ChatGPT Ads in Southeast Asia and Taiwan: the launch guide

Tarun Kapoor Updated 1 October 2026 17 min read

ChatGPT Ads is now available in Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam and Taiwan, following an expansion OpenAI announced in the week of 25 September 2026. These seven markets join the US, Canada, Australia, New Zealand, the UK, Japan, Korea, Brazil and Mexico, plus the September 2026 rollout to select countries in Europe, India, and the Middle East and North Africa. This guide covers the full current market list, how to launch in each new market, why you will usually want one ad account per market (country, currency and time zone are locked at creation), how to localise context hints by language, and how to measure a multi-market programme without mixing currencies and time zones.

Where are ChatGPT Ads available now?

As of 1 October 2026, ChatGPT Ads is available in the US, Canada, Australia, New Zealand, the UK, Japan, Korea, Brazil and Mexico; in select countries across Europe, India, and the Middle East and North Africa from the September 2026 rollout; and in Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam and Taiwan from the week of 25 September 2026.

Why does the Southeast Asia and Taiwan expansion matter?

It opens ChatGPT's conversational ad inventory in seven markets at once, several of them large, mobile-first and fast-growing. For advertisers already running Asia-Pacific programmes it removes the gap between Japan, Korea, Australia and New Zealand, making a genuinely regional ChatGPT Ads plan possible for the first time.

For regional brands and agencies, the practical shift is from “test in one market we happen to have access to” to “plan a regional rollout”. That raises structural questions that did not matter with one market: how many ad accounts, which currencies, which time zones, who owns billing in each country, and how to keep context hints consistent in intent while localised in language. The rest of this guide answers those in order.

One word of calibration. ChatGPT Ads is still a young channel everywhere, and brand-new markets start with little shared benchmark data. Treat the first months in each market as structured learning: fewer campaigns, clean measurement, and tight comparison against your other channels. Our cost guide and ROI guide cover how to frame early tests.

Should you use one ad account per market?

Usually, yes. Country or region, currency and time zone are fixed when an ad account is created and cannot be changed later, so an account set up for Singapore in Singapore time cannot later become your Vietnam account. One account per market — or at least per currency and time zone combination — keeps budgets, pacing and reporting clean.

The logic is straightforward. A daily budget resets on the account's time zone; if your account runs on a time zone several hours away from the market, your budget day and your customers' day do not line up. Spend and reported values are in the account's currency; mixing markets in one account means reporting some markets in a currency that is not theirs. And each account carries one Brand Name and one Legal Name — which matters if different legal entities bill in different countries.

My default recommendation for anything beyond a small test: one account per market, created deliberately, with a consistent naming pattern such as Brand | TH | currency | Ecommerce. Our account setup guide covers naming, the three advertiser names, roles and the 10-account creation limit in detail — agencies running seven new markets for several clients will hit that limit quickly unless clients create their own accounts and invite the agency.

How do you launch ChatGPT Ads in a new Southeast Asian market?

Confirm the market on OpenAI's availability page, decide the account's country, currency and time zone, sign up at ads.openai.com, set the Account, Brand and Legal names, add billing, install tracking on the local site, write context hints in the local language, and launch a small, clearly measured first campaign before scaling.

1. Confirm availability and terms. Check the Ads Manager availability page for the market, and note the currency and any minimum budgets shown there or during signup.
2. Decide the account's permanent settings. Country/region, currency and time zone. Write them down before you start.
3. Sign up at ads.openai.com with your OpenAI login. For a deliberate market setup, skip the website-draft option and create the account first.
4. Set the three names in Settings → General: an internal Account Name that includes the market, the Brand Name customers in that market recognise, and the Legal Name of the entity being billed.
5. Add billing — credit card, business name, billing address, invoice email. Required before delivery.
6. Invite users with the right roles, including local team members as Viewers or Members.
7. Install tracking on the market's site (pixel and ideally Conversions API) and verify events arrive. See conversion tracking.
8. Connect a product feed if you sell online, with local prices and availability. See the product feed guide and bulk feed campaigns.
9. Write localised context hints (next section).
10. Launch a small first campaign with clear success criteria, then scale what works.

What differs market by market?

The mechanics of Ads Manager are the same everywhere; what changes is language, how people phrase questions, local product vocabulary, the landing experience and the legal entity you bill through. Plan each market around those, starting with the language people actually use when they ask ChatGPT for help.

The table below summarises the language picture for each new market. It is general market knowledge to guide localisation, not an OpenAI statement about targeting or language support — OpenAI has not published language-targeting specifics for these markets in the sources I use.

Indonesia

Write hints in Indonesian as people speak it when asking for help, not formal marketing Indonesian. Product categories often have borrowed English terms alongside local ones — include both where real people use both. Landing pages should be in Indonesian with local pricing.

Malaysia

The multilingual market in this group. Decide which language audiences you are serving — Malay, English, Chinese — and write a hint set per language. A single English set will reach some users but miss conversations happening in other languages.

The Philippines

English-language hints work well here, and many advertisers can start with them. Watch for mixed English–Filipino phrasing in how people describe problems, and add hints that reflect it once you have evidence it matters for your category.

Singapore

English-first for most commercial categories, and often the natural first market for regional advertisers because many already have Singapore entities and English-language sites. Be careful not to let Singapore become the de facto “regional” account out of convenience — its currency and time zone are locked to it.

Thailand

Localise fully into Thai. English-only hints and English landing pages will underperform for most consumer categories. Invest in native-speaker review of hints rather than machine translation alone.

Vietnam

Localise fully into Vietnamese, with correct diacritics — unaccented Vietnamese reads as careless and can change meaning. Include local brand and product terms users would name.

Taiwan

Write in Traditional Chinese with Taiwan vocabulary. Do not reuse Simplified Chinese copy or mainland terminology; many everyday product words differ, and the mismatch is immediately noticeable to readers.

How should you localise context hints?

Write context hints in each market's language, describing the conversations people there actually have — not translations of your English hints. Keep the same intent structure across markets (audience, intent, topic), but let the wording, examples and product terms be local. Have a native speaker write or review every set.

Context hints are the primary targeting signal in ChatGPT Ads: short descriptions of the conversations and topics where your product is relevant. A literal translation of an English hint often describes a conversation nobody in the target market has, in words nobody there uses. The approach that works is to keep the strategy constant and localise the language:

1. Start from your intent map. List the buyer intents your English hints target — problems, comparisons, purchase moments. Our writing guide uses the Audience–Intent–Topic framework.
2. Brief a native speaker per market with the intents, not the English wording.
3. Ask them to write how a local buyer would describe the situation, including local product terms, brands, price expectations and colloquialisms.
4. Check against real language. Local search suggestions, forums, marketplace reviews and customer-service transcripts show how people actually phrase things.
5. Keep a parallel reference sheet mapping each localised hint to the English intent it serves, so you can compare performance across markets by intent.
6. Iterate per market. A hint that wins in Singapore says little about the equivalent in Thailand; read each on its own data.

For how context hints differ from keywords — and why literal translation is even less effective than it is in search — see context hints vs. keywords and OpenAI ads intent targeting.

What about creative and landing pages?

Localise the ad and the landing page together. An ad in Thai that lands on an English page, or a Taiwan ad that shows Simplified Chinese or a foreign-currency price, breaks trust at the moment of the click. Match language, currency, pricing, delivery information and payment options to the market the account serves.

• Language: the landing page in the same language as the ad and hints.
• Price and currency: local currency on the page; for feed campaigns, local prices in the feed.
• Delivery and returns: say clearly whether you ship to that market, how long it takes and what it costs.
• Payment: show the payment options people in that market expect.
• Brand Name: use the name customers in that market know. If you trade under a local name, that is the Brand Name for that account.

Our creative guide and ad anatomy guide cover what the ad unit looks like and what earns the click.

How do you measure a multi-market programme?

Measure each market in its own account, currency and time zone, then convert and combine in your own reporting layer. Pull spend and conversions per account through the Insights API, convert currencies at a consistent documented rate, align dates carefully, and compare markets on ratios such as ROAS and cost per acquisition rather than raw totals.

1. Track per market. Each market's site or storefront should send events to the matching ad account. Verify with the event quality checks.
2. Pull per account. The Insights API returns spend, conversions and attributed sales per campaign; the response includes account_currency. For purchases on campaigns optimising toward other events, use non-goal conversion reporting.
3. Keep windows identical across markets so comparisons are fair.
4. Convert currency in one place with a documented rate source and date. Never mix converted and unconverted figures in one table.
5. Mind the time zones. A “day” in each account follows its own time zone. For daily regional totals, either accept slight misalignment and say so, or aggregate weekly.
6. Compare on ratios. ROAS, cost per acquisition and conversion rate travel across currencies; raw spend and revenue do not.
7. Tag traffic. Consistent URL parameters per market make analytics-side reporting reliable.

