The bubble tea business in Singapore, decoded.
Singapore drinks more bubble tea per dollar than almost anywhere in the region — and the business has already crashed once. In 2002 the city hit roughly 5,000 outlets; by 2003 the bubble burst and chains folded overnight. Today it's booming again, yet Gong Cha shut all its Singapore stores in a single day in October 2025. Here is the picture the queue photos skip: the real cost to open, franchise versus independent, the rent-and-hype maths, what Nutri-Grade did to the cup, and the few places a new shop genuinely wins.
~5,000
outlets at the 2002 peak — then the crash
~60
competing brands today
82%
of young F&B closures never profited
~18–20 tsp
sugar in a full-sugar brown-sugar boba
The MOAT Score: is a bubble tea shop worth building?
Before the details, our one-number read. The MOAT Score grades a sector's economic quality on the value-investing lens of Graham, Buffett and Munger — four pillars (Margin, Operating moat, Appetite, Treadmill), each out of 25. Bubble tea scores one notch above the café we cover — a higher gross-margin ceiling and a smaller shop — but it shares the price-taker trap and adds two drags the café doesn't carry: franchise fragility and a sugar-regulation headwind.
Winner-takes-most
Winner-takes-most — hard.
How the score is built
The MOAT Score sums four pillars — each scored 0–25 — from the value-investing lens of Graham, Buffett and Munger. No black box: here is the working.
Margin
8/25Does the average operator actually keep money — real net margin and return on the capital tied up?
This sector: the per-cup gross margin is genuinely high — ingredients cost well under S$2 on a S$5 drink — but franchise royalties skim the top and 82%SourceMTI parliamentary reply, Nov 2025 of young F&B closures never recorded a profit.
Buffett 1979 — “a high earnings rate on equity capital… without undue leverage”; 1986 owner earnings.
Operating moat
5/25Pricing power and a durable competitive advantage — can a typical operator raise prices and have customers shrug?
This sector: even more of a commodity than a café — ~60 near-identical brands compete on hype, price and location, and a franchise brand can be pulled overnight (Gong Cha went dark in 2025).
Buffett, FCIC 2010 — pricing power is “the single most important decision”; 1991 franchise; 2007 moat.
Appetite
13/25Demand durability — steady, recession-resistant repeat demand vs fragile, discretionary or faddish.
This sector: habitual, youth-led demand — Singapore order values run ~2× the region — but more discretionary and fad-cycle-exposed than coffee, with sugar regulation a structural headwind.
Graham, Security Analysis Ch.2 — inherent stability “derives from the character of the business”.
Treadmillinverted · less is better
9/25Capital intensity and structural drag — rent, churn, fashion, discounting. Scored inverted: less treadmill, more points.
This sector: a smaller footprint means lighter fit-out capital than a full café, but brutal rent-per-sqft, queue dependence, franchise fees and Nutri-Grade compliance keep the treadmill turning.
Buffett 2007 — “the worst sort of business… requires significant capital… Think airlines.”
M + O + A + T, out of 100
The MOAT Score is a transparent SGAI judgement on a sector’s economic quality through a value-investing lens — not a verdict on any individual business, and not a comment on an owner-operated livelihood (a sector can score low on capital returns yet work as a job).
The read: a fat-margin cup on a rent-and-hype treadmill — and the brand might not be yours
Bubble tea has the best-sounding pitch in F&B: a S$5 drink whose tea, tapioca and powder cost well under S$2, sold from a box a fraction the size of a café. The gross margin is real. The problem is everything after the gross line — and the fact that, more than coffee, this is a hype business, where demand for any one brand can decay as fast as it spiked.
Singapore has run this experiment to its conclusion before. By 2002 the island had something like 5,000 bubble tea outlets — cake shops and even shoe shops were bolting on a tea counter. Then the price wars came (one-for-one, dollar cups), quality became indistinguishable, and the bubble burst. Each-A-Cup went from 51 outlets in 2001 to about 15 by 2003. Undifferentiated expansion into a fad is the oldest way to die in this category, and the founders who survived 2003 still talk about it.
