Korean Coffee Chains Failed Across the Board in China. What Can China's Great-Leap-Forward Coffee Shops Learn From Them?

A textbook case of what goes wrong when investors rush headlong into the Chinese market.

Another large-scale coffee shop expansion is brewing.

Last month Seesaw announced 45 million yuan in Series A funding from Baifu Holdings, a unit of Hony Capital. The founder of another Shanghai specialty coffee brand, Yingji Coffee, said in a recent interview that he is putting together a 30 million yuan Pre-A round. And GREYBOX, which moved into the Kerry Centre in Beijing's Guomao district late last year, is backed by roseonly, which has already closed a 190 million yuan Series C.

The same fever ran through the market five years ago. Back then the protagonists weren't specialty coffee brands wearing the halo of "consumption upgrading," but coffee shops with Korean backing.

"By 2015, 5,000 chain stores across China, then a listing in Hong Kong"; "By 2015, 1,500 chain stores across China, then a listing in Hong Kong"; "By 2015, 1,000 chain stores across China"; "By 2015, a globally famous coffee chain brand."

That is the arc of the Korean coffee brand Caffebene, from the loud promotional campaign it began in 2012 to its slow decline.

Brands like it include Hollys Coffee, Maan Coffee, Zoo Coffee, MangoSix and Tushang Coffee. Nearly all of them entered China around 2012, paired up with Chinese capital, and tried to take the Chinese coffee market by expanding at a furious pace. In 2014 and 2015, their fastest years, these most representative Korean-style chains opened more than 800 stores in China all at once.

Coffee consumption in China's tempting market really has kept growing. Figures from the London-based International Coffee Organization show it rising 15% a year, against a global average of 2%. Starbucks, the most successful player here, has spent the past year or two expanding at the rate of one new store a day, and has pushed into more and more second- and third-tier cities.

There were also more and more women buying coffee in first-tier cities. Business customers are the mainstay of freshly brewed coffee, and Starbucks, with roughly 40% of the market, aimed mainly at them. Korean-style coffee shops had one enormous advantage: Korean television dramas were everywhere at the time.

In 2012, Rooftop Prince (《屋塔房王世子》) made Hollys Coffee famous, and Hollys promptly made a loud entrance into China. Later, when My Love from the Star (《来自星星的你》) became a sensation, Caffebene signed its star, Kim Soo-hyun.

Five years on, Hollys Coffee is being chased by suppliers for unpaid bills, Zoo Coffee has been taken over outright by Chinese investors and is busy "de-Koreanizing," and Caffebene, the most aggressive of the lot, has simply gone silent. Its China headquarters in the Kaitai Building on Beijing's East Fourth Ring Road is empty. Its website still links to subsidiaries in the United States, Japan, the Philippines, Mongolia and Cambodia — the one place you find nothing at all is China.

Its franchisees are still open. The grand plan is plainly gone.

For China's ambitious new crop of specialty coffee brands, the Korean-style chains are worth studying. They are a textbook case of what goes wrong when investors rush headlong into the Chinese market.

They wanted to attract new customers, but they never created new demand

If you have ever been inside a Korean-style coffee shop, you will have noticed it is nothing like a Starbucks. They tend to be huge — several hundred square meters, sometimes more than a thousand — furnished with fabric sofas, floor-to-ceiling windows and cute stuffed animals scattered around, for a relaxed, homey feeling.

So their target customers were couples, groups of friends meeting to talk, and families. The idea was that a bigger, more comfortable room would keep people in the store as long as possible. Maan Coffee founder Shin Ja-sang once said he believed customers who lingered would inevitably "call their friends over," bringing in more customers.

That is the opposite of what Starbucks does with its business clientele. Starbucks stores are small, with few seats, and rather than encourage customers to settle in, it pushes takeaway. Korean-style shops wanted people to stay longer and buy more. So alongside the coffee, tea, cake and sandwiches any coffee shop carries, they added sweeter things — smoothies, ice cream, pancakes, waffles — plus bread and pastries baked on site: an average of 76 items on the menu, against 41 for the business-oriented brands.

Who they really wanted was women. And for a while it worked: the pretty food and the sweet, doll-like design made these photogenic rooms a selfie mecca for young women. But that was when Korean dramas were at their peak. Some Caffebene stores still put out cardboard cutouts of Kim Soo-hyun, an actor Chinese consumers stopped caring about long ago. Korean-style coffee shops carried more symbols of Korean fashion and trend than anything else, and once the stream of hit dramas dried up, the ground they stood on was gone.

The young women drawn in at the start were mostly there to try something new; they were never typical coffee drinkers. And China's core coffee drinkers — white-collar workers — were trained by Starbucks. Those people were unlikely to switch.

Their success at home in Korea owed more to culture. In the 1930s Korea's "tea room culture" flourished, and you could find one in almost every alley. Despite the name, tea rooms mostly sold coffee, and by the 1960s they had become casual places to gather. In the 1980s the old tea rooms took on modern touches and spread to university districts, city centers and residential neighborhoods. By then they had a new name — "coffee shop" — and no longer sold only coffee: black tea, shaved ice, plum tea and more appeared on the menu.

That was the backdrop when Hollys opened its first Korean store in 1998 and became one of the country's earliest chains. Starbucks arrived in Korea in 1999, but by then the Korean coffee market was reasonably mature and the local chains had fairly steady customers.

China's coffee market has grown fast, but instant coffee still dominates it completely — 70%, according to Euromonitor — with bottled ready-to-drink coffee taking almost 28%. That leaves Starbucks, Costa, Pacific Coffee, the whole crowd of Korean-style shops and now the specialty brands popping up everywhere all fighting over a freshly brewed market worth just 2%. And inside that 2%, the core customers are still white-collar workers.

They could not take the existing coffee drinkers, and they could not hold on to the new ones they attracted. In the end Korean-style coffee shops created no new demand, which left them stuck with an awkward question: why would anyone go to a coffee shop to eat dessert?

A Great Leap Forward expansion that overestimated demand

At its peak Caffebene had more than 600 stores in China. Under the banner of "catching up with and overtaking Starbucks," it expanded wildly, setting a record of 200 new stores in a single year. Its website once declared: "Caffebene's plan for the future is clear — by the end of 2014 Caffebene aims to have 1,000 stores in China, and will keep growing at this pace afterward. By then, wherever you are in China, a first-tier city or a third- or fourth-tier town, if you want coffee there will be a Caffebene waiting for you."

Read today, that is faintly ridiculous. More soberly, Caffebene badly overestimated demand. Even in Korea, where a city dweller drinks an average of 140 cups a year — far above China's 20 — Caffebene had opened only a little over 700 stores in the nine-plus years since it was founded.

Starbucks is the more useful comparison. For the ten years after it entered China in 1999, it added only 30 to 50 stores a year. It began accelerating in 2012, the same year the Korean chains poured in, and even then it opened just 204 stores. Caffebene did indeed "catch up with and overtake" Starbucks on speed. For a new brand with nothing behind it, that was reckless.

The Korean chains did not ignore second- and third-tier cities, which everyone believed held enormous potential — and they overestimated demand there too. Not until 2011 did Starbucks open more new stores in second-tier cities than in first-tier ones, and those were concentrated in the more mature markets around Jiangsu, Zhejiang and Shanghai; it opened more than 300 stores in second-tier cities such as Suzhou and Ningbo. Starbucks only began reaching third- and fourth-tier cities in 2015; of the 23 cities it entered for the first time that year, five were third-tier and 18 were fourth-tier.

China is Starbucks's second-largest market in the world. If there were more room, Starbucks would not be "conservative" in it. This is not to say Starbucks takes all. It is to say that the expansion strategy of the most successful coffee brand in China tells you a great deal about the actual state of the market.

Franchising is not the problem. Having no standard product or service is

To expand fast, Caffebene ran on single-store franchises from the beginning. A franchisee paid the entire fee up front in one payment and in return got full rights to run the store. For a 200-square-meter store in Beijing, the franchise fee alone came to about 3 million yuan; with renovation, equipment and everything else, the initial outlay ran roughly 3 million to 5 million yuan.

