Tencent Built a Flywheel That Spins Beautifully. It Just Can't Leave China.

A street vendor in Guangzhou takes payment by holding up a printed square of black-and-white squiggles. A customer points a phone at it, taps once, and money moves - inside the same app where, ten seconds earlier, they were arguing in a group chat about where to eat. They never left. They booked the table, split the bill, hailed the ride home, and read the news, all without closing one screen. That refusal to leave is the entire engine. WeChat now holds more than 1.4 billion monthly users⁴, and every minute they spend inside makes the next minute more valuable.

The story usually told about Tencent is that WeChat is a flywheel: chat pulls in attention, attention powers payments, payments enable little apps-within-the-app, and all of it generates the data that sells advertising - each turn spinning the next, faster and cheaper. That story is true. It is also incomplete in a way that flatters Tencent badly, because the wheel has nearly stopped accelerating, and the company's quickest-growing engine is bolted on the outside of it.

How the loop actually feeds itself

Start at the center and walk the loop. The original asset is attention: WeChat is the only app in China with over a billion active users, one of just five apps on earth to clear that bar.⁶ Because everyone is already inside for chat, WeChat Pay sits a tap away from every conversation - and it now serves over 900 million users.⁶ Once money moves frictionlessly, mini programs become possible: lightweight apps that run inside WeChat with no download, no separate install, no second account. There were 4.1 million of them registered by 2023.⁶ Those mini programs generate transactions and behavior, which make the advertising slots inside WeChat - and across Tencent's properties - sharper and more targeted. And revenue from all of it funds the next reason never to leave. The genius is that no single piece is the moat. The moat is that they refer to each other.

It did not arrive fully formed. WeChat began as a project at Tencent's Guangzhou research center in October 2010, launched in 2011 as Weixin, and only took the WeChat name internationally in 2012 when it crossed 100 million users.⁵ The flywheel was assembled one layer at a time on top of an app that started as plain messaging - and that incremental construction is exactly why it is so hard to copy. A rival cannot bolt on payments without the attention, or mini programs without the payments. You have to build the layers in order, and Tencent already has.

The wheel is spinning - but it has stopped accelerating

Here is the number the flywheel narrative tends to skip. Combined Weixin/WeChat monthly users grew from 1,343 million at the end of 2023 to 1,385 million at the end of 2024 - a 3% gain.² By the third quarter of 2025 the figure was 1,414 million, up roughly 2% year over year.⁴ A flywheel is supposed to compound; this one is adding users at the rate of a mature utility. The reason is simple and structural: in China, almost everyone who can be in WeChat already is. Meanwhile Tencent's older social asset, QQ, is shrinking outright - its mobile users fell 5% in 2024 to 524 million.² When your input - attention - is at saturation, the loop keeps turning on existing momentum, but it cannot speed up. It can only get more efficient at monetizing a fixed audience.

The fastest growth is coming from outside the loop

If the flywheel were the whole engine, you would expect payments and social to be the fastest-growing lines. They are not. In FY2025, Tencent's total revenue grew 14% to RMB751.8 billion - and the standout was International Games, up 33% to RMB77.4 billion. Marketing Services grew 19%; Domestic Games grew 18%; FinTech and Business Services - the part most tied to WeChat Pay - grew just 8%.³ International games are sold through console and PC platforms, app stores, and Tencent's stakes in studios worldwide. They have essentially nothing to do with the WeChat flywheel. The clearest sign of where Tencent's momentum lives is that international games already reached a record 30% of total games revenue back in 2023.¹ The self-reinforcing loop is the company's foundation. It is not its frontier.

Isn't a closed loop the strongest moat there is?

The fair objection runs like this: of course growth is slowing - the thing is already nearly universal in its home market, and a saturated monopoly is a wonderful problem to have. A walled garden that everyone is inside is the most durable position in tech. There is real force here. The flywheel is genuinely hard to attack, and the lock-in is real enough that competitors litigate over it - ByteDance sued Tencent in 2021, alleging WeChat blocked Douyin links to protect its own short-video feature, seeking about 90 million yuan, with analysts noting the suit was aimed less at the damages than at provoking antitrust scrutiny.⁷ But that is precisely the problem with this kind of moat. The same walls that make the loop self-reinforcing are what regulators came for. The SAMR fine Tencent took in January 2022 was a procedural one - for failing to report acquisitions in advance, not for monopolizing the WeChat ecosystem⁷ - but the direction of travel is unmistakable. A flywheel that depends on keeping users walled in has its growth ceiling set by saturation and its risk ceiling set by the state. Both ceilings are domestic. Both are now in view.

Tencent built one of the cleanest flywheels in business: an app nobody needs to leave, spinning chat into payments into mini programs into advertising, each layer making the next cheaper. It is real, it is documented, and it is not going anywhere. But it is also finished growing in the only place it works. The next chapter of Tencent is being written in international games and advertising - outside the wall, on rails the flywheel never laid. The self-reinforcing loop didn't fail. It succeeded so completely that it ran out of room, and the company's future is now the part of itself the loop was never built to hold.