The modern twist is the franchise model. A franchise buys you a known name and a supply chain — but the name is rented, not owned, and it can be switched off. Gong Cha closed every Singapore outlet on a single day, 1 October 2025, when its local franchise agreement ended. The biggest homegrown chain, LiHO, exists only because that same brand was lost once before — its operator brought Gong Cha to Singapore in 2009, lost the franchise in 2017, and rebranded roughly 80 stores overnight. Milksha exited entirely at the end of 2023. The lesson the lens draws is blunt: in a commodity category, a moat you rent is not your moat.
A barbell market — and the average cup is a price-taker
The category has split into an ultra-cheap value tier (Mixue) and a premium tier (CHAGEE), squeezing the mid-market where most of the recent exits sit. The average operator sets price against the brand next door, not against a captive customer.
- Mixue (value tier)S$2.5
the new ~S$1–4 floor that did not exist in earlier waves
- Mainstream pearl milk teaanchorS$4
KOI / LiHO / the old Gong Cha — the commodity middle
- Premium Chinese tea (CHAGEE)S$9
the up-market end, ~S$7.30–10.20 a cup
Source: SGAI price survey of SG bubble-tea formats, 2024–2026 (indicative; Mixue & CHAGEE menus, Daniel Food Diary / aggregator pricing)
The map: a slice of a ~S$12B F&B market — and the “bubble tea market size” numbers are junk
There is no official Singapore bubble-tea market size. SingStat measures total F&B services receipts at roughly S$12B/yearSourceSingStat F&B receipts, 2025 (online ~24–26%), but does not break out bubble tea. The best independent estimate puts Singapore at about US$342M/yearSingle source — not independently corroboratedMomentum Works × qlub, 2021 dataSecondary reporting of the Momentum Works × qlub 2022 report (2021 data). The best available estimate, not an official figure. — 4th in Southeast Asia. Ignore the aggregator reports claiming the Singapore bubble-tea market is US$9.2MShaky figure — treat with cautionmarket-size aggregatorImplies under S$13M/yr spent on bubble tea across ~60 chains — ~37× smaller than the Momentum Works estimate and not credible (KOI alone reportedly did ~S$45M revenue in 2017). Anchor on SingStat + Momentum Works.; that figure is roughly 37× smaller than the credible estimate and is not believable for a market with ~60 chains.
~S$12B
F&B services / year
SingStat 2025 (bubble tea not broken out)
~US$342M
SG bubble-tea spend / yr
Momentum Works 2021 est. — flag
~60
competing bubble tea brands
Momentum Works 2022 / Vulcan Post 2021
2,431 vs 3,357
F&B closed vs opened (Jan–Oct 2025)
MTI, Nov 2025 — the market runs hot
Boom, bust, and a crowded comeback
The number every old guide half-remembers — the ~5,000-outlet peak — is the cautionary tale, not the present. The category crashed in 2003, rebuilt on premium Taiwanese brands, and is crowded again today.
2002—the first-wave peak — then the bubble burst
2003—mass closures; Each-A-Cup 51 → ~15 outlets
2011—rebuilt on premium brands (KOI 2007, Gong Cha 2009)
2025—~60 brands; crowded, premiumising, consolidating
Stale: the "5,000 outlets" figure older write-ups still cite as if current
Source: NLB Singapore Infopedia (2002 peak ~5,000; Each-A-Cup 51→15 by 2003; ~500 outlets / >10 chains by 2011). 2025 point indicative — there is no official current outlet census.
Outlet counts are indicative of the shape of the cycle, not a census — Singapore keeps no official running tally of bubble-tea outlets, and brand store-locators lag reality badly.
The players: franchise giants, value disruptors — and the exits
The single most predictive question about any chain here is not how long the queue is — it's who controls the brand, and whether the model is being squeezed from the cheap end, the premium end, or both. The mid-market is where the casualties cluster.