According to a 2014 report by RET, Development Trends of Korean-Style Coffee Shops in China (《韩系咖啡店中国发展趋势研究报告》), Korean coffee brands expanded mostly through franchises or joint operations. Company-owned stores accounted for 10% of Caffebene's locations and just 4% of Zoo Coffee's.

Starbucks used to run China on regional licenses, authorizing Beijing Mei Da Coffee, Uni-President and Maxim's Caterers to develop the Beijing, Shanghai and Guangzhou markets. It then clawed those rights back. In 2003, Starbucks headquarters raised its stake in the Uni-President joint venture in Shanghai to 50% and its stake in the Maxim's joint venture in Guangzhou to 51%; in 2006 it bought 90% of Beijing Mei Da, which meant that operation went fully company-owned as well.

Costa and Pacific Coffee were also careful. Pacific Coffee used exclusive agents city by city, requiring each to open at least 10 to 15 stores. Costa instead "tied itself" to large companies through regional agents, working with the Yueda Group in the south and the Hualian Group in the north to run its stores. That set a high bar for who could be a partner, but it gave the brand maximum protection.

Franchising is in fact a standard way to scale fast, and no one in the restaurant business handles it better than McDonald's, whose franchise share in its home US market runs as high as 90%. The point is that McDonald's has a standardized system for its products and its service. Even so, McDonald's kept its China stores company-owned for a long time after it arrived.

The Korean chains had no such system. Visiting Caffebene stores for this piece, we found the menu differed from one to the next: some sold pizza, one had fried pork cutlets. This started well before Caffebene's China headquarters went silent. Caffebene supervised its franchisees loosely, and many stores ran their own promotions and group-buying deals. Toward the end, most Caffebene franchises were essentially "managing themselves," with even coffee bean purchasing and food preparation left to each store. Some stores sold rice bowls. The coffee, naturally, could not be relied on either, and "a hundred Caffebene stores, a hundred flavors" became a common verdict from customers. Plenty of people in the industry trace the chains' collapse straight back to the franchise model.

Big stores made siting harder, and thin foot traffic turned it into a spiral

After China's commercial real estate sector took off in 2007, Starbucks and Costa, which grew along with it, had far more room to expand. The Korean chains arrived late and struggled to get good sites; their big-store format made it harder still.

Shopping malls and office towers in the cores of first-tier cities weren't ideal locations for them anyway, because the rent was too high. Lan Yi, CEO of Pacific Coffee, has said that by industry averages a 150-square-meter coffee shop produces the best sales per square meter. For Korean-style shops routinely running several hundred square meters, rents in commercial centers were prohibitive, so they clustered near residential areas, universities and parks. As a result, 69% of their first-tier-city locations were in the suburbs. In second- and third-tier cities they preferred downtown, where 87% of their stores sat.

Caffebene once told prospective franchisees on its website: "For grade-A prime locations, the rent is frighteningly high. In fact, if you run the business well, grade-B and grade-C locations can make money too. Why cling to one big tree? An entrepreneur has to put every cent where it does the most work. A storefront on a main road can cost tens of thousands of yuan a month, but on a side street a dozen meters off that road, customers won't mind walking a few extra steps, and the rent may be far lower."

The Caffebene on Hongquan Road in Shanghai sits inside a residential neighborhood. It has two floors, plenty of room, and no street frontage. Business there has held up: customers drift in through a weekday afternoon, and staff say there is sometimes no seat to be had on weekends. But that owes a great deal to Hongquan Road itself.

Hongquan Road, packed with Korean restaurants and saunas, is known as Korea Street, and many Koreans live nearby. When My Love from the Star was at its height, this was where everyone rushed to check the place off their list. The street is much quieter now. Most of this store's customers are Korean, and even though Caffebene's headquarters has cut every tie to the China market, the Koreans in the neighborhood still like to come here.

Overall, Caffebene picked its sites more or less at random — for stores the franchisee paid for entirely, Caffebene did not take part in choosing the location at all. Starbucks, by contrast, has a full evaluation system: it weighs local average education levels, average household size, average income, the number of cars passing within an eighth of a mile, the ratio of daytime to nighttime population, safety data, the mix of businesses nearby, and much more.

Patrick O'Hagan, Starbucks's global market planning manager, told the Esri User Conference in San Diego that Starbucks now uses an in-house mapping and business intelligence platform called Atlas to decide where to open, analyzing population, foot traffic, competitors, transport and the commercial mix around a trade area together.

With the Korean chains' big-store format, finding a suitable property was hard enough on its own — and without enough foot traffic, that difficulty became a spiral.

In the end the Korean-style coffee shops blew through like a gust of wind and left a mess behind. Caffebene, Zoo Coffee and Hollys all saw franchisees vanish overnight, again and again, and there were franchisees who felt the companies had taken their money. On the Supreme People's Court website, the register of people who have defaulted on court judgments carries 28 entries tied to Caffebene China, every one marked "entirely unfulfilled."

Some argue the chains' headlong expansion was only a way to gather cash, and there is even a version in which Caffebene never intended to run a business in China at all, but used the country to get itself listed. Another version has it that the money Caffebene raised in China was all shipped back to headquarters in Korea. But Caffebene has not done well in Korea either: figures from the Korea Fair Trade Commission show its store closure rate hit 14.6% in 2015 — 140 stores — the highest in the industry.

Fights with partners never stopped. As early as the end of 2014, word was circulating of infighting between the Chinese and Korean executives at Caffebene's China company. Zoo Coffee is another brand taken over wholesale by Chinese capital, which an insider who asked not to be named confirmed to QDaily; Zoo Coffee is now pushing hard on "de-Koreanization."

The Korean chains were not wiped out in China. The more measured Maan Coffee has done reasonably well. It has no stores in Korea at all — it is a Korean brand that Shin Ja-sang, a Korean, built in China. Maan Coffee has been more careful about partners and locations, generally requiring a partner capable of taking on at least 10 stores. Its mix of company-owned and jointly operated stores has kept management quality up to a point. Shin Ja-sang, too, once promised "3,000 stores in 10 years" at a pace of 100 a year. But as the Korean chains lost ground together, Maan Coffee's growth shifted from speed to steadiness.

The new specialty coffee brands wearing the "consumption upgrading" halo will tell you they are nothing like those Korean shops, and people in the business always produce the same figures to keep their spirits up: someone in Beijing, Shanghai or Guangzhou drinks an average of 20 cups of coffee a year, against 200 in Japan and 140 in Korea. And China's freshly brewed coffee market has been growing fast in recent years, averaging more than 20%.

Meituan's China Catering Report (White Paper 2017) (《中国餐饮报告（白皮书 2017）》) counts 14,807 coffee shops closing across China in 2016, 90% of them in the second half of the year. Shanghai has more coffee shops than any other Chinese city, and Beijing, Shanghai, Guangzhou and Shenzhen together hold 16.4% of the national total. Shanghai's count fell 6% last year; Beijing's fell 20%.

The report says that "distinctive independent coffee shops are opening faster than their customer base is growing; specialty coffee is expensive and hard for the mainstream to take up, and the pool of coffee drinkers is small to begin with… Most of those entering the market now are individual entrepreneurs, and they widely lack basic business skills in store management, training staff and building standards." New entrants should note that once revenue and expansion run into trouble, more problems tend to surface.

The iapm mall on Middle Huaihai Road in Shanghai looks like the Chinese coffee market in miniature. Starbucks has three stores there, holding the traffic to the cinema, the subway entrance and the office towers. Costa is on basement level two, facing Starbucks. The specialty brand Seesaw is on the third floor, right in front of the elevators, on the route to the offices. And stayreal cafe, opened by Ashin of the band Mayday, sits next to Costa.

Starbucks plans to take its mainland China store count to 5,000 by 2021, nearly double what it has now. How much room does that leave a brand like Seesaw, whose rise is still brewing?

As coffee consumption upgrades and diversifies, Starbucks may not take all: Costa has weakened, convenience stores have entered the fight, and there is McCafé riding on McDonald's store network. What happens to this market in China next is something we will take up in another article.

Cover image: Pixabay