Bubble tea, also known as pearl milk tea, came out of Taiwan in the early 1980s, invented in Tainan and Taichung. It is a milk tea base with tapioca balls inside, the “pearls”. For twenty years it stayed a snack. Then it crossed to mainland China and turned into something else entirely: an industry called 新茶饮, “new tea drinks”, worth around 360 billion RMB in 2025 and forecast to reach 650 billion RMB by 2029. Six chains are now listed on public markets. One of them runs close to 60,000 shops. If you sell tea, syrup, powder, toppings or brewing equipment, this is no longer a beverage trend. It is a supply chain with buyers who sign eight-figure contracts.
I am Olivier Verot, founder of Gentlemen Marketing Agency. I have been based in Shanghai since 2012 and my team has worked with dairy, syrup, fruit-puree and tea-leaf suppliers trying to get listed by Chinese chain buyers. I wrote this from procurement conversations, not from a press release. Find me on LinkedIn.
The bubble tea craze in China
There are many varieties of the drink and a wide range of flavours. You can add toppings: tapioca, fruit jelly, pudding, konjac, cheese foam, oat flakes, ice cream. The combinations are close to endless, which is exactly what keeps Chinese consumers coming back. The classic is black pearl milk tea. Matcha and green tea milk follow. Since 2023 the fastest-growing base is not milk tea at all, it is pure brewed tea served cold with almost no sugar.
Chinese consumers drink far more freshly made tea than coffee. That gap has held for a decade. What changed is the machine behind it. In 2018 a foreign brand could open a few shops and ride the novelty. In 2026 the market is run by industrial franchise groups with their own farms, factories and cold chain. You do not compete with them on a street corner. You either supply them, or you sell to their customers on a different shelf.
Who actually owns the market in 2026
The 2025 full-year results, published in spring 2026, made the hierarchy very clear. Scale and supply chain won. Premium positioning did not. Here is where the six listed players stand, based on their 2025 annual reports as compiled by 36Kr.
| Brand | Stores (end 2025) | 2025 revenue | 2025 net result | Listing |
|---|---|---|---|---|
| Mixue 蜜雪冰城 | 59,823 (+13,344) | RMB 33.56 bn (+35.2%) | RMB 5.93 bn (+33.1%) | HKEX, March 2025 |
| Guming 古茗 | 13,554 (+3,640) | RMB 12.91 bn (+46.9%) | RMB 3.12 bn (+108.6%) | HKEX, February 2025 |
| Chagee 霸王茶姬 | +1,013 net additions | RMB 12.91 bn (+4.1%) | RMB 1.19 bn (-52.4%) | Nasdaq, April 2025 |
| Auntea Jenny 沪上阿姨 | 11,449 (+2,273) | around RMB 4.5 bn | profitable | HKEX, 2025 |
| Cha Panda 茶百道 | 8,621 (+226) | RMB 5.40 bn (+9.7%) | RMB 805 m (+70.5%) | HKEX, 2024 |
| Nayuki 奈雪的茶 | 1,646 (-165) | RMB 4.33 bn (-12%) | -RMB 243 m (loss) | HKEX, 2021 |
Read that table twice. Mixue sells the cheapest cup in China, often 4 to 8 RMB, and it earns more than the other five combined. Nayuki sold the most expensive cup, built beautiful flagship stores, and lost money while closing 165 shops. Chagee grew revenue 4% and saw profit cut in half. The lesson Chinese operators took from 2025 is blunt: unit economics beat brand theatre.
The 2025 IPO wave
2025 was the year the sector went public. Guming listed in Hong Kong in February. Mixue followed on 3 March and raised around HKD 4 billion, closing its first day above the 100 billion HKD mark. Chagee listed on Nasdaq on 17 April under the ticker CHA, the first Chinese tea chain on a US exchange. Auntea Jenny listed in Hong Kong later in the year.
Why this matters for a supplier: listed companies publish their cost structure, their supplier concentration and their expansion plans. Before you pitch a Chinese chain, you can now read exactly how much they spend on raw materials, what gross margin they defend, and which product lines they want to add next year. That information used to cost a consultant six weeks of work. It is now in a prospectus.