Who you're really studying — and who already left
| Player | Model | Tier | Brand control | SG signal |
|---|---|---|---|---|
| LiHO TEALargest homegrown (ex-Gong Cha SG) | Mostly company-owned | Mainstream | owned | Born from a lost franchise (2017 rebrand of ~80 stores) |
| KOI ThéSecond-wave pioneer (SG 2007) | Franchise / operator | Mainstream+ | franchise | ~S$45M revenue, ~23% margin (2017, single-source) |
| MixueChinese value giant (SG since 2022) | Franchise / supply-chain | Value (S$1–4) | franchise | The new cheap floor; ~32% group gross margin (HK-listed, global) |
| CHAGEEPremium Chinese tea (re-entered 2024) | Switched to company-owned | Premium (S$7–10) | owned | Dropped franchise, went company-owned; ~13 outlets by mid-2025 |
| Gong ChaExited SG 1 Oct 2025 | Franchise — agreement ended | Mainstream | franchise | All outlets dark in one day; promises a 2026 return |
| MilkshaExited SG 31 Dec 2023 | Franchise / operator | Mainstream | franchise | Full exit — "growing competition and high costs" |
LiHO TEA
Largest homegrown (ex-Gong Cha SG)
- Model
- Mostly company-owned
- Tier
- Mainstream
- Brand control
- owned
- SG signal
- Born from a lost franchise (2017 rebrand of ~80 stores)
KOI Thé
Second-wave pioneer (SG 2007)
- Model
- Franchise / operator
- Tier
- Mainstream+
- Brand control
- franchise
- SG signal
- ~S$45M revenue, ~23% margin (2017, single-source)
Mixue
Chinese value giant (SG since 2022)
- Model
- Franchise / supply-chain
- Tier
- Value (S$1–4)
- Brand control
- franchise
- SG signal
- The new cheap floor; ~32% group gross margin (HK-listed, global)
CHAGEE
Premium Chinese tea (re-entered 2024)
- Model
- Switched to company-owned
- Tier
- Premium (S$7–10)
- Brand control
- owned
- SG signal
- Dropped franchise, went company-owned; ~13 outlets by mid-2025
Gong Cha
Exited SG 1 Oct 2025
- Model
- Franchise — agreement ended
- Tier
- Mainstream
- Brand control
- franchise
- SG signal
- All outlets dark in one day; promises a 2026 return
Milksha
Exited SG 31 Dec 2023
- Model
- Franchise / operator
- Tier
- Mainstream
- Brand control
- franchise
- SG signal
- Full exit — "growing competition and high costs"
The cautionary tale: Gong Cha goes dark overnight
On 1 October 2025, every Gong Cha outlet in Singapore closed at once — website, app and delivery pulled — because the franchise agreement with the local operator ended. A household name for 16 years vanished from the island in a day. The global brand says it will return in 2026 with a new master franchisee and a “Gong Cha 2.0” concept. If you franchise, you are renting demand: read the term, the royalty, the supply lock-in and the exit clause as if the brand could be switched off, because it can.
The customer: young, habitual, and willing to pay more than the region
Bubble tea in Singapore skews young — Gen-Z and young adults — and the habit runs deep: during the 2020 Circuit Breaker, shutting bubble tea as “non-essential” triggered last-order queues and national hand-wringing. The defining trait of the Singapore buyer is spend, not volume: the average order here runs about 2×Single source — not independently corroboratedMomentum Works × qlub, 2022Average bubble-tea order value in Singapore ~2× the SEA regional average — smallest population, highest purchasing power of the six markets studied. the regional average. That means the winning move is usually to trade up, not to chase Mixue down to the floor.
Who actually buys (indicative)
There is no clean public census of the SG bubble-tea customer; this is SGAI's synthesis of survey signals and reporting — directional, not measured. Taste and price are the two stated drivers (69% / 61% in a 2020 SG survey).
- Students & Gen-Z40%the core daily-habit base; price- and trend-sensitive
- Young working adults35%afternoon pick-me-up / treat; willing to trade up
- Families & older adults15%occasional; the segment most moved by sugar concern
- Tourists & passers-by10%low repeat; queue-driven
Source: SGAI synthesis of Statista 2020 (taste 69% / price 61% drivers) + Momentum Works 2022 (order value) + reporting. Segment split indicative, not a measured distribution.
