A talk given at Wiredcraft’s UX/UI Shanghai, February 9th, Shanghai.
Download the presentation:
Key Takeaways
The closest analogy to WeChat in the West is Google Search
Being that services (mini-programs) are built as web apps in WXML, a variant of HTML, WeChat is a browser; a browser is a portal for accessing internet services
In contrast to the West’s popularization of the internet through the desktop computer in the 1990s, the mass of Chinese users first started accessing the internet via mobile devices
The original Super Apps were in fact desktop web browsers like Netscape and Internet Explorer
Mobile apps in the West are generally focused on providing one service, while browsers function as portals, while WeChat in China also functions needs to provide the functionality of a browser
For a Super App to succeed in the West, it requires that existing user behavior lock-ins are unseated
Microsoft’s success depends on whether it can find a new paradigm for accessing information—that could be OpenAI’s Dall-E and ChatGPT which generate new textual and visual data, instead of indexing and displaying what already exists on the internet
Introduction
For the past decade WeChat has been the envy of Silicon Valley thanks to its large user base and its stickiness, and therefore its potential to generate tremendous amounts of revenue. Yet nobody has succeeded in building something similar, what is often called a Super App, or even really attempted it? How come? And what makes China so different from, for example, the US market and the behavior of its users, that an app like WeChat can become successful in one place but not the other?
In December, Microsoft’s ambitions to build a Super App leaked. Why are Microsoft, without a record of building successful consumer apps, attempting this and why now?
In this essay I will use the theory of path dependence to understand how WeChat became popular, and find the service analogous to WeChat in the West, to understand what it would take for a Super App, by Microsoft or somebody else, to become a thing in the West.
A Super App from… Microsoft?
Microsoft recently considered building a “super app” that could combine shopping, messaging, web search, news feeds and other services in a one-stop smartphone app, in what would be an ambitious move by the software giant to expand further into consumer services, according to people with direct knowledge of the discussions. Microsoft executives wanted the app to boost the company’s multibillion-dollar advertising business and Bing search, as well as draw more users to Teams messaging and other mobile services.
Unlike Apple and Google, Microsoft doesn’t operate a mobile app store for smartphone users. By creating an all-in-one app that people don’t need to leave to access its other offerings, Microsoft hoped to emulate a mobile strategy that has worked for Tencent. The Chinese firm’s WeChat app, which combines messaging with shopping, online games, news and a variety of services including grocery ordering, is a source of inspiration for top Microsoft executives, the people said. —The Information
So Microsoft’s ambitions seem to not be fueled by a new idea, but a want to grow its revenue—a typical top-down initiative. Those are rarely successful. Regardless, for the purpose of this essay, let’s pretend we’re the Product Manager that gets asked by their manager to create a strategy for building and launching this Super App in the US and beyond. Before we do that we’re going to need to understand:
What environment did WeChat come out of and what path did it follow to gain popularity?
How did it fit in with existing user behaviors and which new behaviors did it create?
What unique value did it provide to users or where does it fit into the user journey?
(And why target Google Search instead of WhatsApp, which is a messaging service like WeChat?)
Path dependence
Path dependence is a concept in economics and the social sciences, referring to processes where past events or decisions constrain later events or decisions.[1][2] It can be used to refer to outcomes at a single point in time or to long-run equilibria of a process.[3] Path dependence has been used to describe institutions, technical standards, patterns of economic or social development, organizational behavior, and more. In short, it can be stated as the idea that:
“The future development of a system is affected by the path it has traced out in the past.”
So launching a Super App in 2011 is very different from doing it in 2023, as the past 12 years have built up user behaviors and expectations that need to be accounted for. And doing so in the US, which is on a path separate from China, is similarly different.
When WeChat launched in 2011 it was far from as dominant as it is today. It was just one of many messaging apps and it wasn’t even clear why it was so much better than using SMS, which was the standard back then.
After some time battling it out with other messaging apps, a semi-external event occurred: parent Tencent acquired or let WeChat use its banking license to support payments. On top of this, the ‘red packet’ feature was built, allowing users to send red packets to friends and relatives around the Chinese New Year. WeChat became the default messaging app in China, as acquiring a banking license is close to impossible in China, as in most other countries.
There are many different perspectives to approach path dependence from, but I’m going to focus on what I think is the most significant: time. The US reached 20% internet access penetration in the mid-90s and 50% the early-aughts. At both points in time, the desktop computer was the dominant and almost only way to access the internet as mobile phones were still dumb.