What should the first 30 days in a new market look like?

Spend the first week on foundations — account, names, billing, tracking, localised hints — then launch a small number of campaigns around your clearest buyer intents. Use weeks two to four to read results by intent, fix tracking and landing-page gaps, and decide which intents deserve more budget. Do not judge a market on its first few days.

Two habits make this work. First, write the success criteria before launch — a target cost per acquisition or ROAS range, based on what the market is worth to you, not on another market's benchmarks. Second, keep a short log of every change per market, so that when results move you know whether it was the market or your own edits.

Which advertisers should move first?

Advertisers who already sell in these markets with localised sites, local pricing and working tracking should move first; they can test without building new infrastructure. Brands that would need to localise sites, set up local payment or shipping, or create new legal entities should build those foundations before spending on ChatGPT Ads there.

• Strong early candidates: regional ecommerce brands with local storefronts and product feeds; software and services companies with English-first buyers in Singapore and the Philippines; travel, education and financial-comparison businesses with established local sites.
• Build first, then advertise: brands without local-language landing pages, without local delivery, or without a way to bill in the market.
• Watch and prepare: advertisers in categories with strict local advertising rules should check both local regulation and OpenAI's ad policies before launching.

What rollout sequence works for regional advertisers?

Start with the one or two markets where you already have a strong localised site, working tracking and clear economics — often Singapore or the Philippines for English-first brands. Prove the measurement and structure there, then add markets in waves, localising hints and landing pages properly for each before it launches.

The right order depends entirely on where your business already works well; the table is a pattern, not a prescription. What matters is not launching seven markets at once with one English hint set and one account, then trying to untangle the results.

What should agencies in the region plan for?

Plan account ownership and the 10-account creation limit before onboarding clients across seven markets. Have clients create their own market accounts and invite the agency, standardise naming, keep a shared intent map per client with native-written hints per market, and report regionally in one currency with a documented conversion method.

• Account creation limit: once you belong to 10 or more ad accounts, created or invited, you cannot create another yourself. A few clients across seven markets passes that fast.
• Ownership: client-created accounts with agency invites keep billing and ownership with the client.
• Localisation capacity: budget for native-speaker hint writing per market; it is the step most often skipped.
• Reporting: build the regional roll-up once and reuse it across clients.

See our guide to ChatGPT Ads agencies, and for resellers the white-label service.

What mistakes should you avoid?

• Creating a “regional” account in the wrong currency or time zone. It cannot be changed.
• Assuming currencies or minimum budgets. Check the availability page and signup flow.
• Machine-translating English hints. Localise from intents, with native speakers.
• Mismatched landing pages. Language, currency and delivery information must match the market.
• Using Simplified Chinese in Taiwan. Use Traditional Chinese and local vocabulary.
• Mixing currencies in reporting. Convert in one place with a documented rate.
• Launching every market at once. Prove structure and tracking in one or two first.
• Forgetting the Brand Name per market. If you trade under a local brand, set it before launch — later changes pause serving until review completes.

For the fundamentals, start with how ChatGPT ads work and how to run ads on ChatGPT. For how ChatGPT Ads compares with other platforms you may already run in the region, see ChatGPT Ads vs. Google Ads.

Frequently asked questions

Sources and further reading

• OpenAI — “What's new in Ads Manager” weekly note, week of 25 September 2026 (expansion to Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam and Taiwan).
• OpenAI Help Center — Ads Manager availability (authoritative market list; check for currencies and budgets).
• OpenAI Help Center — Ads Manager account setup (locked country, currency and time zone; multiple accounts; retrieved 1 October 2026).
• OpenAI for Developers — Ads API reporting documentation (Insights API, account currency, CTR as ratio).

Tarun Kapoor

Founder & CEO, Context Hints

Twelve years of media buying across GroupM, WPP, Ogilvy & Mather, and Neil Patel Digital. Has personally owned media for Nestlé, Sage, Qualcomm, Aetna, Weight Watchers, Chubb and Novotel.