The economics: the cup isn't the problem — the rent and the royalty are
The internet sells bubble tea as easy money because it fixates on the cup's gross margin. A real outlet is ~S$80–150k all-in independent, ~S$250–460k+ for a prime-mall franchise. The box is small, but the rent-per-square-foot is among the worst in Singapore, and a franchise skims 4.5–6% of revenue in royalties before you keep a cent — plus marketing and supply fees on top. On the macro: tenants reported rent increases of 20–49% in 2025 (MTI), wages are rising under the Food Services Progressive Wage Model (floor to S$2,500 by 2028), and GST kicks in at S$1M turnover.
All-in cost to open a bubble tea outlet
Independent heartland kiosk at the low end; established international franchise in a prime mall at the high end. Fit-out, deposit, equipment, franchise fee and several months of working capital. The smaller footprint saves less than it looks once mall rent-per-sqft is in.
Source: Seedly / operator reports, 2019–2026 (indicative; franchise totals e.g. Gong Cha ~S$244k–462k per outlet, single-source)
Independent = 0%. Typical bubble-tea franchise royalty ~4.5–6% of revenue, plus marketing (~2%) and supply markups — a permanent skim off a thin net line.
Seedly franchise breakdowns, 2019 (KOI ~6%; Gong Cha ~4.5%/mo) — single-source, treat as indicative
Where the bubble tea dollar goes (franchise outlet)
A typical SG franchised bubble-tea kiosk, modelled per S$100 of sales. The cup is cheap to make — but rent, wages and the franchise royalty stack up, and what's left for the owner is thin and fragile.
- COGS (tea, tapioca, powder, cup)−28%72% left
Genuinely low — toppings are near-pure margin · SG operator benchmarks / SGAI calc
- Labour + CPF−24%48% left
Small crew, but PWM floor rising to S$2,500 by 2028 · MOM PWM
- Rent + occupancy−20%28% left
Tiny box, brutal psf; tenants reported +20–49% in 2025 · MTI PQ, Nov 2025 / operator reports
- Franchise royalty + ad fee−7%21% left
~4.5–6% royalty + ~2% marketing; 0% if independent · Seedly 2019 (single-source)
- Other fixed (utilities, licences, POS)−9%12% left
What the owner actually keeps
Verdict: A healthy 12% — there is real margin of safety here.
Illustrative model on SG bubble-tea benchmarks (2019–2026); SG-specific net margins are scarce and the well-run, well-sited owner-operated outlet does better (some report 20%+), while a poorly-sited one can lose money indefinitely. An independent (no royalty) keeps the ~7% the franchise skims. Not financial advice.
Bubble tea & dessert shop
Would a value investor own the average operator here?
A value investor would not want the average bubble tea shop — it's a price-taker in a commodity category, on a rent-and-hype treadmill, where the brand is often rented and a sugar-regulation headwind is rising.
With ~60 near-identical brands and Mixue holding a cheap floor, the average operator sets price against the shop next door, not against a captive customer.
Buffett, FCIC 2010 — pricing power is “the single most important decision”
Recipes are copyable; hype decays (Tiger Sugar). A franchise name is a rented moat that can be switched off — Gong Cha went dark in a day.
Buffett 2007 — an enduring moat protects returns on capital
High gross margin, but rent + royalty + footfall risk pull ROIC back toward the cost of capital.
Buffett 1979 — a high earnings rate on capital, unleveraged
Lighter than a full café (smaller box), but fit-out, equipment, deposit and the franchise fee still bite.
Buffett 2007 — the worst business needs much capital, earns little
Habitual and youth-led, but discretionary, trend-cycle-exposed and pressured by the sugar turn.
Graham, Security Analysis Ch.2 — inherent stability is qualitative
Franchisees rent demand; the agreement can end and the brand can vanish (Gong Cha, 2025).
Buffett 1991 — a franchise must not be “subject to” forces it cannot control
A genuine difference customers can taste or trust — a low-sugar Grade-A position, halal reach, a heartland rent advantage, or an owned brand — not another me-too cup competing on hype.
