Media Buying in Southeast Asia: Geos, CPMs, Offers

Why are media buyers moving budgets to Southeast Asia?

Buyers are shifting spend to SEA because inventory is cheap, volume is real, and the region is not yet priced like the US or UK. A Meta CPM that runs $8-15 in an American nutra vertical often runs $0.50-2 across ID, PH, and VN for comparable placements. Combine that with 680 million people and mobile-first internet adoption, and the math on testing budgets changes fast.

The push is not purely about cost. Shopee and Lazada normalized online buying across the region over the past six years, which trained consumers to trust delivery-based purchase flows even where card penetration stays low. Regulatory attention on nutra and finance claims exists in SEA, but enforcement infrastructure is thinner than the FTC's in the US or the ASA's in the UK, and buyers who assume that means zero risk get burned. Currency also matters: payouts in USD against local ad spend in depreciating currencies like the Philippine peso or Vietnamese dong stretch margins further than the same math in a dollar-denominated market. None of this makes SEA easy. It makes it underpriced relative to what still converts.

Which SEA countries convert best for nutra and finance?

Vietnam and the Philippines convert nutra offers more consistently than Indonesia does per dollar spent, even though Indonesia has three times the population. Thailand leads for finance and crypto-adjacent offers. Malaysia is smaller in volume but useful for bilingual testing before a Thai or Vietnamese launch. Indonesia still moves the most raw volume — it just costs more in wasted spend to get there.

The Indonesia-first instinct is mostly a population-size heuristic, and it is wrong often enough to name. Indonesian nutra placements have been worked by regional media buying teams since roughly 2019, which means the cheap-angle, cheap-creative version of most offers has already run there. Vietnam has a shorter run history in paid nutra, cheaper CPMs across both Meta and TikTok, and COD acceptance that runs comparable to Indonesia's. A buyer testing a new angle with a $500-1000 budget will generally get a cleaner read in Vietnam than in Indonesia, because the noise floor is lower. That does not make Indonesia a bad market — it makes it a scaling market, not a testing one.

What CPMs should you expect per platform and geo?

Expect $0.30-1.20 CPMs on Meta and TikTok across Indonesia, the Philippines, and Vietnam, rising to $0.80-2.50 in Thailand and Malaysia, with Google Display running slightly below Meta in every geo listed. These are ranges pulled from buyer-reported figures and network blog posts, not platform-verified averages, and they move with seasonality, vertical, and account trust score — treat them as a planning band, not a quote.

Statista's regional digital ad spend estimates put SEA's total online ad market growth above 15% annually through the mid-2020s, and that growth is uneven across platforms — TikTok inventory in the region has expanded faster than Meta's has, which is part of why TikTok CPMs still sit slightly under Meta's in most of these geos despite the platform's growing popularity with brand advertisers. Run a small-budget CPM check in your own ad account before committing to any of these numbers. Account history, placement mix, and audience targeting move actual delivered CPM by 2-3x inside the same country.

Why does cash on delivery dominate SEA checkout?

COD dominates because large parts of the SEA population remain unbanked or underbanked, and online card payment still carries a trust deficit that pre-paid checkout does not overcome. The World Bank's Global Findex data has tracked this gap for years — financial account ownership in Indonesia, the Philippines, and Vietnam trails East Asian and Western markets by a wide margin, and cash remains the default settlement method for most retail transactions.

COD funnels look different from card-based ones structurally. The landing page collects name, phone number, and delivery address — no card field, no payment gateway integration, no cart abandonment at the payment step. A courier collects cash on delivery, sometimes verified by phone call before dispatch. That structure removes the single biggest drop-off point in a Western checkout flow. It replaces it with a different one: decline and return rates on COD orders commonly run 30-50%, since a customer who agreed to receive a package feels far less commitment than one who already typed in a card number. Funnels built for SEA account for that shrinkage in the offer's payout math from the start, not as an afterthought.

Which languages do winning SEA creatives use?

