Tap water is not drunk in China. Buying a bottle is a reflex, not a decision. Add the density of retail (in a big city you pass three shops in a hundred meters) and you get a country that drinks packaged beverages the way other countries drink from the sink. The numbers follow. Chinese beverage retail sales reached 3.295 trillion RMB in 2025, with a compound growth of about 7.8% a year since 2019, according to Qianzhan Industry Research Institute (April 2026). Gross margins are high. That is exactly why the category is now the hardest one to enter in China.
Written by Olivier Verot, founder of Gentlemen Marketing Agency, based in Shanghai since 2012. I have run China launches for imported waters, juices, sugar-free teas and functional drinks, and I have also seen several of them pulled from shelves within a year.
The most crowded shelf in China, and the fastest rotating one
Incomes rose, and with them the willingness to pay for a drink made on the spot rather than a bottle off the shelf. That shift created the tea shop boom. Over the past thirty years milk tea (bubble tea) became the most consumed drink in the country. It is also the most contested one, because of its sugar load and its composition, the cream and the starch pearls. It is still everywhere. But a large share of consumers now say they are cutting back, and that single sentence explains most of what happened to the category since.
Here is the part foreign brands underestimate. The Chinese beverage market does not just have more competitors. It rotates faster than any market you have worked in. A flavor is tested, scaled and buried in the space of two seasons.
A drink is tested fast and dies fast
Look at how the new product mix moved between 2024 and 2025 in the Chinese water and beverage aisle, per Qianzhan’s 2026 industry review:
- Ambient ready-to-drink juice: from 13 new SKUs to 26. The hottest category of the year.
- Sugar-free ready-to-drink tea: from 22 new SKUs down to 13.
- Sports drinks: from 16 down to 8.
- Sparkling water and herbal tea: zero new SKUs launched in 2025.
Sparkling water was the darling of 2020. Five years later, Chinese groups stopped launching into it. Not because the drink is bad, because the shelf is saturated and the margin got squeezed. Domestic players treat SKUs like ad campaigns: launch, measure sell-through in eight to twelve weeks, kill or scale. A European brand that plans an eighteen-month launch runway is competing against companies that have already run three cycles in that time.
The other side of that coin is that entry is slowing. Only 196 new beverage companies were registered in China in Q1 2026, far below the peak years. The easy money left. What remains is a market where you win on execution, not on novelty.
Below are the five things I tell every beverage client before they spend a yuan. They replace the generic advice you will find in most guides.
1. Sugar-free and health are the ticket to entry, not your argument
More and more Chinese consumers look for fewer calories and more benefit per bottle. That part of the old advice still holds. What changed is the competitive value of saying it. Sugar-free tea alone is projected at 135.2 billion RMB in 2026, crossing 50% of the whole tea beverage category for the first time, according to the 2026 China Tea Beverage Industry Report by Qince Consumer Research, relayed by HKTDC in June 2026.
When half a category is sugar-free, “sugar-free” is not a claim. It is a spec. Nongfu Spring’s Oriental Leaf passed 8 billion RMB in sales in the first half of 2025 by itself. You are not going to out-zero that.
So put 0 sugar, 0 fat, 0 calories on the front label because its absence disqualifies you, then build the actual argument one level up. What works in 2026: a named functional ingredient with a reason to exist (electrolytes for hot yoga, GABA for sleep, chicory fiber for digestion), a single-origin raw material you can document, or a taste profile Chinese factories cannot copy in six weeks. Functional tea with probiotics or collagen is growing around 35% a year. That is where the premium sits, not in the zero itself.
One practical note. Health claims in China are regulated tightly. “Helps digestion” on a general food label will get your listing pulled and your Douyin ads rejected. Write claims that survive a compliance review before you write the creative brief.
2. Your packaging is your media buy on Xiaohongshu
Take the appearance of the product seriously. The shelf is already full, so the eye decides first. Chinese consumers are sensitive to design and a good one makes a real difference. You can go clean and minimal, which stands out against packaging loaded with colors, mascots and information. Or you can put a message on the label that people want to be seen holding. The difference you create against the other brands is a real advantage.
