Unbundling Apps

The medium of the 'App' shouldn't be the only way we solve problems on mobile devices

As devices have become more powerful and grown to encompass more facets of our lives, apps have proliferated in both quantity and complexity, and it may be time to reconsider whether apps should be this way.

More often than not, apps wind up being more complex than they really should be and create significant user experience overhead as users try to solve their needs. This growing overhead has three root causes:

  1. Medium - apps are fullscreen experiences with access to the device's full capabilities, which suggests to developers to expand into those broad constraints, and they are launched from the springboard which adds to a base level of UI overhead.
  2. Marketplace - the App Store's compensation model (one-time, sub, IAP) and prerequisites (branding and shelf display) suggest developers create products that compete in ways that grow them into businesses, not just tools.
  3. Momentum - apps are central to the knowledge reservoir that exists among device users; they understand and use apps to solve problems, which creates expectation and demand for companies to use the medium of apps to interface with their customers.

The Medium of Apps

Apps as a medium encompass many different needs from the very small to downright massive. Some needs require complex interfaces, others don't really require interfaces at all. Some are single-purpose, others very multi-purpose. It's also worth reminding ourselves that not all apps solve needs, and not all needs are solved (or solved well) by apps as a medium.

Using apps right now incurs a fixed cost of overhead that is worth questioning. Users must:

  1. Make the connection that the need they have may be solved by an app they have (if not, go to the App Store)
  2. Remember the app icon or name
  3. Find the app either spatially on the springboard or via search
  4. Launch the app
  5. Navigate to find the part of the UI where the app does the thing they want from it
  6. Look at or use the thing, hopefully remembering how to
  7. Usually then navigate to another app to do a related thing

Meanwhile, apps litter the springboard and app switcher UIs, and with each app the overall device experience worsens just a little bit as all apps become harder to find and navigate between.

The App Store Drives Apps to Become Businesses

In addition to the fixed costs of the medium, the conditions for feature creep come from the increasingly direct coupling of app <-> business, which act as a variable cost of complexity depending on the market the app is in. The competitive business environment of the App Store creates strong business incentives to:

As app developers add more of these features, their financial costs rise and the app may switch to a subscription model to keep it afloat.[1]

To get a little silly, imagine a wrench you bought was also a brand you had to interact with to use, had its own Instagram account talking about what bolts it turned today, and required a subscription to turn more than 10 bolts a month. That honestly feels like the state of many apps today.

I'm reminded of music albums that only had one or two good songs and 10 tracks of filler, or books that had a few good points among chapters and chapters of blah blah blah. These have to expand themselves to a size necessary to be sold in the markets they're in. Apps are no different.

Where To Go Next With Apps

Right now, apps are the only way to solve user needs on mobile devices [2], and both the grain size feels too large, and the user experience holding it all together is starting to show its limits. There are many apps that would better meet user needs if they were unbundled into some different modalities:

Of course, any system is more than the sum of its parts. These would need to be weaved in elegantly into the overall user experience of the operating system, which likely would involve a near complete overhaul of things like springboard. And there's a strong user education component to introducing these new concepts. It will take years to build up the knowledge reservoir across hundreds of millions of device users, so any changes to apps must be introduced incrementally.

In closing, I think there's a lot of opportunity to create the tooling and user experience context to allow developers to unbundle apps, and so avoid some of the trappings of the current trend of app feature creep. It's a big undertaking, but this feels like where the puck is going.


  1. I'm not anti-subscription; in fact, I think it's often the best way to align user need and business need over time. My critique is that a lot of this is an unnecessary artifact of the business environment and primacy of apps on mobile devices. Most user needs are not worthy of this level of complexity. ↩︎

  2. Aside from voice assistants, which are getting smarter about connecting to services and meeting some needs. Unfortunately their capabilities are largely unknown and invisible to users right now. Hopefully that will change! ↩︎