Asia's online channel is a buying-task map

When a brand asks whether Southeast Asian retail is "moving online," a single percentage rarely answers the question it was actually asking. NIQ's 2025 Asia channel analysis treats online as a channel in its own right, sitting alongside modern and general trade. For planning purposes, the productive move is to map that channel to specific purchase tasks instead of collapsing it into one ecommerce share.

The study draws on actual purchase data from more than 18,000 households across Indonesia, Singapore, urban Thailand, the Philippines, and Hong Kong. That scope is worth stating plainly, because it sets the limits of what the data can settle. This is not a platform GMV ranking, and it is not a full national read for every market. It is a household-purchase view of fast-moving consumer goods and adjacent retail behaviour, taken across markets where the same word — "online" — describes very different things.

Start with the channel structure

NIQ separates the market into modern trade, general trade, and online. Modern trade covers minimarkets, supermarkets, and hypermarkets. General trade covers local markets, traditional stores, and small neighbourhood shops such as sari-sari stores in the Philippines. Online covers ecommerce and digital grocery behaviour.

That split alone explains why one ecommerce strategy rarely travels across Asia. Singapore, urban Thailand, and Hong Kong lean more heavily toward modern trade, while Indonesia and the Philippines retain a large general-trade base. Online is growing everywhere, but it is not displacing the rest of the system along one clean line. It attaches itself to particular missions: discovery, price comparison, planned replenishment, convenience, repeat purchasing.

For a category team, the first practical implication is where to begin. Starting from "online share" invites a budget answer before anyone has defined the problem. Starting from the buying task asks something answerable: is this product being discovered, compared, replenished, bought urgently, or bought again once trust exists?

Mature markets do not behave the same

Singapore and Hong Kong are both mature, high-density markets, and NIQ's analysis still shows them running on different channel logics.

In Singapore, high living costs and cross-border shopping shape how households use retail channels. NIQ highlights cross-border growth and the meaningful share of households that shop outside the market. Online grocery and supermarket delivery matter there because they fit a time-saving, planned-purchase use case — not because every category naturally migrates to a marketplace.

Hong Kong reads differently. Supermarkets remain an important main channel, though growth is flatter, and online food and grocery behaviour attaches to local operators such as HKTVmall and Wellcome rather than to a generic marketplace-first structure. Digital maturity, in other words, does not produce the same channel mix everywhere. Dense cities still split online behaviour by category, delivery trust, basket size, and existing retailer relationships.

The risk for a brand is over-reading maturity. A market can be entirely comfortable with digital and still keep routine FMCG purchases inside supermarket or convenience networks. Digital readiness makes online execution possible; it says nothing about whether the product's core buying mission has moved there.

Thailand needs a food and non-food split

NIQ's Thailand read is useful precisely because the sample is explicitly urban. Modern trade leads, and CJ Supermarket's buyer base reportedly doubled over two years as the chain expanded stores and promotions. Online shopping, meanwhile, is pushed largely by non-food categories, with online food also growing quickly.

The U.S. International Trade Administration's Thailand ecommerce guide adds a further layer: mobile accounts for more than 80% of online sales in Thailand. That figure confirms how online access is reaching people. It does not license the conclusion that every category should be run like a TikTok Shop campaign. Mobile is the interface; the buying task still decides the channel.

For non-food, the decision path tends to run through content, marketplace search, price comparison, reviews, and returns. For food and FMCG it is more operational: availability, delivery reliability, basket economics, promotions, and whether the shopper trusts the channel enough to replenish through it. The same phone supports both paths. The commercial playbook behind them is not the same.

Indonesia and the Philippines: convenience is not always online

Indonesia and the Philippines correct a platform-only reading of Southeast Asia. NIQ's analysis points to the continued strength of general trade in both markets, even as minimarkets and online channels grow.

In Indonesia, general trade remains an important base, while minimarkets gain ground by solving proximity and frequency. Online accelerates when price, delivery incentives, and category fit line up together. Personal care and beauty travel through marketplace and creator-led paths more naturally than everyday staples do, because those products can be discovered, compared, reviewed, and bundled before purchase.

In the Philippines, sari-sari stores hold their position by solving a very specific task: immediate, nearby, small-basket purchasing. For many households, convenience is not a delivery promise. It is a shop within walking distance, open when needed, willing to sell a smaller pack size. Online can grow strongly in non-food without touching that role.

None of this means general trade blocks ecommerce. It means ecommerce has to be mapped around the task it is actually replacing or complementing. Where the task is instant need, neighbourhood retail holds a structural advantage. Where it is discovery or comparison, platforms win more easily. Where it is replenishment, fulfilment reliability and stock availability count for more than top-of-funnel reach.

Use a buying-task map before setting channel budget

The working model is simple:

Buying task	Likely channel pressure	What to measure
Discovery	Creator content, marketplaces, social commerce	Search lift, creator assisted sales, new-buyer mix
Comparison	Marketplaces and price aggregating surfaces	Price position, review depth, conversion by SKU
Replenishment	Supermarkets, minimarkets, delivery operators	Repeat rate, stock-outs, basket size, delivery success
Immediate need	General trade, convenience, local stores	Availability, pack size, distance, frequency
Trust-led repeat purchase	Retailer relationships and owned CRM	Retention, subscription, complaint rate, service recovery

The map also keeps metrics in scope. Platform GMV, online share, household penetration, traffic, recommendation scores, and retail sales are not interchangeable. A marketplace GMV number shows where transactions concentrate. Household-panel data shows what people actually buy. A government ecommerce guide shows infrastructure and payment direction. Each answers a different question, and substituting one for another is how channel plans go wrong quietly.

So what

The decision in front of most brands is not whether to "go online" in Asia. It is which buying task the category needs to win.

For beauty, personal care, home goods, and other comparison-heavy categories, online channels can carry discovery and evaluation. For grocery and routine FMCG, online matters, but the operating system still runs on modern trade, minimarkets, delivery reliability, and stock discipline. In markets with strong general trade, distribution and pack architecture remain part of the ecommerce strategy, because customers do not experience channels the way a strategy deck presents them. They experience them as the fastest way to settle the purchase in front of them.

Use NIQ's channel frame as a planning layer before allocating platform budget: define the buying task, choose the channel that naturally serves it, then attach the metric that can judge it. That sequence avoids the common mistake of treating "online" as a destination when it is one part of a wider Asian retail map.