The delivery subsidy war, and the slide down to smaller cities
Through 2025 the platforms burned close to 100 billion RMB fighting over food delivery. Meituan, Alibaba and JD pushed free-drink vouchers and heavy discounts. Tea shops were the weapon of choice, because a milk tea is cheap to give away and gets people to open the app. Guming hit 2 to 3 million orders a day during the July “zero yuan” campaigns. The catch: net take per order fell to 4 or 5 RMB. Nayuki’s average ticket dropped from 26.7 RMB to 24.4 RMB.
Two things came out of that year. First, price expectations reset downward. The 30 RMB cup that was normal in 2021 is a hard sell in 2026. Second, growth moved out of Shanghai and Beijing. Third-tier cities and below now account for about half of national GMV and 53% of new store openings, according to the 2026 New Tea Drinks White Paper by CIC. Guming was opening more than 500 stores a month at one point, almost all in county-level markets.
The premium wave came and went in about four years. Nayuki and early Heytea proved that Chinese consumers would pay 30 RMB for cheese-foam tea in a designer store. Then the economy cooled, delivery subsidies trained everyone to wait for a discount, and Mixue took the country from the bottom. Premiumisation did not disappear, it just got squeezed into a narrow band of city-centre stores.
Health is now the product, not the label
Ask any Chinese chain buyer what their R&D brief looks like and sugar comes up first. In store, more than 60% of orders now specify a reduced-sugar or no-sugar option. On the packaged side, unsweetened pure tea grew over 70% a year and reached 35% of the Chinese tea-drink category in 2025, on track to pass half in 2026. Functional drinks, meaning teas with an added claim around sleep, digestion or recovery after a late night, hold roughly 15% of the category.
Practical consequences for a foreign supplier:
- Non-dairy creamer is dying. Chains have spent three years replacing vegetable-fat powder with fresh milk and real cream, because Xiaohongshu users started reading ingredient lists out loud on camera. If your product still contains hydrogenated fat, it will fail the buyer’s first screen.
- Origin claims sell. Single-origin tea leaf, named production area, harvest date. Chagee built its entire menu story on named tea cultivars. Buyers want documentation they can print on a cup.
- Sweetener systems, not sugar. Erythritol, allulose, stevia blends that survive being shaken with ice. That is a technical sale, not a commodity sale.
- Residue standards are the gate. Organic certification and low pesticide residue have moved from nice-to-have to mandatory on global sourcing lists. Chains have been burned by food safety stories and will not risk it.
These chains are now exporting, and that changes your negotiation
Mixue passed 4,700 shops outside mainland China across 14 markets and opened its first South American store in São Paulo in April 2026. Heytea opened in Toronto in February. Chagee reported overseas GMV up 84.6% in the fourth quarter and is pushing into South Korea, the US and Europe. Southeast Asia is getting crowded, so the brands are going further, as Caixin Global reported in March 2026. China Daily covered the same push from the brand side.
This is the part most European suppliers miss. If you get listed by a Chinese chain, you are not selling into China only. You are selling into their whole international network, and they will ask you to support them in Indonesia, Vietnam and Brazil at Chinese prices. Mixue already built a factory in Vietnam. Heytea set up warehouses in the UK and the US. Decide early whether you can serve that footprint, because the buyer will ask in the second meeting.
Why beverages are so popular in China
Pearl milk tea first attracted young female students. Those students grew up. The habit stayed and spread to working professionals and to men, which is why you now see 35-year-old engineers queuing at a Chagee counter at 3pm.
Western countries built a coffee lifestyle. Chinese people drink before, during and after a meal, and they always have. Some like the drink-and-snack combination that chewy toppings give. Others simply like the taste and the number of choices. The cup is also social currency: it gets photographed, shared, and used as a small daily reward.