That was true in 2018. In 2026 there is a mechanism behind it. Xiaohongshu is now north of 400 million monthly users and it works as a search engine for products. Somebody in Hangzhou types 无糖气泡水 推荐 (sugar-free sparkling water recommendation) and reads twenty posts before buying anything. Those posts are photographs. If your bottle looks good on a café table in daylight, users produce that content for free and your paid content costs less to make. If it does not, you are buying every impression.
Design for the platform, concretely:
- The brand name must be readable in a 3:4 vertical crop at thumbnail size. Most European labels are horizontal and disappear.
- One color that is yours. Users search by look before they search by name.
- A Chinese name on the bottle, not only on the back sticker. People cannot search what they cannot type.
- A shape that survives being held in one hand in a selfie. Tall and thin photographs better than short and wide.
Then seed it properly. The engine on Xiaohongshu is KOC volume, not one big KOL. A hundred small accounts posting real usage build the search results a buyer sees. One celebrity post builds a spike and nothing else. Our Xiaohongshu strategy guide details how the seeding and the search layer fit together.
3. Seasonality and limited editions keep you alive between launches
A beverage brand in China does not have a launch. It has a calendar. Chinese players ship a seasonal flavor, a festival pack or a co-branded edition every six to ten weeks, and that rhythm is what keeps them in the algorithm and on the convenience store shelf. A brand that shows up once a year with a “new campaign” gets rotated out.
The mechanism is simple. Retail buyers reallocate shelf facings quarterly and reward brands that give them a reason to re-merchandise. Platform algorithms reward new product pages with fresh traffic. Consumers who bought once need a reason to come back that is not a discount, because a discount trains them to wait for the next one.
What I recommend to a foreign brand with a small portfolio: keep one hero SKU permanent and run two to three limited runs a year around real Chinese moments. Osmanthus in autumn. A Chinese New Year gift pack in a red carton that survives being given to a colleague. A summer format for the June to August peak, which is when half the category’s volume is sold. Limited runs also let you test flavors cheaply. If the osmanthus edition sells out in three weeks, you have your next permanent SKU and you did not pay for a full production line to find out. Co-branding with a Chinese partner is the cheapest version of this, since the partner brings an audience you do not have to buy.
4. Convenience stores and instant retail own the impulse
Beverages are an impulse purchase. Which means distribution is not a logistics topic, it is a marketing topic. Two things changed in the last two years and both matter to your plan.
First, convenience stores are under pressure. The China Chain Store and Franchise Association’s 2026 report, covering 56 chains and about 159,000 stores, found average daily revenue per store fell to 4,453 RMB in 2025, down 3.9% year on year, while footfall per store dropped 8.7%. Net store openings slowed to 7,572 in 2025, against 9,570 the year before. Huxiu covered the findings in detail in May 2026. Buyers in these chains are therefore ruthless about facings. A slow SKU is delisted in one review cycle.
Second, the traffic went to instant retail. Meituan Instashopping, Taobao Shangou and JD Daojia now deliver a cold drink in thirty minutes, and the Ministry of Commerce research institute expects instant retail to pass 1 trillion RMB in 2026, on more than 60 billion orders growing above 25% a year. Drinks and snacks are among the categories driving that growth.
What that means for a foreign brand, in order: get into instant retail inventory early, because listing there is cheaper and faster than negotiating national convenience chain distribution, and because it gives you sales data by district that you can then show to a distributor. Cold chain matters more than pretty displays. A drink that is not cold is not bought. And do not neglect the machines: vending in China is a real channel with real constraints, and it works for single-serve formats near offices and gyms. If you are going the classic route, finding the right distributor is a marketing exercise before it is a commercial one, because good distributors pick brands that already have demand signals.
5. Price against Chinese players, not against your home market
This is where most European and American beverage brands lose. Mainstream sugar-free tea sits at 5 to 6 RMB. Bottled water starts around 2 RMB. An imported drink landing at 15 to 20 RMB is not slightly more expensive, it is in another category, and the consumer will treat it as an occasional gift rather than a daily habit. Habit is where the volume is.
