One participant in a week-long mobile diary sent us fourteen photos of receipts and shopping bags in seven days. A Saturday morning run to the wet market near her block for fish, pork and greens. A Sunday online order for rice, detergent, and two cartons of soya milk (delivered, she noted, in a two-hour slot). A Tuesday stop at a supermarket in the MRT station on her way home, because she'd forgotten garlic. Eggs from a convenience store at 10pm on Wednesday (her son had eaten them all). And a Friday visit to a Japanese specialty grocer for one bottle of mirin. Five channels, one household, roughly one week of meals. When we asked at the start of the study where she did her grocery shopping, she said "mostly online".
Here's what doesn't add up. Singapore is small, dense, and well served by supermarkets, so you might expect households to consolidate, pick one or two stores, and save time. The data suggests they are adding channels instead, and our diaries show them splitting even single meals across several stops. This post looks at the rules people use to decide which item goes where, why "mostly online" rarely means mostly online, and what that means for grocery brands and retailers trying to research a shopper who is never in just one place.
The numbers behind a fragmented basket
The first thing to pin down is how much of the food budget is actually in play. The SingStat Household Expenditure Survey 2023 found that households spent $1,422 a month on food, and that 67.9 per cent of it went to food and beverage serving services (meals out, from hawker centres to restaurants). So the grocery basket, the part people actually cook, is roughly the remaining third. That smaller basket is where the channel fight happens, and it's fought item by item.
Online is growing, though from a lower base than the hype suggests. The same survey shows monthly household spending on food bought online rising from $17 in 2017/18 to $86 in 2023, taking the online share of food spending from 1.4 per cent to 6.0 per cent. Enterprise Singapore notes that online sales as a share of total retail doubled from 6 per cent in 2019 to 12 per cent in 2020, the pandemic jump most of us remember. And the latest SingStat Retail Sales Index for April 2026 puts online sales at 13.0 per cent of supermarket and hypermarket sales. That figure is lower (much lower) than many participants' own estimates of how much they buy online.
Physical stores are hardly retreating either (whatever the think pieces say). Sheng Siong reported FY2025 revenue of S$1.57 billion, up 9.9 per cent, and opened 12 new stores in the year. FairPrice Group describes a network of over 570 touchpoints spanning supermarkets, hypermarkets, convenience stores and online. I read those two facts together as a sign that growth online and growth in stores are happening at the same time, in the same households.
Why the wet market survives on a narrow list of items
The wet market is the channel everyone assumes is dying, and the numbers are mixed. The Straits Times reported in May 2025 that the National Environment Agency manages 83 markets, that the median age of stallholders is 63, and that many patrons are around 60 and above. It also reported occupancy averaging 96 per cent over five years. So the markets are dying? No, that's not quite right. The stalls are full, and the people on both sides of the counter are ageing, which is a different problem with a longer fuse.
What keeps younger households coming back, at least some of the time? In our diaries it is almost never the full shop. It is a short list of items where people want to see, touch, or ask (fresh fish, cuts of pork for a particular soup, greens for a festive dish). And it is relationships. "The uncle knows I want the belly with less fat" is a sentence we have heard, in some version, more than once. That kind of service is hard for a supermarket chiller to replicate, and I suspect it's the real reason wet markets hold on to the protein basket long after rice and detergent have moved online.
There's a supply story underneath this too. The Singapore Food Agency's 2025 food statistics note that Singapore imports more than 90 per cent of its food, from over 180 countries and regions. Shoppers do not quote this. But they talk constantly about origin, and a stallholder who can say where today's fish came from carries an advantage that a printed label doesn't quite match.
Two questions decide where each item goes
We began by typing households, the online household, the market household, the supermarket household, and the first round of diaries scrambled every type. Actually, the household is the wrong unit altogether, because almost every household used several channels in the same week. The item is the unit. What decides the channel for an item seems to come down to two questions. Do I need to see it or touch it before I buy? And do I need it tonight, or can it wait? I've started calling this the Touch and Timing Matrix, though I'm still testing whether a third axis (price sensitivity, probably) deserves its own place.
The Touch and Timing Matrix
The Stock-Up Order is where online grocery has won, and it's easy to see why. Heavy items, predictable use, and delivery fees that reward a bigger basket. A MoneySmart comparison from September 2025 listed a $59 threshold for free delivery at FairPrice and a $100 minimum order at Sheng Siong, and participants describe padding a cart with toilet paper to cross the line. That padding is itself a channel rule. The Inspection Run is the opposite, and it's where the wet market and the specialty grocer still set the terms. The two top quadrants are about time, and they're where convenience stores and station supermarkets quietly win small, frequent, higher-margin baskets.
Does the matrix hold for every household? Probably not. Families with a domestic helper shift the Inspection Run to someone else (and the brand decision often shifts with it). Older shoppers treat the wet market trip as a social outing, which pulls items across the line that a younger shopper would happily order online. And fresh produce ordered online sits uneasily in the matrix, since some shoppers trust one retailer's picker and not another's.
What shoppers say about their grocery habits, and what the diaries show
Ask people to describe their grocery shopping and you get a tidy story. Follow them for a week and the story develops more stops.
| What participants say | What their week actually shows |
|---|---|
| "I buy mostly online now" | Online covers the heavy, predictable items, while fresh protein and top-ups still come from three or four other places |
| "I go wherever is cheapest" | Price decides the stock-up basket, but convenience decides the evening top-up, even at a visible premium |
| "Wet market is for old people" | Younger shoppers still make a planned trip for specific items, often before festivals or for a parent's recipe |
| "I plan my meals for the week" | Plans hold for about three days, then a mid-week supermarket or convenience store stop fills the gap |
| "I'm loyal to FairPrice" or "I'm a Sheng Siong person" | Loyalty holds for the big basket, while small baskets go to whichever store is on the route home |
That first row is the one that misleads the most. People count online orders because they are memorable (a delivery arrives, a notification pings, a receipt lands in the inbox). They undercount the unplanned stops, because a $6 purchase on the way home doesn't feel like shopping. This is one reason stated channel share runs so far ahead of the 13.0 per cent in the SingStat figures, and it's a pattern we see across categories, as I wrote in why Singapore consumer research keeps misleading brands. The same undercounting shows up with prepared food, which we traced in our research on food delivery behaviour.
Following a shopper who is never in one place
If you only intercept shoppers in one store, you study one quadrant and call it the market. So we start with the week. Mobile ethnography lets participants log each trip as it happens, with photos of the basket and a line about why they chose that stop. The diary captures the $6 purchases that memory drops, and it's the method behind the scenes in our ethnography of Singapore's flavour wars.
In the diary follow-up call: "Show me the one item this week you would never buy online, and the one you would never buy at the market. Tell me why, as if you're explaining it to a new neighbour." The two answers usually reveal the touch and timing rules faster than any direct question about channels.
Then the diaries need context from the shelf and from the person. Mystery shopping shows what a shopper actually finds at 9pm in a station supermarket, whether your product is in stock, faced, and priced as planned, a gap we describe in what happens when merchandising plans meet the supermarket floor. And in-depth interviews unpack the household politics (who shops, who cooks, whose mother's recipe sets the shopping list) that a diary only hints at. If you're new to commissioning this kind of work, our guide to market research in Singapore covers how the methods fit together. And brands launching into the fresh or chilled aisle may find a useful parallel in who actually buys plant-based food in Singapore, where the channel shapes the trial.
The basket is split, and the rules are consistent
Singapore households are not disloyal, and they are not confused. They run a fairly consistent set of rules, item by item, and those rules send the same product to different channels depending on whether it needs inspecting and how soon it's needed. A brand that asks "which channel do our shoppers use?" will get a single answer that is wrong. The better question is which quadrant your product lives in, and whether you are present, in stock, and visible there. I don't think the Touch and Timing Matrix is the final shape of this (price and household composition may need their own treatment). Fourteen photos in seven days, from a shopper who told us, sincerely, that she did it mostly online, is a fair summary of why the diary comes before the survey.
What food and retail brands ask about grocery shopping behaviour
Why do Singapore consumers use so many different grocery channels?
How much grocery shopping in Singapore happens online?
Are wet markets still relevant to grocery shoppers in Singapore?
How do you research grocery channel choice in Singapore?
What should grocery brands do with a fragmented shopper?
Finding out which grocery channel actually decides your product's sales
Singapore shoppers split one week of meals across five channels, and each product follows its own rule about where it gets bought. We use mobile diaries, mystery shopping, and in-depth interviews to show which rule governs your category and where your brand needs to be present when it matters.
Request a quote →