Coffee has not lost ground either. Both categories are converging. Guming fitted more than 12,000 stores with coffee machines. Mixue’s Lucky Coffee passed 10,000 outlets and moved into Malaysia and Thailand. Auntea Jenny launched its own coffee sub-brand. If you supply syrup or powder, the same buyer now handles both menus. We covered that convergence in our piece on China’s tea and coffee shakeout.
Xiaohongshu, the map of Chinese consumer choice
Bubble tea is not as simple to order as coffee. You cannot just say “a tea please”. Sugar level, ice level, base, topping, cup size. So the internet filled up with posts explaining how to build your drink. The main platform for that is Xiaohongshu.
Xiaohongshu, or Little Red Book, is a social platform where users post advice, product reviews and photos. The community is very active. More importantly for a B2B supplier, it has become a search engine. Chinese buyers, including procurement staff, type product questions into Xiaohongshu before they type them into Baidu. A tea-leaf exporter with twelve well-structured Xiaohongshu notes will get found by chain R&D teams. A tea-leaf exporter with a PDF brochure will not.
Add to that the generative search shift. Chinese professionals now ask DeepSeek and Doubao questions like “which European suppliers of fruit puree serve tea chains”. The answer is assembled from Chinese-language pages the model can read. If your company only exists in English on a .com domain, you are invisible in that answer. Getting cited means publishing Chinese technical content, with specifications, certifications and named use cases, on domains Chinese crawlers actually index.
Where a foreign brand actually fits
I get this call once a month. A European or American founder wants to open bubble tea shops in China. My answer has not changed in five years: do not. You will fight Mixue on price, Guming on density and Chagee on brand, with none of their cold chain. There are three doors that do open.
1. The ingredient
Tea leaf, fruit puree, syrup, milk powder, cream, plant-based milk, toppings, sweetener systems. This is the largest and most realistic door. Chains buy in volume and they renew their menu every few weeks, which means constant demand for new inputs. The buying decision is technical: stability in cold liquid, colour retention, cost per cup, and paperwork. Your entry point is the R&D team, not the marketing team. If you also sell tea as a finished product, our guide on how to sell tea in China covers the consumer side.
2. The equipment
Automatic tea brewers, sealing machines, ice machines, dosing systems, cold chain hardware, water filtration. Chinese chains are automating aggressively because they cannot train 60,000 shops to hand-brew consistently. European and Japanese equipment makers have a genuine technical advantage here. The sales cycle is long, twelve to eighteen months, but one contract covers thousands of stores and the aftermarket parts run for years.
3. The franchise, in reverse
Instead of bringing your concept to China, take a Chinese concept home. Master franchise rights for Mixue, Chagee or Cha Panda in your own country are being negotiated right now. These groups want local partners who know local food regulation and retail leases. It is a different business from exporting, and it is the fastest route to revenue for a food operator with a network.
Case study: a Slovak syrup producer
Zuzana runs a family fruit syrup and puree company in Slovakia, third generation, strong in Central Europe, no Asian sales. She came to us in 2024 wanting to sell to Chinese tea chains. She had already spent about EUR 40,000: two trips to food expos in Shanghai and Guangzhou, a translated brochure, a distributor in Guangdong who placed one container and then went quiet. Twelve months, zero repeat orders.
The expo approach failed for a simple reason. Chain R&D teams do not source at general food fairs, and the distributor she signed had no relationship with any chain, only with small wholesalers. She was also pitching the wrong thing: a fruit syrup range built for European yoghurt and pastry, too sweet and too thick for a cold shaken tea.
What worked took nine months. We built a Chinese-language technical site with per-product specifications: Brix, acidity, behaviour at 4°C, colour stability after four hours, dosage per 500 ml cup, EU organic and pesticide residue certificates as downloadable files. We ran Baidu SEO on procurement keywords, not consumer keywords, and mirrored the same technical content on Xiaohongshu and a WeChat official account so it could be forwarded inside a buyer’s internal chat. Zuzana’s team reformulated two SKUs down to a lower sugar load with a stevia and erythritol blend, because that is what the briefs asked for.
