How To Build The Next Super App
And Why It Won't Look Like WeChat
In this post, I present a novel framework for thinking about super apps, and argue that the future might actually be a lot closer than we think. If you want to keep up with my thoughts on the creator economy, entrepreneurship, and other dispatches from my life, please subscribe!
What Even Is A Super App?
I have long been fascinated by WeChat’s success in China. Now, with Elon Musk’s recent stated desire to turn Twitter into an “everything app,” I thought it would be interesting to share some of my thoughts on what it might look like for a super app to take hold in the West.
I believe that super apps are the most effective way to break outside the walled gardens of the Apple and Google App Stores and move into a new world of innovation backed by consumer choice and true competitive dynamics.
I also believe that WeChat is simply a prototype of what the next “everything app” is going to look like, and that to spend time analyzing WeChat is to miss the forest for the trees.
Let’s start by unpacking how super apps actually work. There are two important pieces: a host app and a mini-app layer.
The host app is usually a big, generic, broadly-applicable service—often something like a social network, messenger, or peer-to-peer payments app.
The mini-app layer is a bundled runtime that lets third-party developers extend the functionality of the host app. Some of these extensions are small and trivial (e.g., a mini app that lets you book a movie ticket or order a piece of clothing), while others are experiences that could be seen as full-on apps in their own rights (e.g., dating, games, ride hailing, etc.).
The mini-app layer is what turns a generic, vanilla app into something that is infinitely contextual and useful, and this abstraction means that the host app doesn't need to change itself in order to "fit" these new product experiences. The host app simply becomes a platform for launching all kinds of app experiences—much like our phones’ home screens.
Existing Social Media Apps Are Already “Super-ish”
There are a lot of people who have thought about how to build the next WeChat. Products like Facebook Instant Games and Snap Minis have even attempted it. While neither of those took off, that doesn’t mean it can’t be done—and I’m sure that people will keep trying.
I want to propose a different way of thinking about it. A way of thinking that gets us to the same end state much faster and more efficiently—but does it simply by leaning into patterns that already exist.
Consider the following.
When we use our phones, there are really only a handful of apps in which we spend the majority of our time. Indeed, the bulk of time spent on mobile is spent inside social networks like TikTok, Instagram, Facebook, and Twitter.
We engage with the native content inside those apps—feeds of photos, videos, tweets, etc.
When we want the next level of functionality, we don’t close the app we’re in and switch contexts to a different app—you don’t tap “Buy Now” on an Instagram ad and have it open a deep link to another native app. Rather, the contexts “stack,” and the app presents a WebView where you can, for example, buy something from someone’s Shopify store. Then, when you’re done buying, you close the WebView and you’re right back in the feed.
This same behavior happens when you tap an ad on Twitter, or someone’s Link in Bio on TikTok, or a link sticker inside an Instagram Story.
Every social app uses a WebView to open links. This means that, in effect, the beginning of a universal mini-app layer is already available inside every social app.
What’s Next?
With this in mind, I would argue that the easiest way to create something that embodies "Super App Dynamics"—including capturing ecosystem value in the same way as WeChat—would be to take one of the existing big social apps and lean further into making the WebView feel like a native experience. This might even include standardized, platform-level hooks between the native app and the WebView for things like payment info, identity, and other core services.
This approach has two outsized benefits:
It busts outside the walled gardens of Apple and Google, and all the various app store policies, taxes, and other headaches. A mini-app layer powered by the open web keeps things at arms length from the host app, which would certainly find itself butting heads with Apple if it tried to build something bespoke.
Because this mini-app layer is just the web, you don't need to bootstrap a new developer ecosystem and convince developers to build and support platform-specific mini-apps, nor do you need to build new infrastructure and architecture to be able to support these apps inside the host app.
What we've been building with Koji is this second piece of the equation—the development environment, runtime layer, and creation experience that lets ordinary social media users create and share all kinds of web-based mini apps, to do everything from sell ebooks to share games, crowdsource content, and more.
Koji’s mini apps can already be shared on every social network, including via the Link in Bio on TikTok, Instagram, and Twitter, as well as via Link Stickers in Instagram stories.
As the world continues to move in this direction, we believe that it is inevitable that every major app begins to be used as a super app.
Bravo! The logic is sound! The vision is clear!