenue.”

Coming next, we focus on the telco sector—unpacking how communication providers in APAC are using agentic AI to unlock operational efficiency, reimagine customer engagement, and build entirely new models of business value.