The mechanism that closed it was the WeChat private domain. Every inbound contact went to a sales rep’s personal WeChat, not to a form. Chinese R&D staff move fast and they ask questions at 11pm. Answering in Chinese within ten minutes, with a spec sheet attached, is what separates a supplier from a brochure.
Result after fourteen months: 31 qualified inbound enquiries, 6 sample rounds, 2 signed contracts. One regional chain with about 400 shops, one second-tier national brand for a seasonal limited edition. Combined first-year value around EUR 480,000, with the seasonal contract renewed for a second run. Not a fairy tale, and it took over two years from her first attempt. But it is repeatable, and the technical content keeps generating enquiries without further spend.
How to break into the Chinese tea market
Being visible online is not optional, because Chinese buyers and consumers research heavily before they try anything new. In food, brand awareness carries extra weight: people are careful about what they eat, and China has a memory of past food safety scandals. A supplier nobody can verify is a supplier nobody buys.
Here is the order I recommend to food and beverage clients in 2026:
- Chinese-language technical base. A site on a domain Chinese crawlers reach, with real specifications and certificates. This feeds Baidu, Xiaohongshu search and the generative engines at the same time.
- Xiaohongshu as a search surface. Notes answering the questions buyers and consumers actually type. Not lifestyle photography.
- WeChat private domain. An official account for credibility, plus personal WeChat for every lead. This is where Chinese B2B deals are actually discussed.
- KOC and affiliate on the consumer side, if you also sell a finished product. Small creators with 5,000 to 50,000 followers, paid on performance, beat one expensive celebrity.
- Douyin for anything consumer-facing. Interest-based discovery means people buy your drink without ever having searched for it. See our notes on selling a drinks brand in China.
- AI in customer service. A bilingual assistant on WeChat that answers spec questions instantly and hands over to a human for pricing. It is what makes the ten-minute response time above realistic.
You can also read our Strategic Guide to Export Food Brands in China
FAQ
Is it too late to enter China’s new tea drinks market?
Too late to open shops, yes. The retail layer is consolidated and the top six chains control the traffic. It is not too late upstream. Chains reformulate constantly and add menu items every few weeks, so they are always evaluating new ingredients. The category is still forecast to grow from 360 billion RMB in 2025 to around 650 billion by 2029. Growth of that size needs new suppliers, especially in low-sugar systems, real dairy and origin-certified tea leaf.
How do I get in front of a Chinese chain’s R&D team?
Not at a general food expo. Publish Chinese technical documentation online so their sourcing staff finds you when they search, then use a WeChat account as the contact channel. Specialist trade events like Hotelex and the China Tea Expo work better than broad food fairs. Expect three to six months of contact before a sample request, and treat the sample round as the real pitch. Answer in Chinese, fast, with numbers.
Which is the better target: the giants or the mid-size chains?
Start with mid-size. Mixue makes most of its own inputs through a self-owned production system across seventeen fruit-growing regions and five plants, so its supplier list is short and price pressure is brutal. Chains in the 300 to 3,000 store range still buy externally, decide faster, and pay more per unit for differentiation. Win two of those, then the giants will return your calls with a reference to check.
What does a realistic first-year budget look like?
For an ingredient supplier, plan EUR 60,000 to 120,000 for year one: Chinese technical site, Baidu SEO, Xiaohongshu and WeChat content, plus a Chinese-speaking commercial contact. Add sample production and shipping, and certification translation. That is far less than an expo-led approach and it compounds, because the content keeps generating enquiries after the spend stops. Retail entry, by contrast, starts in the millions and I do not recommend it.
Work with us
GMA is a China digital marketing agency based in Shanghai. For food and beverage suppliers, we build the Chinese-language technical presence that chain buyers and R&D teams find when they search, on Baidu, Xiaohongshu, WeChat and the generative engines. We handle the Chinese-language sales channel, from first enquiry to sample follow-up, so your leads do not die in a form. Contact us if you supply tea, dairy, syrup, toppings or brewing equipment and want to reach Chinese chains.
Read also:
