Startup ventures in Asia have been growing over the past few years; however, many young companies think marketing is optional. However, marketing is not a choice, but a must, according to analysis, and without a proper strategy and budget, no amount of effort will make the campaign successful.

As an entrepreneur says, “it takes years to make an impression,” while a partial campaign “simply does not work.” Surveys prove the point: companies that spend less than five per cent of their budget on marketing only grow in half of the cases, while 80 per cent of those that invest 5-10 per cent of their budget grow considerably.

Localise your strategy for diverse markets

There is great diversity among Asian markets, and blanket marketing techniques rarely work. Simply mimicking Silicon Valley approaches is not always helpful; for instance, the approach that works in Singapore will not necessarily work well in rural Southeast Asia. In successful Asian startups, unique playbooks are crafted because the company takes time to learn local culture and behaviours rather than mimicking global trends. Grab and Gojek gained traction by taking advantage of local trends such as cash payments and motorbike taxis as opposed to Uber.

The other key thing is trust, since most consumers in many Asian markets are skeptical and need to be won over before they fully embrace the product or service. In addition to creating demand through the use of attractive content, startups must be able to earn customers’ trust. The use of content such as guides and customer success stories helps to build credibility. PropertyGuru became a trusted name in property listing by offering simple guides in local languages.

Build credibility with content and PR

Mistake: Assuming all marketing is paid advertising. In Asia, earned media and content authority are huge assets. Neglecting public relations and expert commentary is a missed opportunity. Detutu Media notes that founders who shift from shouting into the void to borrowing existing audience through expert quotes and media coverage can land their first press mentions in 60–90 days. In other words, rather than creating content for no followers, startups should contribute expert insights to established publications. Those quick media wins instantly transfer the outlet’s credibility to the startup.

Also Read: The future of marketing isn’t about AI, it’s about judgment

Diversify channels and content

Overreliance on one platform is another mistake entrepreneurs make. 72 per cent of Asian companies say social media is their most used digital channel. Startups spend money on Facebook, TikTok, and Instagram ads. It’s a big mistake because the entrepreneur relies solely on paid advertisements while ignoring other platforms.

Overlooking content marketing and SEO may become even costlier. One marketing company suggests that “over-reliance on paid ads without content authority” is not an efficient strategy. The meaning behind that statement is clear: you spend money on ads but haven’t created any blog posts, videos, or emails for SEO and organic marketing.

It’s a huge mistake to skip influencer and community marketing, too. Consumers in Asia are loyal to the recommendations of influencers. According to Nielsen, approximately 80 per cent of Asian social users are more willing to purchase products recommended by influencers.

However, startups do not make use of that opportunity since fewer than half of the companies in Asia consider influencer marketing crucial. Entrepreneurs should use local influencers who will engage their audience and connect better with them. The same is true about email marketing. In conclusion, diversify your marketing efforts. Use paid ads, owned content, and earned marketing on social platforms.

Measure everything and stay agile

Firstly, do not fly blind. Analytics should not be skipped or neglected. Research proves that in a certain number of Asian companies (for instance, six per cent in Singapore), there is no monitoring of any kind of marketing metrics whatsoever. This is a path to failure. A startup has to measure performance and act according to its results. As a famous marketer says, acting on instincts rather than numbers is often seen but not advisable. 

Also, be prepared to pivot. If something does not work in Asia, change it. According to one marketing manual, a company should stay “agile” while working on its strategy. For example, if a copy of a landing page is confusing to the audience (ShopBack found out that it was true about the word “cashback” in SEA), then change the copy and make it beneficial to users.

Also Read: Driving the future: How AI is rewriting the global marketing roadmap for the Land Rover Defender

Invest wisely (but invest enough)

Finally, don’t starve marketing of resources. Underfunding and expecting magic are traps. As one founder remarks, too many startups launch once and “disappear” because they lack a follow-through plan. A sustained budget and timeline are essential. Hiring an all-star in-house marketing team may be out of reach, but shortcuts like always choosing the cheapest agency or tool often fail. A Singapore marketing guide cautions against “choosing agencies based only on price”. Instead, pick partners for fit and expertise.

Beware shiny-object syndrome (e.g., unvetted AI content) as well. Using generic AI text without human review can hurt authenticity. Always ensure creative quality and local relevance, even on a budget. Finally, embrace the mobile reality: Asia is largely mobile-first.

Currently in APAC, there are over 1.4 billion people (~51 per cent penetration) who use mobile internet. If your website is not optimised for mobile or neglects applications and messaging services that are popular among your local audience, then you’re missing out on customers. The digital presence of a startup needs to be fast-loading and native to how Asians consume media.

Marketing in Asia is much more complicated than in Western countries. Avoiding all these mistakes due to poor localisation, brand-building, and/or data, the underinvested startups can actually transform their marketing into a competitive advantage. Marketing done correctly can be the difference between earning your startup’s first paycheck and celebrating its tenth anniversary.

Conclusion

Learning from failures is crucial. The best thing for any startup from Asia would be planning, adapting to cultural contexts, using paid as well as earned media platforms, and constantly measuring the results. In an area where everything depends on context and trust, the most intelligent startups steer clear of making mistakes and use data-driven advertising campaigns.

—

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 Diversify or die: The importance of multi-channel marketing in Asia appeared first on e27.

Indonesia is one of the world’s fastest-growing digital economies: young, mobile-first, and home to more than 270 million people. It is also one of the region’s most dynamic and layered markets, full of opportunity for companies that take the time to understand how it really works.

That opportunity was the backdrop for Scale Up to Global 2026: Gateway to ASEAN, a closed-door session hosted by e27 in partnership with Thailand’s National Innovation Agency (NIA) at the Mandarin Oriental Jakarta on 23 July. The premise was straightforward: as Southeast Asia’s largest economy, Indonesia is a natural gateway for regional expansion, and the event was designed to help a group of Thai scale-ups explore that market by meeting the local partners, enterprises and investors who could help them land.

Rather than a run of stage pitches, the afternoon was built around conversation, bringing together Indonesian enterprise leaders, founders, investors and government representatives for a panel discussion, followed by roundtables with each visiting company and open networking. The emphasis throughout was practical: partnerships, pilots and distribution.

The five Thai scale-ups

The showcase featured five Thai companies, each already deployment-ready at home and each addressing a sector that matters to Indonesia:

• MUI-Robotics, an NIA-backed deeptech company whose AI-Nose platform digitises smell and taste for quality control, safety and environmental monitoring in industries such as food and beverage, cosmetics and chemicals.
• ViaBus, a transit-technology company that digitises public transport end to end, from real-time passenger information to driver tools and fleet-management systems, already operating across Malaysia, Thailand, Laos and the Philippines.
• Precision Dietz (Dietz.asia), a telemedicine platform for chronic-disease and non-communicable-disease care that connects hospitals, clinics and home-based monitoring into a more continuous model.
• ALIVELOOP, a circular-materials platform that turns hard-to-recycle packaging waste, such as multi-layer foil, into industrial-grade material, with the goal of building circular supply chains across the region.
• PraIn FinTech (ChillPay), a Bank of Thailand-regulated payment-gateway provider expanding from domestic payments into cross-border commerce, connecting Thai merchants with customers across the region.

Their sectors map neatly onto Indonesia’s own priorities: AI-driven quality control, digital health, sustainable packaging and cross-border payments.

Inside the panel discussion

The panel, moderated by e27 Co-Founder and CEO Mohan Belani, paired an incoming operator with three players who know the local market well. On the incoming side was Intouch Marsvongpragorn, Co-Founder and CEO of ViaBus. Representing the local view were Abhishek Pansari, COO of distribution company Baskit; Agustine Gunawan, SVP of partnerships at digital-health platform Alodokter; and Bayu Seto, a partner at Living Lab Ventures, the corporate venture arm of Sinarmas Land. Over an hour, the conversation moved from the big-picture opportunity to the practical steps that make a cross-border expansion work.

Also Read: Southeast Asia’s AI future is being written in Vietnamese, Thai, Indonesian

One country, many markets

A recurring theme was that Indonesia is best understood not as one market, but as many. ViaBus, which operates across Malaysia, Thailand, Laos and the Philippines, drew a contrast with its home market: Thailand is highly centralised, with most activity in Bangkok, while Indonesia is an archipelago of islands, each with distinct cultures, regulators and local players.