Assessment uses the value-investing lens on SG bubble-tea unit economics (2019–2026). A lens on economic quality, not a verdict on an owner-operated livelihood.
Model your own survival line — drag the sliders (set royalty to “Independent” for the no-franchise case):
Break-even
227
cups a day, just to break even
Monthly net
−S$620
Rent / revenue
27%
danger
Verdict: At these numbers you'd join the 82% of young F&B closures that never turned a profit.
Illustrative model on SG bubble-tea benchmarks (2023–2026). Per-cup gross margin is genuinely high, but rent and franchise royalty decide the outcome. Set royalty to “Independent” to see the no-franchise case. Rent gauge: ≤12% healthy, 12–20% tight, >20% danger. Not financial advice.
Nutri-Grade: the rule that rewrote the cup
Bubble tea is the poster child of Singapore's sugar policy. Since 30 December 2023, freshly-prepared drinks — bubble tea included — fall under Nutri-Grade, and it reshapes the menu, the marketing and the product itself.
What the rule actually requires
- Grades A–D based on sugar and saturated fat per 100ml (A green, D red).
- Drinks graded C or D must display the Nutri-Grade mark on the menu next to the listing; A and B are optional.
- Grade D drinks cannot be advertised at all, across every medium. (Grade C can still be advertised — a point many guides get wrong.)
- Toppings must declare their sugar content on menus and signage — this is what catches the pearls, jelly and brown-sugar add-ons.
Why it bites bubble tea hardest
A full-sugar brown-sugar boba can carry roughly 18–20 teaspoonsEstimate / modelled, not a measured figureMount Alvernia Hospital / NUHS, 2019A regular bubble milk tea with pearls at 100% sugar ≈ ~20 tsp; brown-sugar boba ≈ ~18.5 tsp. Hospital education material — directionally solid; exact figure varies by recipe. of sugar — past an adult's entire recommended daily intake in one cup. That is why most chains now default to lower sugar and lean on the 0% / 30% / 50% / 100% customisation. The compliance overhead is real, but so is the opening: a brand built genuinely Grade-A can advertise freely while sweeter rivals can't, and can carry the Healthier Choice halo into the bargain.
How to actually open one (in the order that matters)
A bubble tea kiosk is a Food Shop, so the licensing stack is the same as a café's. As ever, the cheap SFA licence is the last easy step — the premises are the bottleneck.
The approval stack, sequenced
- ACRA — register the business.
- URA change-of-use (~S$500) — the unit must be approved for F&B use. Gate 1.
- SCDF fire safety — plans via a Qualified Person → Fire Safety Certificate. Gate 2.
- WSQ Food Safety Course + a Food Hygiene Officer.
- SFA Food Shop Licence — S$195/yr, in-principle approval ~7 working days (the cheap, late step).
- Then concept-specific: Nutri-Grade menu marks + topping sugar declarations; MUIS halal if you want the wider market; signboard.
Two things older guides get wrong
First, from 19 Jan 2026 the SFA SAFE framework replaced the old A/B/C/D hygiene grades (now A/B/C on track record, with a “NEW” tag for shops under a year) — a new shop cannot display an “A” until it builds a record. Second, there is no ban or mandatory charge on disposable cups in Singapore today (a widely-shared “2024 ban” claim is false) — but sustainability is a live policy direction worth designing for.
Where a new bubble tea shop actually wins
Another me-too cup competing on hype loses — the category already has ~60 of those. The openings are where a real, defensible difference meets Singapore's actual buyer.
Grade-A as the brand
Most chains treat Nutri-Grade as a constraint to dodge. Build genuinely low-sugar, real-ingredient drinks that default to A — advertise freely while sweeter rivals can’t, and ride the War-on-Diabetes tailwind.
Halal reach (MUIS)
Many premium chains skip halal certification, leaving a large addressable segment on the table. MUIS cert is a concrete, defensible way to widen the market most players ignore (verify your supply chain qualifies).
Heartland rent arbitrage
Rent is the single biggest margin lever: a heartland kiosk can run a fraction of a trophy mall’s rent. A loyal repeat base near where people live beats a high-footfall unit that needs huge volume just to clear the lease.
Own the brand, not rent it
A franchise can be switched off (Gong Cha, 2025). An owned concept — even a small one — keeps the demand you build. If you do franchise, negotiate the term, royalty, supply lock-in and exit clause hard.
Drink + dessert hybrid
Pure tea is commoditised. Pairing a signature dessert or a defensible format raises ticket size and differentiation on the same Food Shop licence and kitchen.
The honest AI edge
Demand forecasting, inventory and labour scheduling trim waste off thin lines, and queue/loyalty data sharpens the menu. Real, but it won’t rescue a bad lease or a me-too cup. Use it to sharpen a good business, not save a doomed one.
Questions founders ask
How much does it cost to open a bubble tea shop in Singapore?
Realistically S$80,000–150,000+ for an independent heartland kiosk, and S$250,000–460,000+ for an international franchise in a prime mall (fit-out, deposit, equipment, franchise fee and working capital). The bubble tea box is small, but mall rent-per-square-foot is among the highest in Singapore, so the headline saving over a café is smaller than it looks.
Is the bubble tea business profitable in Singapore?
The gross margin on a cup is genuinely high — ingredients (tea, tapioca, powder) cost well under S$2 on a S$5 drink. But the net depends entirely on rent and cups-per-day, and of F&B businesses that closed within five years, 82% had never recorded a profit (MTI, Nov 2025). A well-sited, owner-run outlet can do 20%+ margins; a poorly-sited one can lose money indefinitely.
Is a bubble tea franchise worth it in Singapore?
A franchise buys a known brand and supply chain, but you pay a 4.5–6% royalty plus marketing and supply fees, and the brand is not yours — it can be pulled. Gong Cha closed all its Singapore outlets overnight on 1 October 2025 when the franchise agreement ended; LiHO exists only because its operator lost the Gong Cha franchise in 2017 and rebranded ~80 stores. Read the term, the royalty, the supply lock-in and the exit clause carefully.
What licences do I need to open a bubble tea shop in Singapore?
A bubble tea kiosk is a Food Shop, so the stack is: ACRA registration, URA change-of-use, an SCDF Fire Safety Certificate, a WSQ Food Safety Course plus a Food Hygiene Officer, and the SFA Food Shop Licence (S$195/year). The real bottleneck is the premises chain, not the cheap licence — and from 19 Jan 2026 SFA replaced the old A/B/C/D hygiene grades with the SAFE framework, where a new shop opens carrying a "NEW" tag.
How does Nutri-Grade affect a bubble tea shop in Singapore?
Since 30 December 2023, freshly-prepared drinks (including bubble tea) must carry a Nutri-Grade A–D mark when graded C or D; Grade D drinks cannot be advertised at all; and any topping must declare its sugar content on the menu. A full-sugar brown-sugar boba can carry around 18–20 teaspoons of sugar — well past an adult day’s recommended intake — so most chains now default to lower sugar or push the 0%/30%/50% customisation. A genuinely low-sugar, Grade-A brand can turn the rule into a selling point rather than a constraint.
Want a report like this for your company or sector?
Tell us what you're trying to figure out. We'll build you a Deep-Context report — the same depth, focused on your business.
About this report. Built with SGAI's Deep-Context Engine — human-directed, AI-accelerated. Figures draw on SingStat, an MTI parliamentary reply (Nov 2025), HPB Nutri-Grade, SFA, NLB Singapore Infopedia, MOM, IRAS, HKEX filings and SG market reporting (2017–2026). Bubble-tea-specific net margins are scarce and franchise figures are largely single-source — we flag them, range them, and separate Singapore data from global. Outlet counts are indicative; Singapore keeps no official bubble-tea census. The MOAT Score is a transparent SGAI judgement on economic quality, not a verdict on an owner-operated livelihood. Verify fees and regulatory steps with each agency before acting.
We're not a grant office — we build AI and intelligence for Singapore SMEs. Talk to us.