China on the other hand reached 20% in the mid-aughts and 50% around 2015, when smartphones were ubiquitous and capable as personal computing devices and had all but replace dumb phones. Most people who upgraded their phone around this time got a computer with internet capabilities for free and may not even have had a PC at home. For the Chinese mainstream, the smartphone was the way they built the habits as a netizen.
User behavior—and what is a Super App?
So what user needs does WeChat satisfy and what behaviors exist around its usage? And what is a Super App, really?
“WeChat is an OS.”
There’s been many ways of explaining what WeChat is. One of them is that is is in fact not just an app but an OS; an idea popularized by the blogs of Silicon Valley venture capitalists like Sequoia and a61z from around 2015.
An effect of WeChat functioning as an OS, was said to be that the device (ex: iPhone and iOS) would become irrelevant, so WeChat was supposed to be a threat to Apple’s market share.
Clearly, Apple is still doing well in China (24% China smartphone sales in 2022Q4) and commanding a premium, so WeChat does not seem to be an OS.
“WeChat is an Everything App.”
Another similar idea is that WeChat is an app that does everything, so you don’t need any other apps on your phone.
WeChat does a lot, but not everything, and most importantly it doesn’t do it all best. Just like in other markets, there are still successful dedicated apps.
If WeChat really was a Super App in the sense of doing everything a mobile user could want to do, why would other apps exist?
In a lot of cases, like booking a taxi with Didi, the clicking through from WeChat is often the first time the user tries out a service. The simple WeChat mini-program is fine for something you do once or seldomly, but if you do it frequently the janky user experience (like pressing back and losing your location) is going to be frustrating so most users end up installing the Didi app after a few times. So WeChat actually funnels users to third party services.
Let’s think how this works in the West. When we hear about something new and want to learn more about it, we don’t open a messaging app like WhatsApp but instead we open a browser and perform a search. From that, we’ll jump to websites or apps, in the same way that Chinese users will search in WeChat to find the mini-program or official account.
When WeChat was rumored to add mini-programs in 2016, there was speculation on how they would split revenue with third party developers (Source: WeChat’s Next Step Toward a SuperApp, The Information).
In 2023, it is clear that no revenue sharing is needed—third parties develop mini-programs for the same reason that businesses optimize for Google Search: so that they will receive traffic from WeChat’s massive user base.
So here’s the theory we’re going with: WeChat is a gateway to the internet for mobile users, filling the same function as the browser on desktop.
“Super App” simply means providing a range of rudimentary services across a number of areas
It is a starting point, but not an end point
After booking a Didi through WeChat a few times, you will probably move on to downloading the native app and use that instead
This is similar to how a web browser works: use it for initial discovery, then move on to a dedicated service
The closest analogy to WeChat in the West, then, is search, mainly Google Search. Now it makes sense that Microsoft would target Google Search instead of WhatsApp with their coming Super App.
Will Microsoft’s Super App ambitions succeed?
The challenge with building a Super App like WeChat in the West is that there are already established user behaviors (path dependence: ‘lock-in’) and dedicated apps that make the value of such an app diminutive.
A new paradigm is needed, that breaks with current behavioral patterns, and offers enough value for us to learn a new behavior.
Enter ChatGPT. After starting work on this essay an presentation, Microsoft announced a follow up investment in OpenAI, the creator of ChatGPT, giving them 49% ownership and exclusive rights to use version 3.5 and up of GPT.
Microsoft’s plan is to build ChatGPT-like capabilities into Bing and thereby disrupt Google’s search dominance. Is this going to succeed? Let’s try to look at what we learnt about path dependence and WeChat’s path to understand if the Generative Bing has what it takes to become a Super App.
Absolutely. ChatGPT is the fastest growing service ever, reaching 100M users in 2 months. It is a significantly different way of interacting with information on the internet.
Yes. There’s going to be numerous Generative services fighting to gain dominance.
What’s Microsoft’s hongbao? Unclear, and likewise, it is unclear if they’ll be able to come up with one.
This follows from #3; creative work requires experimentation and autonomy. Will the Bing team be able to try things out like the WeChat team did? Unlikely. But the same goes for competitors like Google, Meta, and Baidu. So perhaps the answer will come from a much smaller player?
Addendum, 2/4 2023
While finishing up this essay, Microsoft’s new Bing with GPT integrated was leaked. And, Google’s press conference to counter this has been announced for February 8th.
The new Bing.com front page, where the search box is now replaced by a chat box:
Results page after entering a query:
If that is what is coming, kill me now.