You have three honest options. Sell at 8 to 12 RMB with a documented reason (imported water source, real fruit content, a functional ingredient), which is the cold-brew premium band and it does work. Or localize production and compete at 5 to 6 RMB, accepting that your “imported” story disappears. Or stay high, accept a small volume, and build the brand through gifting, hotels, cafés and specialty retail rather than mass distribution.
What does not work is the fourth option most brands pick by accident: import at high cost, price high, and then fund permanent discounts on Tmall and Douyin to move stock. You lose margin and you teach the market your real price at the same time.
What still works from the old playbook
Traceability and origin
Play on your origin, especially as a foreign brand. Imported still reads as better quality for a large part of the market. After the food scandals of the past two decades, Chinese buyers care a lot about where a product comes from, and transparent information reassures them. Nongfu Spring built a whole advertising history on naming its water sources, and it is a domestic brand. If a Chinese company thinks origin is worth that much airtime, yours is worth documenting properly: the spring, the orchard, the farm, with photos and a date, not a flag on the label.
Product testing, in person and on video
If you can, organize free tasting in a public spot. A consumer who tried your drink and liked it is far more likely to buy, because he already participated. The presence of other people around pushes him further, since what others think carries real weight in China. Group consumption is visible in this market: shoppers follow the purchases of the people next to them. If the tasting attracts a crowd, a few buyers are enough to start the rest.
The 2026 version of that is the same psychology on video. A Douyin live session where the host drinks the product on camera and shows the comment count is a public tasting with a bigger room. Same trigger, lower cost per person reached.
Reputation, and the AI layer on top of it
Word of mouth is very powerful in China. Chinese consumers trust each other more than they trust the brand. Good products and bad ones get classified fast because information moves fast, and reversing a bad reputation is very hard once it is set.
What is new: a growing share of buyers now asks an AI assistant. DeepSeek, Doubao and Baidu’s AI answers are being used as a first filter for “which sugar-free tea is actually good”. Those models answer from what exists in Chinese online text: Xiaohongshu posts, Zhihu threads, Baidu Baike, media articles. If your brand has no Chinese-language content with a consistent Chinese name, the model will not mention you, and it may repeat a competitor’s claim about your category. Getting cited there is not a paid placement, it is a content problem: publish structured, factual Chinese content about your product, your origin and your ingredients, and make sure your Chinese brand name is used identically everywhere.
Online presence and e-commerce
Chinese consumers are very active online to get and share information. It is faster and more reliable for them than asking a shop employee. An online presence is not optional, and seeing the brand repeatedly is what builds the confidence to try it. Several sales channels should run in parallel: Tmall or JD for credibility, Douyin for discovery and volume, Xiaohongshu for the search layer, and WeChat for what happens after the first purchase. That last one is the piece most brands skip. A WeChat group of 2,000 buyers who reorder every month costs almost nothing to run and is the only asset in China that no platform can take away from you.
Case study: Sinead, an Irish beverage brand
Sinead runs a family-owned Irish drinks company. Spring water base, botanical infusions, three flavors, strong in Ireland and the UK. She came to us after eleven months in China with a distributor in Guangdong.
The situation was bad and clear. About 40,000 bottles landed, roughly 9,000 sold in eleven months. Retail price 19 RMB against 6 RMB for the domestic equivalent. The distributor had placed the product in premium supermarkets in Shenzhen and Guangzhou, where it sat next to imported juices and did not move. Her Tmall flagship store had 300 followers.
What had been tried: a Weibo account posting photos of the Irish countryside, two paid KOL videos on Douyin at 60,000 RMB total, and a price promotion that moved stock for one week and nothing after. The Weibo account had 1,100 followers, most of them inactive. The KOL videos produced 400,000 views and 60 orders.
What we changed, and why. First the price story. We stopped defending 19 RMB and repositioned the brand at 12 RMB in a smaller 330ml format, which put it in the cold-brew premium band where consumers already accept paying more. The 500ml stayed for gifting and horeca. Second, the search layer. We put 180 KOC posts on Xiaohongshu over four months, all built around one question people actually type: which bottled drink to bring to a Pilates class. Not around Ireland. Ireland became a detail inside the post, not the headline. Third, we listed on Meituan Instashopping in Shanghai and Hangzhou before touching new physical retail, because it gave district-level sell-through data in six weeks instead of six months. Fourth, a WeChat mini-program with a repurchase discount for anyone who scanned the bottle cap.
Why it worked: the Xiaohongshu content created search demand, instant retail satisfied it within thirty minutes while the intent was still there, and the WeChat layer turned first purchases into a base she could talk to for free. The origin story still did work, but as proof, not as the hook.
Result after ten months: about 71,000 bottles sold over the period, monthly reorder rate at 22% on the WeChat base, and two regional distributors approaching her instead of the reverse, because they could see the instant retail numbers. She is still not profitable in China. She has a business that grows, which at this stage is the right target.
If you sell a specific category
The five points above are the marketing principles that apply to any drink. Each category then has its own channel rules, price bands and regulations. We treat them separately:
- Soft drinks and soda in China
- Selling tea in China, and the tea and coffee shakeout
- The juice market in China
- Sports and energy drinks
- Bottled water in China
FAQ
How much budget do I need to launch a beverage in China?
For a serious first year, plan 40,000 to 80,000 USD in marketing on top of your product and logistics costs. That covers Chinese naming and label compliance, a Xiaohongshu KOC program of a few hundred posts, one e-commerce storefront, and content production. Below 25,000 USD you can test, but you cannot build search presence, and without search presence a beverage brand in China stays invisible. Above 150,000 USD only makes sense once you have proven sell-through in at least two cities.
Do I need a Chinese brand name?
Yes, and you should register it as a trademark before you use it publicly. Consumers cannot search a name they cannot type, distributors will invent one for you if you do not, and once a squatter registers it you are renting your own name. Pick a name that sounds close to the original, means something positive, and is three characters or fewer. Test it with real Chinese consumers, not with your translator.
Should I start with e-commerce or with physical distribution?
E-commerce and instant retail first, physical distribution second. Online gives you sell-through data in weeks and costs a fraction of a distributor agreement. A Chinese distributor will ask what your online numbers look like before committing shelf space, so you need those numbers anyway. The one exception is horeca: hotels, gyms and cafés can be approached in parallel because they buy on the product, not on brand awareness.
How long before my beverage brand is profitable in China?
Two to three years for most imported drinks, and that is with a working strategy. The first year covers registration, label compliance, naming and the first content base. The second year is where reorder rates decide whether you have a business. Anyone promising profitability in twelve months is either selling you something or planning to discount their way to volume, which is not the same thing as profit.
Is sugar-free really mandatory?
For anything sold as a daily drink, effectively yes. Sugar-free tea alone is projected past 135 billion RMB in 2026 and more than half the tea beverage category. Sugar still sells in indulgence and gifting occasions, in tea shops, in soda for young consumers, and in the lower-tier city market. But if your positioning includes the words healthy, natural or premium, a sugar figure on the label kills the argument.
Why do Chinese brands launch and kill products so fast?
Because their cost of trying is low. Contract manufacturers can produce a small run in weeks, Douyin and instant retail give sales feedback in days, and shelf space is negotiated quarterly. So the rational move is to launch many variants and keep the two that work. Foreign brands with central approval cycles and annual planning cannot match that speed, which is why they should compete on a defensible product rather than on flavor novelty.
Does the imported label still carry a premium?
Less than it did, and it varies by category. In water, dairy and infant products, imported still commands trust. In tea, coffee and functional drinks, Chinese brands are now seen as equal or better, and several of them out-innovate foreign competitors. Treat imported origin as evidence supporting a specific claim, water source, fruit variety, brewing method, not as the claim itself.
Working with us
Gentlemen Marketing Agency has run China launches for imported waters, teas, juices and functional drinks since 2012, from Chinese naming and label compliance through Xiaohongshu seeding, Douyin selling and instant retail listing. We work on the whole chain because in beverages a good campaign on a badly priced product changes nothing. We are based in Shanghai and we know which distributors actually move volume in this category.
If you want to market your drink in China, start by getting the price band and the search layer right. Contact us and we will look at your product, your target price and your category, and tell you honestly whether the numbers work before you ship a container.
Read also:
