Across Southeast Asia’s technology hubs, the AI conversation has moved quickly from “Can it help developers write code faster?” to a harder question: “Can it be trusted to work on the codebase by itself?”

That shift matters. The first wave of generative AI in software engineering was largely about assistance: autocomplete tools, chat-based coding helpers, boilerplate generation and faster documentation. The next wave is different. Autonomous AI agents can plan multi-step tasks, inspect repositories, refactor code, generate tests, review pull requests and attempt end-to-end software changes with less direct human prompting.

Also Read: Singaporean founders’ Lightsage bags US$4M to decode how AI agents choose software

For founders and CTOs, especially in fast-moving startup markets such as Singapore, Vietnam, Indonesia, Thailand, Malaysia and the Philippines, the appeal is obvious. Engineering talent is expensive, product cycles are short, and investors expect teams to do more with less. If AI agents can help small teams ship faster, the productivity upside is difficult to ignore.

But new data suggests Southeast Asia may be entering an uncomfortable phase: companies are deploying autonomous agents faster than their engineering systems are ready for them.

According to the Agoda AI Developer Report 2026, which surveyed more than 800 software developers and technology executives across Singapore, Thailand, Vietnam, Malaysia, Indonesia, the Philippines and India, 53 per cent of developers say AI agents are already in production or broad organisational use. Adoption is higher in several key markets: 58 per cent in Singapore, 64 per cent in Thailand and 59 per cent in Vietnam.

Yet only 38 per cent of developers believe their codebase is mostly or fully ready for full autonomy.

That gap between usage and readiness could become one of the defining engineering risks for startups in the region.

Productivity gains are real, but so are the risks

The report captures why founders are moving quickly. Across the region, 55 per cent of developers say AI tools now save them at least seven hours a week, up sharply from 18 per cent a year earlier. In Thailand, 70 per cent of developers report saving seven or more hours weekly. In Vietnam, the figure stands at 63 per cent.

Developers are also no longer restricting agents to simple tasks. The report shows they are testing or using agents across much of the software lifecycle: 71 per cent for documentation generation, 70 per cent for code refactoring, 66 per cent for autonomous code generation, 63 per cent for research and 62 per cent for pull-request reviews.

For a startup, that can translate into faster feature releases, shorter debugging cycles and more time for senior engineers to focus on architecture rather than repetitive implementation work. In a region where many companies operate across fragmented markets, multiple languages, different regulatory systems and uneven infrastructure, any tool that improves engineering velocity can feel strategically important.

The problem is that autonomous agents do not understand a company’s software the way experienced engineers do. Much of a startup’s codebase is shaped by context that is rarely written down: why a workaround exists, which payment flow breaks in a specific market, which customer segment depends on an old API, or which “temporary” patch has quietly become mission-critical.

A human engineer can often infer those trade-offs from experience, Slack history or conversations with colleagues. An AI agent, by contrast, depends on what it can read, retrieve and reason through. If the codebase is poorly documented, tightly coupled, weakly tested or filled with hidden dependencies, the agent may make confident but flawed assumptions.

Also Read: The hidden cost of AI coding: Why proof will matter more than prompts

That is where productivity can turn into technical debt.

Startups face the sharpest version of the problem

The readiness gap is particularly relevant for early-stage companies. The Agoda report found that among startups with one to 50 employees, 63 per cent report production or broad AI agent adoption, but only 37 per cent say their codebase is structurally prepared for autonomous execution.

The same pattern appears in larger growth-stage companies. Among organisations with 201 to 1,000 employees, 63 per cent report active production or broad agent use, while only 32 per cent believe their codebase is ready.

This is not surprising. Southeast Asian startups often build under pressure. Teams optimise for product-market fit, market launches, fundraising milestones and customer growth. Clean architecture, internal documentation, automated tests and platform governance can fall behind.

That trade-off is understandable in the early days. But autonomous AI changes the cost of messy systems. A codebase that is merely annoying for humans can become actively dangerous for agents. An undocumented dependency may lead to a broken workflow. A missing test may let a regression into production. A vague instruction may result in an agent modifying the wrong part of the system.

In other words, AI does not eliminate technical debt. It can expose it, accelerate it or multiply its consequences.

What “agent-ready” really means

The report points to five areas that companies need to assess before giving agents more autonomy.

The first is technical readiness. A codebase needs to be modular, decoupled and supported by reliable tests. If an agent cannot isolate the impact of a change, it is more likely to break adjacent services.

The second is integration readiness. Agents need access to the right development environments, internal packages, sandboxed databases, continuous integration pipelines and testing systems. Without these, they may generate code that looks plausible but cannot be safely validated.

The third is economic readiness. AI agents consume tokens, make repeated attempts and sometimes fail before producing useful output. Startups need visibility into the cost per successful task, not just the novelty of automation.

The fourth is governance readiness. Companies must define which tasks agents can complete independently, which require human approval and which are off limits. This is especially important in regulated sectors such as fintech, healthtech, insurtech and digital lending, where a faulty deployment can have legal or consumer-protection implications.

The fifth is workforce readiness. Engineers need to learn how to brief agents, review outputs, design better systems and validate changes. The role shifts from writing every line of code to directing, constraining and auditing automated execution.

Also Read: AI governance is moving from promises to proof

As Siba Prasad Hota, Senior Technical Lead and Architect at Sakhatech Information Systems, puts it in the report: “The goal should be AI autonomy with human oversight where consequences are significant, rather than replacing human ownership entirely.”

The Omise example

Regional payments company Omise offers a useful example of a more controlled approach. Rather than letting developers run unconstrained agents across production repositories, the company built a governed agentic platform for impact analysis, feature development, testing and code review.

During a major platform upgrade, Omise used AI agents to assess downstream code impacts, helping engineers complete in weeks what would previously have taken months. According to Sylvain Dormieu, Director of Engineering at Omise, the company achieved a 37 per cent productivity gain based on story points.

The more important lesson is not simply that AI improved output. It is that Omise treated agents as part of an engineering system, not as a shortcut around one. Junior developers could generate baseline code faster, while senior engineers focused on business logic, architectural decisions and risk. Accountability remained human-led.

That distinction is crucial for Southeast Asia, where startups are often scaling across markets with different payment rails, compliance requirements, logistics networks and consumer behaviours. Autonomy without context can be brittle. Autonomy inside a governed system can be powerful.

The founder’s takeaway

For startup leaders, the message is clear: high AI adoption does not automatically equal engineering leverage.

The companies that benefit most from autonomous agents will not necessarily be those that use the most tools. They will be the ones that make their systems legible to machines and accountable to humans.

That starts with refactoring for context: breaking down monoliths where practical, improving internal documentation, strengthening tests and making dependencies clearer. It also means building gateways rather than blanket bans, so agents operate inside controlled environments with automated checks. Finally, companies need to match autonomy to risk. Documentation, test scaffolding and research may be suitable for greater automation. Production deployments, security-sensitive changes and customer-facing systems still require strong human review.

Also Read: Your offshore vendor’s AI is running on your code: Do you know which one?

Southeast Asia’s startups have often won by moving faster than incumbents. In the age of AI agents, speed will still matter. But the advantage may shift towards teams that combine speed with discipline.

The next phase of software engineering will not be defined by whether companies use autonomous AI. Many already do. It will be defined by whether their codebases, workflows and leadership teams are ready for what autonomy actually demands.

The post The AI agent boom is exposing Southeast Asia’s startup codebase problem appeared first on e27.

In part one, I traced the early years of building Ngital, the frustration with an industry built on vanity metrics, the scramble to land the first clients, and the realisation that reshaped how the agency approached its work. Part two picks up from there: what trust actually looks like once that work is underway, the mistakes that don’t make it into founder highlight reels, and where the business goes next.

What trust actually looks like

The organisations on Ngital’s client list, United Nations, Education Ministry of Malaysia, Prime Bank Investment Limited, BRAC Aarong Enterprise, UCB Bank, Nordge Bank, Fantasy Kingdom, Concord, Rupayan, Marie Stopes Bangladesh, 10 Minute School, didn’t choose us because of a clever pitch.

They chose us because someone trusted the work, and the work held up, and the relationship survived at least one moment where something didn’t go according to plan.

That last part matters more than most agencies acknowledge publicly. The test of a real agency relationship isn’t a campaign that delivers everything it was supposed to. That’s the expected outcome, not the differentiating one. The test is what happens when a campaign underperforms, or a deliverable is delayed, or the brief changes halfway through execution, or the results are there but the client’s internal stakeholder is unhappy for reasons that have nothing to do with the work.

How you communicate in those moments, the speed, the honesty, the quality of the proposed solution, determines whether you keep a client for three months or three years.

One habit we built that made a genuine difference was what I call proactive transparency. Rather than waiting for a client to notice a problem, we developed the discipline of surfacing issues before they became complaints. This is uncomfortable at first. Nobody enjoys calling a client to say that something isn’t performing as expected. But the alternative, allowing the client to discover it themselves and then presenting a retroactive explanation, is significantly worse for the relationship and for the agency’s credibility.

There is also something to be said for learning to decline work. We have turned down clients whose expectations were misaligned with what honest digital marketing can deliver. We have walked away from briefs that were structured in a way that would make success impossible to achieve or define. Early in the company’s life, this felt like a risk. Later it felt like a filter that made the business stronger. The clients who stay with Ngital for years are not the clients who were easiest to onboard. They are the clients where the relationship was built on honest communication from the beginning.

The failures I won’t pretend didn’t happen

There is a particular kind of agency failure that doesn’t make it into case studies: the client you onboarded because the revenue was attractive, knowing somewhere in the back of your mind that the brief was unrealistic.

I’ve been in those situations. You tell yourself that you’ll manage it, that the team will figure it out, that the client will eventually adjust their expectations when they see what’s possible. Sometimes that happens. More often, it produces a six-month relationship that ends badly for both sides, with a significant amount of internal resource consumed and a reference that you can’t use.

Also Read: Why money won’t save Bangladesh’s startups: The ecosystem readiness crisis

We also made pricing mistakes, particularly in the early years. Underpricing is a trap that is easy to fall into when you need clients and difficult to escape once you’re in it. A client onboarded at a price point that doesn’t cover the cost of serving them properly will eventually become a client served poorly, because the economics don’t allow the time and resource the work actually requires. We had those clients. The right solution, when you recognise the problem, is to have the pricing conversation early and directly, not to hope that quality work will justify a rate increase that nobody has agreed to.

Operational complexity is another lesson that arrived gradually rather than at once. As the team grew, the coordination overhead grew faster. Communication that had worked informally when there were five people stopped working when there were thirty. Processes that had existed in my head needed to be documented in places where other people could actually follow them. The transition from a founder-run operation to a process-run operation is something nobody fully prepares you for, and I made every common mistake along the way, holding on to decisions I should have delegated, documenting processes too late, assuming that people understood expectations that had never been explicitly stated.

Cash flow is its own category. Service businesses are structurally vulnerable to the gap between work completed and payment received. There were periods where this gap created real pressure, where the revenue was technically there, in the form of outstanding invoices, but the operating account didn’t reflect it. Managing that gap requires more financial discipline than most founders building service businesses in the early stage naturally have.

Building in Bangladesh: The real picture

The case for building a serious digital company in Bangladesh is stronger than the conventional narrative suggests, and weaker in the specific ways that aren’t usually discussed.

The talent is real. Bangladesh produces engineers, designers, marketers, and strategists who are capable of work that competes at any level. The challenge is not the existence of talent, it’s the work of finding, developing, retaining, and properly compensating it in a market where the pricing pressure on services is significant and the competition for good people is increasingly intense.

The market is maturing faster than many established agencies are adapting. Consumer behaviour has changed dramatically. Social commerce is real. Performance marketing at scale is real. The appetite for digital advertising among serious Bangladeshi businesses, banks, real estate developers, healthcare organisations, educational institutions, FMCG brands, has grown substantially. The conversation I am having with clients today is fundamentally different from the one I was having five years ago. Clients are more sophisticated. They ask better questions. They push harder on performance data. That is entirely a positive development, even when it makes the work harder.

The pricing environment remains challenging. The Bangladeshi market still has a significant segment that purchases digital marketing primarily on the basis of cost. This creates pressure that is difficult to avoid entirely and important not to capitulate to entirely. An agency that wins on price is an agency that has accepted a margin structure that will make it difficult to invest properly in talent, technology, and process. That investment is the only sustainable path to quality.

International exposure has been important to Ngital’s development in ways I didn’t fully anticipate when we started. Working with clients like the United Nations, the Education Ministry of Malaysia, and Nordge Bank, and earning recognition from platforms like GoodFirms and TechBehemoths globally, has shaped how we think about quality benchmarks. When your reference point for good work is only the local market, you can find yourself setting standards that are locally competitive but globally mediocre. Exposure to international clients and international evaluation raises that floor.

Also Read: Bangladesh’s startup ecosystem is entering a new phase of investability

The opportunity beyond Bangladesh is something I think about more now than I did in the early years. Dhaka is not a geographic limitation. The question is whether the company you’ve built is capable of delivering at a level that earns business from clients who have global options. That is a quality question, not a location question.

What scale actually means

This is worth being honest about, because the conventional narrative around agency growth is misleading.

More clients does not mean a better agency. More employees does not mean a more capable organisation. More campaigns running does not mean more value delivered. Revenue growth that outpaces the growth of the team’s capacity to serve clients well is not growth, it’s degradation disguised as success.

The healthiest periods of Ngital’s development have been the ones where we grew client relationships before we grew client numbers. Where we deepened the quality of what we were doing for the clients we had before we added the next one. Where we hired for capability and then found the work to justify the hire, rather than winning the work and then scrambling to find someone to do it.

The distinction matters because clients can feel it. A client served by an agency that has slightly more capacity than it needs experiences something different from a client served by an agency that is perpetually running at 110 per cent. The former gets strategic thinking. The latter gets execution, if they’re lucky.

We have also learned, slowly, and not without some painful examples, that not every client is the right client. The right client is not simply the one who pays the retainer. The right client is one whose objectives are realistic, whose internal processes allow the agency to do its best work, whose stakeholders are aligned on what success looks like, and whose timeline for results reflects how digital marketing actually works rather than how they would prefer it to work.

That filter, applied consistently, is a growth strategy. It is also the thing that allows an agency to maintain a 5.0/5.0 rating across platforms when it has served more than 200 brands, because the work being delivered matches the expectations that were set.

Technology, AI, and the question every agency needs to answer

The conversation about artificial intelligence in digital marketing is happening at a level of abstraction that doesn’t always help practitioners.

The honest picture, from where I sit: AI is already changing what our team does and how it does it. Research processes that used to take hours can now be meaningfully accelerated. Content workflows, creative ideation, performance analysis, reporting, keyword research, audience analysis, all of these are being reshaped by tools that have become genuinely useful rather than theoretically interesting.

What this means operationally is that the tasks AI handles well are tasks our team no longer needs to spend as much time on. The time freed up needs to go somewhere. The question is whether it goes into lower billing or into higher-value work, into the strategic thinking, client understanding, and judgement-intensive activity that AI handles poorly.

My view is that the agencies that will be in serious trouble are the ones whose primary value proposition was volume. The agencies that existed to produce a high quantity of social posts, reports, or templated campaigns at a predictable price point, those businesses have a structural problem, because the cost basis for that type of work has collapsed.

The agencies that will be fine are the ones whose primary value is judgement. Strategic thinking about which channels to use and why. The ability to translate a business problem into a marketing solution that is testable, measurable, and genuinely connected to commercial outcomes. The capacity to build and maintain client relationships through the inevitable periods where things don’t go according to plan. The cultural knowledge of a market that isn’t legible to an algorithm.

Also Read: Bangladesh: An emerging investment sweet spot in South Asia

Ngital needs to be, and is working to become, the second type. That requires investing in people who can think strategically, not just execute tactically. It requires building internal processes that use AI to handle the commodity work while protecting the space for genuine analysis and creative judgement. It requires being honest with clients about what AI can and can’t do, including resisting the temptation to overstate its capabilities, which is a mistake I’ve seen other agencies make in ways that will eventually damage their credibility.

What I would do differently

The question I find most useful when reflecting on building Ngital is not “what went well”, it’s “what would I tell myself at the beginning that would actually have changed my decisions.”

A few things are clear:

• I would specialise earlier. The instinct to say yes to everything in the early stage is understandable but costly. An agency that does everything is an agency that is remarkable at nothing. The positioning advantages of genuine specialisation, in a sector, in a channel, in a type of business problem, compound over time in a way that generalism doesn’t.
• I would document processes from month three, not year three. The cost of not having documented processes is not visible when the team is small, because people compensate with effort and direct communication. It becomes very visible when the team is 20 people, and catastrophically visible when it’s 60.
• I would price for the work I was actually doing, not the work I hoped the client thought I was doing. Underpricing is a mistake that creates a client relationship based on a false premise. Eventually the economics force a correction, and that correction is almost always harder than the honest conversation would have been at the start.
• I would hire differently. Not necessarily differently in the sense of more or less experienced people, but differently in the sense of being more honest about what the role needed and more rigorous about whether the person genuinely fit it. The pressure to fill a seat quickly is real, but the cost of filling it with the wrong person is almost always higher than taking the time to find the right one.
• I would have built the author and thought-leadership side of the business earlier. Writing, through The 5.0 Agency, through the prompt engineering work, through the content we now produce at scale, has been one of the most effective ways to demonstrate expertise and build the kind of credibility that converts into serious client relationships. The knowledge was there from the beginning. The formal articulation of it came later than it needed to.

For founders who are starting now

The lessons I’d offer someone building a digital agency, or any service business, in Bangladesh today are specific rather than general.

Results are the only currency that survives scrutiny. Relationships open doors. Credentials create initial credibility. But in a service business, the thing that keeps clients is the quality of what you deliver against the expectations that were set. Everything else is table stakes.

Trust compounds slowly and disappears quickly. The organisations that have trusted Ngital with serious budgets and serious briefs didn’t make that decision based on a pitch. They made it based on the accumulated evidence of how we operate. That takes time to build and very little time to damage.

A service business becomes scalable when systems are stronger than heroics. The founder who solves every problem personally is building a business with a very clear ceiling. The founder who builds systems that allow the team to solve problems is building something that can actually grow.

Saying no is a revenue strategy in disguise. The clients you decline because they’re the wrong fit are protecting the time, energy, and reputation that will serve your better clients. This is easier to understand in theory than to practice when the pipeline is thin. Practice it anyway.

Hiring is one of the highest-leverage activities a founder can do, and one of the most underdisciplined. The quality of the team, compounded over time, is the quality of the company. Cheap hiring is almost always expensive in total.

The Bangladesh market is not a ceiling. It is a foundation. The question is whether you build on it or allow it to limit you. The companies that will matter from Bangladesh in 10 years are the ones that are building now with a global reference point for quality, not a local one.

Failure is not the opposite of success, it is its prerequisite. I have never learned as much from a campaign that worked as I have from a client relationship that didn’t. The work is in taking those lessons seriously rather than filing them away.

Also Read: Bangladesh after the ballot: Why the emerging market king may be entering its strongest growth decade

Where this goes next

Looking at Ngital today, the team, the client roster, the partnerships, the processes we’ve built, I feel something I can only describe as cautious satisfaction. Not confidence exactly, and certainly not complacency. Something closer to the awareness that the company is better than it was, but not yet what it needs to become.

The digital marketing industry in Bangladesh is in a genuinely interesting phase. The market is large enough now to sustain serious agencies. The client sophistication is high enough that serious agencies can differentiate from mediocre ones on merit rather than just on relationship. The technology landscape, including AI, is creating capability advantages for agencies willing to invest in learning how to use it properly.

There are things I want to build that don’t yet exist inside Ngital. Deeper proprietary research capability. A more structured approach to developing junior talent into senior practitioners. A product layer that doesn’t depend entirely on client retainers. Regional capability that makes the Dhaka headquarters a genuinely global base rather than a Bangladesh-facing one.

None of that is simple. Some of it will probably take longer than I currently think, require more iteration than I’m currently planning for, and produce lessons I can’t yet anticipate.

I’ve made enough mistakes at this point to be appropriately humble about predictions. What I’m less humble about is the direction. The conviction that digital marketing, done honestly and built around real business outcomes, creates genuine value for the organisations that trust it with their budgets, that hasn’t wavered. It’s gotten stronger.

The story of building Ngital isn’t finished. In many of the ways that matter most, it’s still early.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post What building Bangladesh’s agency actually looks like from the inside – Part 2 appeared first on e27.

A good product can open the door to a market, but it does not guarantee business success.

Many companies reach a point where their product or service has potential, customers are showing interest, and the team is working hard, but growth remains slower than expected. The challenge is often not the product itself. It is the business structure surrounding it.

Sustainable growth requires more than selling a product. It requires the right strategy, operations, market positioning, customer approach, financial planning, and people.

For companies looking to grow, the question is not simply “How can we sell more?” but “What needs to change inside and outside the business to support sustainable growth?”

Growth starts with understanding the business

Before looking for new customers or entering new markets, companies need a clear understanding of where they currently stand.

This includes understanding:

• Who the ideal customers are
• What makes the company different from competitors
• Which products or services generate the most value
• Where revenue is coming from
• Which costs are limiting growth
• Where operational problems are occurring
• What customers actually need
• Which markets offer realistic opportunities

Without this foundation, companies can spend significant time and resources pursuing opportunities that may not fit their business.

A clear business assessment can reveal problems that are not always visible from day-to-day operations.

The difference between being busy and growing

One of the challenges businesses face as they expand is confusing activity with progress.

A team may have more meetings, more leads, more projects, and more customers, while profitability or efficiency remains unchanged.

Growth should therefore be measured through meaningful business outcomes.

Depending on the company, these may include:

• Revenue growth
• Profitability
• Customer retention
• Market share
• Operational efficiency
• Sales conversion
• Customer acquisition cost
• Employee productivity

The right metrics depend on the company’s objectives. What matters is having a clear connection between daily activities and strategic goals.

Also Read: The AI productivity paradox: Why finance must  move beyond automation 

Strategy must become action

A business strategy is valuable only when it can be translated into execution.

Companies often have ambitious plans such as entering a new market, launching a new service, expanding their sales team, or improving their digital presence. The difficult part is turning those ideas into realistic steps.

A practical growth strategy should answer five basic questions:

• Where are we now?
• Where do we want to go?
• What is preventing us from getting there?
• What resources do we need?
• What should we do first?

Breaking a large objective into measurable actions makes execution more manageable and allows leadership teams to track progress.

Entering a new market requires more than demand

Market expansion can create significant opportunities, but entering a new country or customer segment without preparation can also create unnecessary risks.

Companies should consider factors such as customer behaviour, competitors, pricing, regulations, distribution channels, partnerships, cultural differences, and local business practices.

A market may look attractive from the outside but become much more challenging once a company evaluates the actual cost and complexity of operating there.

This is where structured market research and business planning can make a significant difference.

The objective is not simply to identify a market with potential. It is to determine whether the company has a realistic path to succeed in that market.

Operations become more important as companies grow

What works for a small company does not always work when the company becomes larger.

Processes that were previously managed informally may become inefficient. Communication can become slower. Responsibilities may become unclear. Customers may experience inconsistent service.

Companies therefore need to continuously review how work gets done.

Improving operations does not always mean implementing expensive technology. Sometimes the biggest improvements come from clearer responsibilities, better processes, stronger communication, and more effective performance measurement.

Technology can support these improvements, but it should serve the business strategy, not replace it.

Also Read: How to use AI to win (Hint: It has nothing to do with being more productive)

Customers should remain at the centre

Growth is ultimately connected to customers.

Companies can invest heavily in marketing, technology, or expansion, but if they do not understand their customers, those investments may not produce the expected results.

Customer feedback can reveal opportunities to improve products, services, pricing, communication, and the overall customer experience.

Businesses should continuously ask:

• Why do customers choose us?
• Why do some customers leave?
• What problem are we actually solving?
• What would make our customers choose us again?

These questions can provide valuable insights for both strategy and innovation.

Sometimes the business needs an outside perspective

Leadership teams are often deeply involved in daily operations. This can make it difficult to identify problems objectively.

I believe an outside perspective can help companies assess their current position, challenge existing assumptions, identify opportunities, and develop practical strategies for growth.

The role of a business advisor is not to make every decision for the company.

It is to help the leadership team make better-informed decisions.

Depending on the company’s needs, this can involve business strategy, market research, growth planning, operational improvement, partnerships, sales development, market expansion, and other areas of business development.

Also Read: The cheapest way to stop your AI product from regressing

There is no single formula for business growth

Every company has different challenges.

A startup may need help validating its market and building a scalable business model. An established company may need support entering a new market. Another business may have strong sales but inefficient operations. Others may need to reposition their brand, improve their customer strategy, or identify new growth opportunities.

For this reason, business advisory should not be based on a one-size-fits-all approach.

The focus should be on understanding each company’s specific situation, objectives, resources, and challenges before identifying the most suitable path forward.

Sustainable growth is built, not chased

The strongest companies are not necessarily those that grow the fastest.

They are the companies that build the foundations needed to continue growing.

A strong product matters. But so do strategy, customers, operations, people, financial discipline, market knowledge, and the ability to adapt.

For companies looking toward their next stage of growth, the goal should not simply be to do more.

It should be to build a business that can do better, smarter, and more sustainably.

That is where strategic business advice can create real value, not by replacing the company’s vision, but by helping turn that vision into a practical path forward.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post Why a good product isn’t enough for sustainable growth appeared first on e27.

In global logistics, the problem is rarely a shortage of new technology. Ports, shipping lines, freight forwarders and warehouse operators are constantly pitched tools promising to predict delays, automate yards, optimise routes or cut emissions. The harder question is whether those tools can work inside real trade networks, where a single container journey may involve terminals, trucks, ships, customs systems, warehouses and multiple data owners.

That is the gap PSA International and Granite Asia are now trying to address.

The Singapore-headquartered port operator and the Asian private capital platform have launched the G&P Strategic Innovation Fund, a US$50 million vehicle aimed at backing technology companies building for ports, logistics and global supply chains.

Also Read: Singapore’s Neptune Robotics secures US$52M to fuel global rollout of AI-powered vessel cleaning

The fund will co-invest with Granite Asia’s existing funds and focus on solutions that can move beyond pilots into practical, scalable deployment.

The timing is not accidental. Artificial intelligence, robotics and automation are beginning to change how goods are planned, handled and tracked. But supply chains remain deeply fragmented, especially across Southeast Asia, where modern mega ports sit alongside manual warehouses, paper-heavy customs processes and uneven digital infrastructure.

For PSA, which operates across 45 countries and handled more than 105 million TEUs in 2025, the fund is a way to bring external innovation closer to the operating floor. (A TEU, or twenty-foot equivalent unit, is the standard measure of container volume.) For Granite Asia, which manages and co-manages US$11 billion in assets, it offers a route into one of the world’s most complex industrial technology markets.

From pilots to ports

Corporate venture funds are not new in logistics. Many large operators have set up accelerators or innovation arms to test new software, sensors and robotics. The challenge is that pilots often stay pilots.

A warehouse robot that works in a controlled demonstration may struggle with irregular floor layouts. An AI model trained on one terminal’s data may not generalise well to another. A digital freight tool may be useful only if enough customers, carriers and intermediaries agree to share information.

The new PSA-Granite fund appears designed around that bottleneck. PSA brings physical infrastructure, operating knowledge and access to trade flows. Granite Asia brings technology investing experience and a portfolio-building lens. Together, they want to identify companies whose products can survive the messy reality of global logistics.

The fund sits within a broader strategic partnership between the two organisations, covering technology insights, venture engagement, talent development and innovation partnerships. It also ties into PSA’s wider AI strategy, which includes operational excellence, customer solutions, trusted data foundations, enterprise productivity and sustainability across its port ecosystems.

“Innovation, automation and artificial intelligence are increasingly becoming critical differentiators in global supply chains,” said Ong Kim Pong, Group CEO of PSA International. He said the partnership reflects PSA’s push to strengthen operational performance and connect its port ecosystems more effectively.

Also Read: Teleport powers Capital A’s rebound, but thin margins show logistics remains a hard road

The phrase matters. PSA has been framing its future around a “Node to Network” vision; the idea that ports are no longer just physical nodes where boxes are loaded and unloaded, but connected platforms in wider supply-chain networks. That shift requires technology not only at the quay crane, but across forecasting, inland transport, inventory visibility and emissions tracking.

Why Southeast Asia matters

For Southeast Asia, the fund lands at an important moment. The region is becoming a larger manufacturing and trade hub as companies diversify supply chains beyond China, while e-commerce, regional consumption and cross-border trade continue to grow. At the same time, logistics costs in many Southeast Asian markets remain high due to congestion, fragmented trucking networks, customs friction and inconsistent warehousing standards.

Singapore is already one of the world’s most advanced maritime hubs, but the region around it is far more uneven. This creates both a problem and an opportunity: technology that can prove itself in complex Southeast Asian supply chains may have relevance in other emerging markets too.

AI could help predict vessel arrival times, reduce idle time at terminals or optimise equipment allocation. Robotics could improve container handling, warehouse operations and inspection work. Automation could reduce paperwork and improve coordination between ports, shippers and inland transport providers. But these tools depend on reliable data, integration with legacy systems and trust among industry players.

That is where PSA’s operating footprint may give the fund an advantage. Startups in logistics often struggle to access real-world environments where they can validate products at scale. A global port operator can provide not just capital, but also use cases, domain expertise and potential deployment pathways.

Also Read: Sensors, predictions, premiums: How Willog turned shipment data into an insurance biz

Jixun Foo, Senior Managing Partner at Granite Asia, said the fund brings together Granite Asia’s technology investment platform with PSA’s operational scale and supply-chain know-how. He added that the aim is to support entrepreneurs whose solutions can deliver real-world impact for ports, logistics and global trade.

Strategic capital, not just financial capital

The fund also reflects a broader shift in venture and growth investing. In capital-intensive or operationally complex sectors such as logistics, climate and industrial automation, money alone is often not enough. Startups need customers, deployment sites, regulatory understanding and technical feedback.

That is why strategic investors have become more important in parts of the technology market where adoption cycles are long. A port automation startup, for example, cannot scale like a consumer app. It must pass safety checks, integrate with critical infrastructure and prove reliability under heavy operating conditions.

Granite Asia’s track record gives the partnership a financial platform to work from. The firm says it has invested in 121 companies valued at more than US$1 billion and supported 70 IPOs worldwide. Its mandate spans private equity, private credit and public capital, giving it flexibility across company stages.

Still, the success of the G&P Strategic Innovation Fund will depend less on the size of the cheque and more on whether it can help portfolio companies cross the gap between promising technology and repeatable commercial adoption. In logistics, that is usually where startups stumble.

For PSA, the move is also defensive. Ports are under pressure to become faster, greener and more resilient while dealing with volatile trade patterns, labour constraints, climate risks and geopolitical disruptions. Operators that can use data and automation well may gain an edge not only in efficiency, but also in customer stickiness.

Also Read: DBS doubles down on private markets with US$110M AI IPO fund

The US$50 million fund is modest compared with the scale of global logistics. But if deployed well, it could serve as a test bed for technologies that make supply chains less opaque and more adaptive. In a region where trade remains central to economic growth, that could matter far beyond the port gate.

The post PSA, Granite Asia launch US$50M fund to scale supply-chain innovation appeared first on e27.

This week, two numbers came out of the same region within days of each other, and almost nobody put them side by side.

On September 3, Singapore’s central bank committed SG$220 million over three years to fintech and AI infrastructure under a fourth-generation innovation scheme. Around the same time, a Tracxn landscape report showed Southeast Asia’s native AI startups had raised US$4.1 billion through July, more than double all of 2025. On paper, that reads like a boom.

Except one company, Kling AI, absorbed 68 per cent of that total in a single Series D. Strip that round out and the region raised roughly US$1.3 billion, spread across just 23 disclosed deals, down from 41 the year before. And separately, no dedicated Southeast Asia private equity fund closed at all in the first half of 2026 — the worst showing on record — while three pan-Asian mega-funds raised a combined US$39.2 billion that has no obligation to land in this region specifically.

The tool gap closed, the timing gap didn’t

Here is the part of the AI story that gets the applause: building software no longer requires a computer science degree. By one recent estimate, 84 per cent of people using AI coding tools in 2026 have no engineering background. Lovable reportedly hit US$400 million in annualised revenue this year, four times what it was eight months earlier. A non-technical founder took a Lovable-built tool to over US$800,000 in ARR in nine months.

I am one of these people. I scored 147 on the PSLE and was streamed into Normal Technical — not the track anyone expected a founder to come from. I built my first product during downtime on ambulance shifts, using AI as my technical co-architect. So I understand, from the inside, why “anyone can build now” feels like the headline.

Also Read: Singapore’s AI dividend will depend on what happens after the pilot phase

But building access was never the real gate. Information timing is, and that gate has not moved.

Institutional investors have spent years buying lead time most consumers don’t know exists. Search data has been shown to run 4 to 10 weeks ahead of reported revenue for consumer-facing businesses; card transaction data offers a 2 to 4 week head start on earnings surprises. None of that is illegal or even secret — it’s a mature industry with vendors and case studies. It’s just priced for institutions, not for the person deciding whether to buy the stock, the shift, or the property everyone will be talking about in a month.

Consumers can feel it, even if they can’t name it

Consumer AI usage keeps climbing — Prophet’s 2026 research puts adoption at 73 per cent, up from 45 per cent in early 2024. But belief that AI will be trustworthy enough to lean on for real decisions has dropped roughly 30 per cent in the same window. That is not a contradiction; it’s a fairly accurate read of what’s happening. The same report notes that “businesses that own the agents will have a structural advantage in maintaining consumer relationships, driving full-journey engagement, and capturing data” — which is a polite way of saying the agent works for whoever deployed it, not necessarily for the person typing into it.

Agentic commerce is walking straight into that gap. Consumer surveys this year show 65 per cent of US shoppers trust AI to compare prices, but only 14 per cent trust it to actually place an order on their behalf. OpenAI paused its own Instant Checkout feature in March, shifting focus back to discovery rather than transactions — a fairly candid admission that the trust and evidence layer underneath these systems isn’t ready, even as the commerce layer races ahead.

Speed without an evidence trail is its own risk

It would be dishonest to hold vibe coding up as the clean counterexample to institutional advantage, because the same “move fast, verify later” instinct shows up there too, with its own cost. Security researchers who scanned over 1,400 production vibe-coded applications found 65 per cent had security issues and 58 per cent carried at least one critical vulnerability. Georgia Tech researchers tied 35 CVEs in a single month directly to AI-generated code. Building got easier. Building something that holds up under scrutiny did not.

Also Read: The hidden economics of autonomous AI agents

I think about this the way I was trained to think about a scene before I touch a patient: check the evidence, understand what’s actually in front of you, then act — not the other way round. That habit, more than any framework, is what shaped the signal-intelligence work I do now at OnTheRice, where the working rule is that a claim only counts once it’s timestamped, sourced, and checked against what actually happened later. I don’t say that to sell the product. I say it because the discipline is the point, and it’s transferable to anyone building in this window, not just to me.

What Southeast Asia’s builders should actually be arguing for

The uncomfortable version of this moment is that AI has quietly widened, not narrowed, the distance between who acts on information first and who reads about it after the fact. Capital in this region is barbelling into a handful of infrastructure mega-rounds while dedicated regional funds can’t close.

Consumer-facing AI is scaling in usage while shrinking in trust. And a genuine wave of non-traditional builders — paramedics, ex-guild leaders, people who never touched a computer science classroom — is proving the tools no longer gatekeep who can ship. That part is real, and it matters.

But shipping fast and shipping trustworthy are different achievements, and only one of them closes the actual gap. The founders in this region who deserve the next round of attention and capital are not the ones building faster checkout flows or flashier agents.

They’re the ones building the boring, auditable infrastructure that gives an ordinary person in Singapore, Jakarta, or Manila the same kind of advance notice a hedge fund buys for itself — evidence attached, timestamped, and checked against reality afterward. Everyone can build with AI now. The next competitive line is who’s willing to build something worth trusting first.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post Everyone can build with AI now. Almost nobody can see what’s coming appeared first on e27.