Companies in Southeast Asia (SEA) are at the forefront of a technological revolution, with artificial intelligence (AI) playing a pivotal role in reshaping industries. From streamlining operations in manufacturing to delivering personalised customer experiences in retail, AI has become an essential tool for businesses seeking to remain competitive in the fast-evolving digital landscape.
According to IDC, the AI market in the Asia-Pacific region is projected to grow from US$20 billion in 2023 to over US$110 billion by 2028. However, while AI promises vast potential, it also comes with risksโparticularly the danger of sinking significant resources into projects that fail to deliver the expected returns.
The path to successful AI adoption is not about throwing money at the latest technologies but about making strategic decisions that balance risk, cost, and benefit. Without proper planning and execution, AI investments can quickly turn into costly missteps. Here’s how SEA companies can maximise their AI returns while avoiding the common financial traps.
Start small with targeted use cases
One of the most common missteps for organisations adopting AI is attempting large-scale transformations without first building a solid foundation. The draw of AI’s transformative power can drive firms to embark on ambitious projects. Still, a more prudent approach is to start small and focus on targeted use cases that can deliver measurable outcomes.
“Many organisations believe in launching large AI projects to see instant results, but incremental progress often drives long-term success,” says Joe Ong, Vice President and General Manager, ASEAN, Hitachi Vantara. Rather than trying to overhaul entire operations, businesses should identify specific pain points where AI can address immediate needs and generate quick wins.
For example, local banks can begin by deploying AI for high-priority fraud detection rather than attempting a full-scale operational transformation. Companies can quickly measure outcomes by focusing on a manageable project, fine-tuning their models, and establishing organisational credibility. These early successes help secure stakeholder confidence and justify further investment in AI initiatives.
Moreover, targeting well-defined use cases helps firms avoid resource wastage. According to a SAS study, 40% of executives in the region reported setbacks due to the premature scaling of AI projects. Companies can learn from early implementations by taking a measured approach and gradually expanding their AI capabilities across departments and functions.
Invest in flexible and scalable infrastructure
A major challenge with AI adoption lies in the significant computing power and data storage requirements. Traditional on-premise infrastructure can be expensive and difficult to scale, often resulting in underutilised resources or bottlenecks during peak usage. SEA companies should consider flexible, scalable infrastructure options such as cloud-based platforms or hybrid cloud setups to overcome this.
Cloud solutions allow businesses to access computing resources on demand, enabling them to scale up or down as needed. This flexibility reduces upfront capital expenditures and ensures that resources are aligned with project demands. “Flexible infrastructure aligns AI capacity with real-time business demands, keeping costs in check,” says Ong.
In addition, sustainability considerations are becoming increasingly crucial for businesses in the region. AI requires substantial energy, and inefficient infrastructure can contribute to higher operational costs and a larger carbon footprint. Companies can adopt green data centres, energy-efficient hardware, and renewable energy sources to reduce environmental impact while maintaining performance.
For example, using energy-efficient servers and optimising data centre cooling systems can significantly save time. Firms can also integrate sustainability into their infrastructure strategy to balance technological growth with responsible resource management.
Manage stakeholder expectations with realistic timelines
Another critical aspect of maximising AI’s ROI is managing expectations, particularly regarding the time it takes to see meaningful returns. Stakeholders often expect immediate results, but AI implementation is a gradual process that requires iterative improvements and refinements.
Forrester’s 2024 Q2 AI Pulse Survey found that nearly half of AI decision-makers anticipated returns within one to three years. However, rushing projects to meet these expectations can lead to subpar implementations and missed opportunities for optimisation. SEA firms should adopt a phased approach, starting with small pilot projects demonstrating tangible value before scaling up.
Governments in SEA can learn from phased AI adoption strategies in the broader Asia-Pacific region. For instance, South Korea’s US$7 billion AI development plan prioritises initial research and talent development investments, followed by broader deployment across key industries. This methodical approach ensures that each phase builds on the previous one, creating a sustainable path to growth.
Companies can emulate this strategy by setting realistic milestones and tracking performance metrics at each stage. Early wins, such as improved customer engagement or enhanced supply chain efficiency, help maintain stakeholder enthusiasm while providing the time needed for more complex implementations.
Avoid the overload: Prioritise high-impact AI projects
The rapid advancement of AI technologies often tempts organisations to explore multiple initiatives simultaneously. However, this can lead to resource strain, project overlap, and unclear returns. SEA firms should prioritise a few high-impact projects with clear objectives and measurable outcomes to avoid this pitfall.
Rather than wasting resources on numerous experimental projects, companies should focus on initiatives that address core business challenges. For example, an AI-powered recommendation engine that improves customer retention can deliver faster and more measurable benefits than speculative projects with uncertain value.
Tailoring AI solutions to local markets can further enhance their effectiveness. In Indonesia, for instance, deploying AI chatbots that understand local dialects improves customer interactions and satisfaction. Customisation ensures that solutions are relevant and resonate with end users, resulting in higher adoption rates and better ROI.
Joe Ong highlights the importance of strategic prioritisation, stating, “Companies that focus on targeted, high-impact AI initiatives tend to see better results and avoid common pitfalls associated with overambitious planning.” Organisations can maximise their resources and achieve consistent outcomes by narrowing their focus.
Foster a data-driven culture with robust governance
AI success is intrinsically tied to data. Even the most advanced AI models can underperform without access to high-quality, well-managed data. To maximise ROI, companies must foster a data-driven culture where data is treated as a strategic asset and governance frameworks are in place to ensure its proper use.
Breaking down data silos is a critical step. Many organisations suffer from fragmented data systems that limit the effectiveness of AI initiatives. Companies can ensure that AI models are trained on comprehensive and diverse datasets by creating a unified data ecosystem, improving accuracy and performance.
However, data accessibility should be balanced with strong governance to address risks such as privacy breaches, AI model bias, and regulatory non-compliance. SEA firms should establish guidelines for ethical AI use, data protection, and performance monitoring. For example, financial institutions handling sensitive customer information must implement robust security measures to protect against breaches.
Collaboration across departments is also essential. Cross-functional teams comprising data scientists, business analysts, and domain experts can work together to identify opportunities, refine AI models, and troubleshoot issues. This collaborative approach fosters innovation while ensuring that AI projects align with business goals.
The road to sustainable AI success
Maximising AI’s ROI is a complex but achievable goal for SEA companies. With its AI market expected to grow rapidly at a compound annual growth rate (CAGR) of 24% from 2023 to 2028, the region is primed for AI expansion, driven by abundant data and high digital adoption rates. By starting with targeted use cases, investing in flexible infrastructure, managing stakeholder expectations, prioritising high-impact projects, and fostering a data-driven culture, businesses can navigate the challenges of AI adoption while avoiding costly mistakes.
As Joe Ong concludes, “AI success is not about speed but about strategic, phased adoption. Incremental progress ensures the delivery of consistent and measurable returns without overwhelming the organisation.” With a structured approach and a focus on long-term sustainability, SEA firms can harness this growth, unlock AI’s full potential, and create lasting value in a competitive market.