The company’s most useful takeaway was counter-intuitive: you may not need to start in Jakarta at all. Pansari agreed from the distribution side. Indonesia’s 270 million people are far from homogeneous; tastes, spending power and online behaviour vary widely, and trends move fast, so a single go-to-market model rarely travels well.

Regulation comes first

For companies in regulated sectors, compliance is the first thing to plan for, and it usually takes longer than newcomers expect. Gunawan spent close to a year securing a place in the Ministry of Health’s regulatory sandbox, an investment that now sets Alodokter apart.

Pansari pointed the same way from consumer goods: Halal and BPOM certification can take five to seven months, with more steps than in some neighbouring markets. The lesson for founders is to build regulatory timelines in from the start, because treating certification as an afterthought is one of the most common reasons a launch slows down.

The value of a local partner

If there was one point of consensus, it was the value of a strong local partner, for reasons beyond market knowledge. Pansari described Southeast Asia as a relationship-driven, trust-based environment: even the best product benefits from someone who can open the right doors, both to the market and to the right conversations.

Some consumer brands, he noted, came to Baskit only after trying to go it alone. Marsvongpragorn agreed while staying flexible on structure: a joint venture, vendor relationship or channel partnership can all work, but a local partner, sometimes more than one, is what activates each region.

Localisation as strategy

Gunawan offered one of the sharpest framings: in Indonesia, localisation is not cosmetic adaptation but core business strategy, and pricing is decisive. He pointed to a premium US hospital-information system that, within a month of launching, was matched by cheaper local alternatives that slotted into existing systems.

Indonesian teams are quick and resourceful, so a product priced above what the market will bear can be undercut fast, and because higher costs are passed on to the customer, pricing strongly shapes adoption. Pansari’s beauty-sector example echoed it: a brand that had thrived in China, Taiwan and Thailand found Indonesia far more price-sensitive, a market where a fresh product line may be needed every three to six months to stay ahead.

Also Read: Why agritech is key to securing long-term food resilience in Indonesia

A city as a sandbox

Seto used his time to introduce a different route into the market. Living Lab Ventures is wholly owned by Sinarmas Land and has grown from managing its parent’s balance sheet into a cross-border fund manager with external investors, backing Asia-Pacific companies that are ready to expand into Southeast Asia.

Its distinctive asset is Sinarmas Land’s BSD City: a fully private, self-operated city of around 6,000 hectares (larger than Pattaya, and roughly a tenth the size of Singapore), home to some 500,000 residents and, by Seto’s account, the second-highest GDP per capita in the country after Jakarta.

Because the group manages the city’s roads, transport, water, fibre and hundreds of CCTV cameras feeding a single command centre, BSD can serve as a live sandbox where startups run controlled pilots, backed by investment and a go-to-market programme, before committing to a national roll-out. AI runs through much of it, from traffic management upward, though Seto was candid that not every experiment works: an autonomous-bus pilot paused when regulation was not yet ready.

That kind of controlled test, he suggested, is a signal in itself, showing whether Indonesia is ready for a product, or the product ready for Indonesia. Living Lab, he added, is now actively looking for AI investments. His advice: validate the model in a captive, high-spending-power environment first, then scale with confidence.

Playbooks that have worked

Seto grounded the pitch in examples. In 2023, Living Lab invested in a Melbourne-based loyalty app for its Southeast Asian expansion and connected it into the Sinarmas ecosystem, with venues such as Plaza Indonesia and Aeon adopting it.

In 2025, it partnered with a century-old Japanese technology company to bring around 20 intellectual-property assets, spanning semiconductors, logistics and healthcare, into Indonesia through an accelerator that matched them with local startups; the cohort produced four joint ventures.

The logic is a repeatable, win-win trade: the foreign partner gains market access, while the local startup gains new innovation and handles the localisation. A product does not have to be the best in the world, Seto said. It has to be adapted into the right business model for Indonesia.

How foreign companies can stand out

Companies have been coming to Indonesia to expand and partner for years, Belani noted, prompting the panel to consider where the biggest opportunities now lie. The answers pointed less to the product than to the approach: co-development, genuine IP collaboration, and true partnership rather than going it alone.

Marsvongpragorn closed with a fitting metaphor: in a fragmented market, where Bangkok alone has around 200 bus operators, the goal is not to fight for slices of a small cake, but to work with others to bake a bigger one. For founders weighing a move into Indonesia, the message was encouraging and clear: come with an open mind, the right local partner, realistic timelines, and a willingness to adapt. For those who do, the opportunity is substantial.

The post Localise, partner, adapt: A playbook for entering Indonesia appeared first on e27.

Three weeks ago, I had coffee with the chief risk officer of an Indonesian insurer who has spent the last eighteen months building an AI-driven pricing model for one of their health products. He told me the model now produces better predictions than any of their senior actuaries can defend. He also told me he was not entirely sure what to do about that.

That conversation — repeated across enough insurance C-suites — is the part of the industry’s AI story the actuarial profession is still processing in private.

I spent three years as a financial risk country manager at one of Indonesia’s largest life insurers, and have sat through enough pricing committees and product launch reviews to recognise what is happening. The discipline that holds insurance together — actuarial science — is older, more rigorous, and more sober than the rest of finance gives it credit for. It is also, more than any other part of financial services, built on a single foundational assumption: that risks can be pooled across populations and priced once.

AI is breaking the pool.

What insurance was built on

For three centuries, insurance has operated on a deceptively simple model. Gather a large population of similar risks. Estimate the expected loss across that pool. Charge each participant a premium covering their share of the expected loss plus an insurer margin. The healthier subsidise the sicker. The lower-risk subsidise the higher-risk. The pool, at scale, is more stable than any individual within it.

This is not just a pricing technique. It is the social contract of insurance. Risk pooling is what makes insurance economically valuable for the individuals who buy it.

How AI breaks the pool

Three mechanisms are reshaping the foundation at the same time.

• Granularity. AI models can now distinguish risk at the individual level with precision that classical actuarial models cannot match. Driving telematics, wearable data, health records, and social media activity — all feed into models that price each policyholder as a pool of one. The risk pool, conceptually, dissolves.
• Continuity. The premium that used to be set once a year, based on an underwriting moment, can now be repriced continuously against real-time data. The “policy period” becomes an accounting convenience rather than a risk reality. Adverse changes in behaviour are caught — and priced — within days.
• Selection. With granularity and continuity comes the most consequential change. The healthiest, lowest-risk segments of the market discover they are overpaying inside the traditional pool and migrate to insurers that price them individually. The pool that remains is, by definition, riskier than the original. Premiums rise. The cycle repeats. Adverse selection — which insurance regulation has spent decades preventing — becomes structural.

Also Read: Investors bet on algorithms and insurance to tame Asia’s climate-health crisis

Why ASEAN is more exposed than the markets writing about this

Most of the public conversation about AI and insurance is happening in the United States and Europe, where regulatory frameworks for AI pricing have begun to take shape. ASEAN is running behind those discussions, not for lack of capability but because the region is running behind in everything to do with AI risk.

The exposure here is sharper for three reasons.

The data is more available. Indonesia, the Philippines, Vietnam, and Thailand have generated enormous volumes of digital behavioural data through super-apps, mobile banking, ride-hailing, and digital health. That data is less regulated and more granular than what insurers in mature markets can legally use. The capability to price individually is further ahead than the regulatory ceiling that constrains it.

The actuarial profession is thinner. The major actuarial associations in Indonesia, Malaysia, and the Philippines have credentialled practitioners in the low thousands per market, not the tens of thousands. A shock to the profession from AI displacement lands harder when the bench is smaller.

The supervisor’s capacity is uneven. Insurance regulators have been building AI competence steadily, but the depth of model risk expertise inside ASEAN supervisors is still less than what assessing AI-driven pricing genuinely requires.

What is at stake

The pool, once broken, is difficult to reassemble. The traditional life and health products that have served Indonesian households for forty years rely on cross-subsidy across populations. If the lowest-risk customers leave for individually priced products, the pool that remains is more expensive to insure — and the premium increases force the next layer out, until the product becomes unaffordable for the population it was designed to serve.

This is not theoretical. The same dynamic is already visible in segments of the US auto insurance market, where telematics-based pricing has fragmented what used to be a pooled product. It will appear in ASEAN health and life insurance next.

What regulators should be doing

Three policy responses would meaningfully shape how this lands.

Define the floor. Regulators should specify a minimum level of pooling — categories of customers that must be priced together regardless of what AI models could distinguish. Certain health conditions, genetic information, age beyond a band — these can be enforced as non-disaggregable by rule.

Require disclosure on continuous pricing. Customers whose premiums are being recalculated based on ongoing data should be told how, at what frequency, and against what data. The opacity of continuous pricing is the most consequential consumer protection question of the decade.

Also Read: The AI stack trap: Why more AI tools aren’t translating into more growth

Build supervisor capacity now. AI insurance pricing will be deployed faster than supervisors can train to assess it. The window to bring model risk expertise into OJK, MAS, BNM and their equivalents is short.

The macro stakes

Insurance was built on the assumption that risks could be pooled. AI is making it possible to price individually at scale. The institutions that succeed in the next decade will be the ones that figure out how to use the new capability without dismantling the social contract the old model rested on. The regulators that succeed will be the ones who defined what counts as pooling before the market decided for them.

The actuarial profession will adapt. It always has. What will not adapt easily is the public’s trust in insurance as a mechanism that distributes risk fairly. That trust is harder to rebuild than the pricing model is to replace.

—

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 How AI is dismantling the risk pool in insurance appeared first on e27.

In the tech ecosystem, we often talk about the next big thing in terms of code and cloud scalability. But as we move further into 2026, I’ve realised that the most complex system I am currently architecting isn’t a SaaS platform. It is a physical puzzle hunt in the heartlands of Singapore.

I call it The Kampung Files.

If you search for it, you won’t find a landing page or a venture capital pitch deck. That’s because it is currently a live pilot (running June–August 2026), designed as a navigational and cognitive challenge for adults aged 55 to 65. It is my contrarian answer to a fundamental question: How do we apply high-level systems thinking to the real world?

High-tech logic for high-touch reality

While the industry is obsessed with the Metaverse, we are ignoring a massive, systemic friction point: the ageing human mind in an increasingly complex urban environment.

The Kampung Files applies the human rapid loop to social infrastructure. We aren’t just entertaining seniors; we are pressure-testing their cognitive readiness and navigational sovereignty. We use AI to architect the logic of the puzzles, but the vibe is 100 per cent human, rooted in the history of Toa Payoh’s hawker centres and town plazas. It’s a response to the Global Loneliness Epidemic, treating social isolation not as a feeling, but as a failure of neighbourhood architecture.

Solving non-profit fatigue with systems thinking

The social sector is often plagued by symptom management. If seniors are lonely, we give them a tea session; if they are inactive, we give them a gym. But as Stanford Social Innovation Review argues, real change requires moving from fixes that fail to Root-Cause Architecture.

Also Read: Tokenised assets have moved on-chain. The liquidity has not followed

This is the same logic driving my upcoming masterclass at The Foundry Singapore, where we will be applying Radical Essentialism to social impact. We aren’t teaching theory; we are performing a Systemic Audit of how organisations operate. By turning a neighbourhood into a game board, or a social hub into a high-agency ecosystem, we reveal the leverage points where human connection actually happens.

The build engine for the streets

Designing a community puzzle hunt requires a level of originality and intent that no AI can simulate. It requires an understanding of a specific neighbourhood’s pulse. However, the execution, the logistical maps, the logic gates of the puzzles, and the pilot timelines are accelerated by an Agentic Workflow.

By separating the architecture of intent from the mechanics of execution, we allow the human-in-the-loop to focus on what matters: the cultural nuance and the user experience. This allows us to move from an idea to a street-ready pilot in weeks, not years.

The conclusion: Architect for the soul

The ultimate hack for 2026 is realising that innovation that doesn’t reach the street level is just noise. Whether it is redesigning the leadership intelligence of a company through systems like the Octiverse or helping a 60-year-old navigate Toa Payoh with more confidence, the goal is identical: to use technology to lower the friction of being human.

Stop building for the machine. Start architecting for the Kampung.

—

The post The most sophisticated AI strategy is a puzzle hunt in Toa Payoh appeared first on e27.

The post Will Bitcoin hold its US$64,500 line in the sand, or are we heading straight for a test of the US$63,170 support level? appeared first on e27.