Didi Is Really Taking Ride-Hailing Overseas. How Much of a Chance Does It Have?

Forced to pull back at home, Didi has no choice but to go abroad.

Didi and Uber are about to go to war again — this time, not in China.

Lately, a few Chinese users whose Android phones are set to English may have found the app switching to an English interface. That is Didi's international version, still in closed testing and not yet in app stores.

Last Thursday, Didi CEO Cheng Wei and President Jean Liu announced a new International Business Department, and put Zhu Jingshi, who had been in charge of investment, in charge of "launching substantive global business expansion."

Didi has settled on Brazil as the first stop. It has already named the head of its Brazil team: Gu Tao, formerly of the President's Office, who reports directly to Zhu Jingshi.

At the end of last month, Didi put US$100 million into the Brazilian ride-hailing app 99 Taxi, at a valuation of US$1.25 billion. After the news broke, Uber added another US$60 million to its own team there.

After months of talk about going global, Didi has finally started, in Brazil. Its opponent hasn't changed: it is still Uber, the company it bought out and made peace with in China.

Didi Has No Choice but to Go Abroad

In a speech at the Wuzhen World Internet Conference in November, Cheng Wei said: "Uber didn't leave China because of government protection. It was the result of fighting it out in the market. Didi's home game is over. Next we're going to try an away game."

Didi's attitude toward markets abroad changed last October, when Jean Liu said twice in a single week that Didi would go global. Before that, Didi had only talked about partnerships — letting users abroad open Didi and hail a car from someone else's platform. A Chinese user landing in the United States, for instance, could use the Didi app to call a Lyft.

After Didi bought Uber China, its home game looked finished. But what Uber left behind was not a market big enough to support a valuation in the tens of billions of dollars.

According to data the Shenzhen-based venture-capital cloud service provider VC SaaS gave QDaily, the number of people opening either app each day has fallen sharply since the two merged.

Between August 1 and November 17 last year, the share of users opening Didi Chuxing daily dropped from 4.5% to 3.4%, a fall of 27%. Uber China's share went from 0.5% to 0.1%.

Last October, led by Beijing, Shanghai, Guangzhou and Shenzhen, more than a hundred cities issued their own rules for ride-hailing, most of them imposing strict requirements on drivers' household registration, licenses and vehicle models. Didi fought for months and still could not change Beijing's and Shanghai's final decision to push out drivers without local household registration. The ride-hailing revolution the internet companies had brought was cut back, by government order, to a high-end taxi service.

Cheap rides on Didi Express are where Didi's real advantage lies. As early as May 2015, only weeks after it launched, Express was taking more orders a day than Didi's premium car service. Now, hemmed in by regulation, even the pricier premium service has new rivals.

Local governments in Guangzhou, Zhejiang and Guizhou have all started nurturing home-grown ride-hailing apps. After Shouqi Yueche announced a tie-up with Dazhong Chuxing, owned by Shanghai Dazhong Taxi, it laid out an expansion plan in the last week of December, opening itself to local franchise partners — mostly companies backed by local governments.

Worse, other internet companies have decided that the new rules put them back on the same starting line as Didi. Meituan, which serves local eating, drinking and entertainment, recently began testing a ride-hailing service in Nanjing.

To keep its users and orders from shrinking, and to keep its huge system running, Didi has had to bring taxi companies in as partners. "Promoting the integration of taxis and Express" was something Didi worked on all through the second half of last year.

It rolled out "smart dispatch," which hands Express orders to taxi drivers automatically, weighing distance, traffic and the balance of supply and demand. Between August and November last year, at least 200 taxi companies signed deals with Didi.

In this reorganization, the taxi business has been folded formally into Express. The Express, taxi and Uber units were merged into one Quick Mobility business group, run by the executive who previously oversaw Express and platform operations, and who now also runs the taxi business.