Winning SEA creatives run in six languages: Bahasa Indonesia, Filipino (often mixed with English as Taglish), Vietnamese, Thai, Bahasa Malaysia, and straight English for urban Singapore and Malaysia audiences. Direct translation underperforms. Creative that wins locally is written by someone fluent in the market's slang, humor, and purchase objections — not machine-translated from an English-language VSL script.

Testimonials need to sound like the market, not like a translated American ad. A Filipino audience responds to Taglish code-switching in a way a straight Filipino script does not replicate. Vietnamese creative performs better with local currency figures (VND, not USD conversions) and domestic celebrity or influencer references where usage rights allow it. Thai script direction and reading patterns differ enough from Latin-alphabet languages that text overlay timing in video ads needs separate testing, not a reused template. A single English master ad, dubbed six ways, will underperform six creatives built natively in each language — the cost difference between the two approaches is usually smaller than buyers assume, and it shows up directly in CTR.

How do you spy on ads across SEA markets?

You spy on SEA ads with a combination of Meta's Ad Library for scale signals, a paid tool like AdSpy or BigSpy for searchable creative archives, and manual checks from a real local device or mobile proxy — because automated spy tools frequently get served decoy creative when regulated-niche advertisers detect a datacenter IP or crawler pattern. None of the three methods alone gives a complete picture.

The Ad Library is genuinely useful for one specific thing: confirming that an advertiser is actively running a meaningful number of live ads right now, which is a scale signal worth acting on. It is not reliable for seeing the actual converting creative in nutra or finance, since advertisers running regulated offers commonly cloak based on IP and device fingerprint, and a lookup from a US-based research machine or a known spy-tool IP range gets served a compliant placeholder instead of the real ad. AdSpy's published pricing sits in the $149/month range for standard access, and it indexes a large creative set, but it inherits the same blind spot — its crawlers are still crawlers. The reliable workaround stays manual: a real SIM card or residential mobile proxy located in the target country, browsing Facebook and TikTok at different times of day the way a genuine local user would. It is slower than a spy-tool dashboard. It also shows you what is actually running, which the dashboard frequently does not in these verticals.

Country-specific execution — payout tables, top offer categories, and network shortlists for Indonesia, the Philippines, and Vietnam — gets covered in the linked country pages. This page is the map. The next three are the routes.

Frequently asked questions

What is the best entry-level country for media buying in Southeast Asia?

Vietnam is often the cleaner starting point for nutra testing, not Indonesia, despite Indonesia's much larger population. Vietnamese CPMs run lower and the market carries less saturation from years of prior media buyer activity, so a small test budget produces a more readable signal there than in Indonesia's more competitive inventory.

How much does it cost to run test budgets in SEA compared to the US?

CPMs across Indonesia, the Philippines, and Vietnam commonly run $0.30-1.20 on Meta and TikTok, roughly a fifth to a tenth of comparable US nutra placements. That gap lets buyers run wider creative and angle tests for the same total spend, though actual delivered CPM still depends heavily on account history and targeting.

Why do SEA nutra and finance funnels rely on cash on delivery instead of card checkout?

Large portions of the SEA population remain unbanked or underbanked, per the World Bank's Global Findex data, and online card payment still carries a trust deficit in the region. Removing the card field and collecting cash on delivery lowers checkout friction, though it introduces 30-50% decline and return rates that funnels must account for.

Can I rely on the Meta Ad Library to see what nutra ads are actually converting in SEA?

Not reliably. The Ad Library is a solid signal for confirming an advertiser is running ads at scale, but regulated-niche advertisers frequently cloak creative based on IP and device fingerprint, so automated lookups often return decoy or compliant-placeholder versions instead of the live converting ad.

Do I need separate creative for each SEA language, or can I dub one master ad?

Native creative built separately for Bahasa Indonesia, Filipino, Vietnamese, Thai, Bahasa Malaysia, and English consistently outperforms a single dubbed master ad. The cost gap between the two approaches is usually smaller than buyers expect, and the difference shows up directly in click-through rate across markets.

Sources

Named rather than linked — verify before relying on any figure below.

• Meta's Advertising Standards and Ad Library documentation
• AdSpy's published pricing page
• The World Bank's Global Findex Database
• Statista's Southeast Asia digital advertising spend estimates