The Didi Bus unit, set up in 2015, has been renamed the Public Transit unit and placed under a new Smart Transportation FT team led by Zhang Wensong, the chief scientist Didi poached from Alibaba. Plugging in taxis, buses and other public transit only generates transaction volume for Didi, not revenue or profit.

Even in courting taxi companies, Didi runs into state-backed rivals such as Shouqi. Unlike signing up Express drivers, Didi's famous on-the-ground sales push is not necessarily more efficient than one big company dealing directly with another.

After buying Uber China, Didi was valued at US$38 billion. That is the valuation of a company that owns the movement of 800 million city dwellers. A high-end taxi company is not worth that much. Neither is a map app with taxis and buses plugged into it.

Expanding abroad, and making money where the rules are looser, became the obvious choice in front of Didi.

In speeches and interviews late last year, both Cheng Wei and Didi CTO Zhang Bo talked about plans to start going global.

Now That It's Really Going Abroad, What Will Didi Do?

At the company's annual meeting twenty days ago, Cheng Wei said the next phase of Didi's globalization would lean less on the capital alliances of the last one. "We'll build an elite unit, a Didi expeditionary force, and send it to Brazil, Japan, South Korea and Europe," he said.

In South Korea, designated-driver services still hand out cards in the middle of the night; in Brazil, the government subsidizes public transit far less than China does and illegal buses are everywhere, so a minibus business would have a future. Those were the examples Cheng Wei used to argue that Didi should be "confident" about going out.

But none of these businesses are on the rise in markets abroad to begin with.

Last November, Didi announced a strategic partnership with the car rental group Avis Budget: the two would offer Chinese users cars from Avis and Budget in nearly 175 countries and regions. The service went live this January, letting Chinese travelers rent from both companies directly inside the Didi app.

The effect on Didi's globalization is small, and this kind of deal does nothing to answer the question of what supports a US$38 billion valuation.

Last July, Uber struck a deal with Hertz, the world's second-largest car rental company, to lease mid-size cars — Camrys, Passats and the like — to drivers on the Uber platform at relatively low prices, which effectively turned Hertz into a car supplier for a ride-hailing company. Hertz had said publicly as early as January last year that ride-hailing apps were already hurting its growth.

The person Didi is sending to Brazil is the strategy vice president responsible for financing and acquisitions, not an executive from Express or the premium car service.

In that same annual-meeting speech twenty days ago, Cheng Wei listed "globalization" as one of his keywords for 2017. "The next phase of Didi's globalization won't rely as much on the capital alliances of the last phase," he said.

By "capital alliances" he meant the "anti-Uber coalition" Didi built across Southeast Asia, South Asia and the United States while it and Uber were still burning subsidies at each other in China.

In December 2015, the company — then still called Didi Kuaidi — announced a four-way plan to connect its product with Lyft, Ola and GrabTaxi. As Uber's rivals in several of its biggest markets, the four said they would link their apps so that a user arriving in a new region could use a partner's service directly.

Investment was what made that cross-border alliance possible. In May 2015, Didi put US$100 million into a US$530 million round for Lyft. It also joined funding rounds for GrabTaxi and Ola.

But all of these companies were already valued highly, and Didi's stakes in them were small.

The loose alliance changed once Didi bought Uber China. Uber China merged into Didi at a value of US$7 billion, and Uber Global took a 17.7% stake in Didi. After Lyft and Didi connected their services in the US, the others never went any further.

In the email laying out the internal changes, Cheng Wei and Jean Liu wrote:

"The International Business Department will take the innovative products, technology and services we have developed in China and roll them out in global markets, serving a broader population and more diverse cities."

The man Didi is sending is Zhu Jingshi, who ran strategy and investment. Zhu was an executive director at Goldman Sachs' Asia-Pacific division; in the second half of 2014 he followed Jean Liu from Goldman to Didi, which then set up a strategic investment department. In his two years at Didi he handled fundraising, and led a US$3 billion round in 30 days — which Cheng Wei singled out at Didi's 2015 annual meeting as "breaking the record for fundraising by any unlisted company in the world."

If Didi meant to expand the way Uber did, it would be sending executives from the premium car or Express businesses, not the strategy vice president who raises money. Which suggests its international business will still run on acquisitions and investment. Counting a US$2.5 billion loan from China Merchants Bank, Didi can now draw on roughly US$10 billion in cash. Uber has about as much: as of last June, its balance sheet held more than US$11 billion.

At The Wall Street Journal's WSJD Live conference last year, Jean Liu said Didi was assessing which new markets to enter. If Didi decided the companies already in a market were not strong enough, it would go in and compete head-on.

For now, Didi has chosen investment plus technical support as its way into international markets, with Brazil first, and perhaps the Japan, South Korea and Europe that Cheng Wei named after that.

Which Markets Are Left?

Clay Shirky, the New York University professor whose book about social media, Cognitive Surplus, was a bestseller and who has spent the past few years watching Chinese tech companies, once gave us two ways to think about this.

"In practice, Chinese companies' global strategy will start out as a regional strategy. Just as Xiaomi will do better in India than in Brazil, a Chinese ride-hailing company will find it easier to expand into Malaysia and the Philippines than to start in Nigeria or Argentina," he said.

The other direction is countries and regions that draw enough Chinese tourists. A startup there doesn't have to begin by winning local users; capturing what Chinese tourists do abroad already gives it some scale.

But these markets aren't sitting empty waiting for Didi.

Four months after leaving the Chinese market, Uber is back in China's app stores — with an app built for Chinese users to hail rides abroad.

If you registered before Uber China pulled out, your old account still works and all your trips are still there. Outside mainland China, Uber wants you to use this app. According to Trip.com Group's latest Chinese New Year outbound travel data, eight of the top ten destinations for Chinese tourists are markets Uber already serves. Didi's international version, built on its Lyft partnership, can so far only call a Lyft, and only in the United States.

Uber has spent the past two years trying just about every market there is.

Two years ago, Beijing was the hundredth city Uber entered. Over the two years that followed, Uber fought Didi hard while also moving into 400 cities around the world. Its astonishing valuation helped its competitors raise enormous sums too — everyone wanted to be the next Didi.

Grab, Ola and Lyft, which had formed the "anti-Uber coalition" with Didi, were among the biggest. Their markets — North America, India and Southeast Asia — were also where Uber spent the most. The subsidy war from China played out there as well, as investors followed Uber's lead and tried to grow a local Didi of their own.

None of these are the weak markets Jean Liu described. But with the money Didi has today, it is not easy to expand across several global markets, rebuild the China business and buy these companies all at once — unless Uber crushes them first, in which case there may be no need to buy them.

They may not want to sell, either. Lyft recently said its 2016 ride volume tripled year-on-year, to 160 million rides. That is still far behind Uber, which completed 78 million rides in the US in December alone, against Lyft's 18.7 million.

We counted up the cities outside mainland China with more than 2 million people.

Outside mainland China, 1.231 billion people live in 215 cities of more than 2 million:

Among them, the US, India and Southeast Asia already have local companies worth US$3 billion or more, and Uber is very active in all three — India, for instance, is now its largest market by orders outside the US. The Didi-Uber fight could keep going there the way it did in China: large populations, a market with potential.

That leaves 810 million people in 152 cities of over 2 million. Already bigger than the Chinese market.

Apart from a few war-torn or desperately poor places, most of these markets are underdeveloped because of regulation.

Uber has been in Western Europe for years, for instance, but faces tight restrictions in most cities. Not one local ride-hailing startup there has reached a US$1 billion valuation.

Japan, Canada, South Korea and South America are much the same.

Japan bars non-professional drivers from carrying passengers, so for now Uber only hails taxis by phone and delivers food there, waiting for the rules to catch up.

These markets will open. Premium cars plus carpooling run far more efficiently than taxis — even taxis hailed by phone. As one city after another finds its transport working better under the new model, change will reach other places sooner or later.

North America already shows it. In 2012, Uber could not even get into New York, and its CEO said publicly at an event that corruption was the main thing blocking the technology. Today New York has changed its rules, and Uber has driven the price of a shared ride there down to a fraction of a traditional cab fare.

Quebec, to the north, tells the same story: Uber could not get in in 2012 and now has permission to run a trial. Presidents, mayors and transport ministers change every few years, and policy does not stay fixed forever. When these markets do open, Uber will hold little first-mover advantage over Didi, and in theory the two will be back on the same starting line.

But that will probably come only after both have tried their luck in Latin America, Japan, South Korea and Southeast Asia.

Succeeding Abroad Will Be Much Harder for Didi Than Winning at Home

Uber today is valued at US$68 billion and still isn't profitable, which makes it easy to assume that all it does is burn money.

But the company, founded in 2009, did not raise its first round of more than US$100 million until August 2013.

By then it had already spread to dozens of cities worldwide, testing each new one cheaply with its famous three-person teams: when Uber moved into a new city, the market was usually opened by three people — a city general manager, an operations manager for drivers and a specialist who handled everything else.

The 500 cities Uber runs today sit on every inhabited continent on earth and talk to each other across more than 20 time zones. Its engineering teams in Europe and the US have spent years learning to build software together.

Behind the capital and the loud CEO, Uber has become an extremely efficient multinational.

Didi, for its part, does not lack the ability to run a big company. Its recent annual meeting drew 6,800 employees.

In the year and a half it went head-to-head with Uber China, Didi added a business unit every time it added a service: starting with taxis and premium cars, then Express, designated driving, carpooling, buses, car rental, corporate cars and minibuses, one after another, each with its own product, technology, operations and marketing teams. Growing fast, Didi also turned quickly into a bloated big company, its headcount climbing from just over 1,400 in 2014 to where it is today.

Because the Chinese market is so large, Professor Shirky observed, Chinese companies had already reached enormous scale and learned how to run a huge organization before they ever needed to go abroad.

Didi's size is not only a matter of headcount. By its own figures, its system completes 20 million orders a day — taking the request, assigning the order, planning the route, navigating, collecting payment — and processes 2,000 terabytes of data daily.

But the 400 cities Didi covers are all in the UTC+8 time zone. It has no experience hiring teams abroad to open new markets. And only now is it starting to deal with its own internal efficiency. As Cheng Wei put it in a closed-door conversation last November, after a round of growth this fast there is a lot of catching up to do: scale can be bought by burning money, but a team and technology cannot simply be bought.

"In our first five years, Didi hopes to become the world's largest one-stop mobility platform. In the second five years, we hope to make Didi a world-class technology company." That was the future Cheng Wei described in a speech late last year.

Going abroad will not necessarily solve everything, either. Bloomberg reported recently, citing people familiar with the matter, that Uber's early employees may sell up to 10% of their shares — which means Uber will not be going public any time soon. And holding on to employees may be harder for Didi than it is for Uber.

Didi is now in its fifth year. With policy shifting at home and competition abroad starting from zero, it still has a lot to learn.

In its long fight with Uber, Didi won on capital, on traffic from its investors, and on a huge sales force that went door to door regardless of cost. Didi shows where Chinese and foreign internet companies differ in competition: American internet companies can take almost any market in the world and yet cannot get into China. At the same time, the internet companies that succeeded in China's market — Tencent, Alibaba, Baidu among them — have never managed to get their products into markets abroad.

Roelof Botha, the well-known Silicon Valley investor who backed YouTube and Instagram early, has said that there is too much money in the market, which lets many startups avoid facing the final question.

"With that much money around, it's like the exam date keeps getting pushed back. You don't need to study any harder, you just need to cram later," he said.

Now Didi has to sit the real exam.

Cover image from VisualHunt.com