My immediate reaction to seeing yet another call from some website, shopping chain or mall to please install their app for all of these amazing benefits and indispensable features is to wonder just how many narrow-purpose apps any given person can have on their phone and use regularly.

Just think about your own app usage – there’s the general stuff like your alarm, calendar, email, browser, chat (i.e. WhatsApp), phone (haha),¹ SMS, and maybe a torch or note-taking app depending on your needs. That all gets used.

And then there’s the Facebook app and the Twitter app, and maybe one or two more for sites you use regularly (I’m quite fond of the Inoreader and Hacker News apps). At best, these provide a few aesthetic improvements on the websites and maybe a more solid feel, but with HTML5 and all the rest of that great new stuff, the differences between say, the Facebook app and the Facebook mobile website² are fast becoming academic.

In an article provocatively titled “Mobile Apps Must Die”, Scott Jenson writes of what he calls the golden rule for user experience (UX) design, which he formulates thusly:

Value > Pain

That is, a product must provide more value to a user than the pain it causes them. Because installing an app requires a user to open their platform’s app store, search for the relevant app, download the app (making certain they’re on wifi first if they want to save data) and then only be able to use it, the value it provides to them has quite a high threshold to overcome versus, hey, click this link! And so the user either has to already really like the app’s brand enough for the app to have value for them, or the app has to do something very valuable to them.

Benedict Evans asks brands considering apps a simple question:

Do people want to put your icon on their homescreen?

Fundamentally, there are only a few things people want to put on their homescreens. I probably don’t need the app you’re advertising for your shopping mall, or the app for an online store I’ll only use once to make a very specific purchase, or the forum app for a forum I’m only visiting because it appeared on the front page of a Google search I just did about some question I need answered.

And the app your website is directing me too is just a wrapper over some HTML or a webview, well, then the whole exercise becomes completely farcical.

There’s also the perception that apps are faster than the web. This is, in many cases, because the web is mostly being misused. Some background first: A few months ago, there was a bit of a stir over the announcement of Facebook’s Instant Articles: a service which allowed publishers to deliver their articles directly to the Facebook mobile app. Relevant concerns about the Facebook monopoly and the growth of its walled garden³ abounded.

Facebook’s rationale was this:

As more people get their news on mobile devices, we want to make the experience faster and richer on Facebook. People share a lot of articles on Facebook, particularly on our mobile app. To date, however, these stories take an average of eight seconds to load, by far the slowest single content type on Facebook. Instant Articles makes the reading experience as much as ten times faster than standard mobile web articles.

As discussed in the articles I’ve linked to above, this was not an unfounded concern. Though the web was built for displaying simple textual documents with a couple of images, it has grown and been stretched to accommodate full-blown applications and a lot of advertising. How positive a trend this is is up for debate (it does help me earn a living by explaining CSRF and friends), but what it means in practice is that most modern websites are so stuffed full of images and videos, advertising and analytics, and third-party scripts and resources that they’re damnably slow and difficult to navigate on our less powerful, but increasingly used mobile devices.

The web was built for efficient delivery of articles, but we’ve been layering other things on top for so long we’ve forgotten that, and made it necessary for Facebook to come in and built a custom system for delivering articles to mobile devices in less than eight seconds (and without ads obstructing their content from every angle). This also explains the rise of things like Flipboard.⁴

So websites try to be like apps, end up being too slow, get ditched for actual apps, and then this ditching becomes fashionable such that even lightweight websites spring for app versions of themselves. All the while, quietly in the background, the web and its browsers get more and more features to help websites become more like apps.

Despite the title of this article, I don’t think “apps versus websites” is really a versus topic at all. The correct response to “does my website need an app” is mu. An app is a different beast from a website, fulfilling different functions – where apps give you persistence in the user’s mind and space, websites give you discoverability, and where apps give you hooks into a user’s device’s deeper functions, websites give you the ability to link out to the rest of the web.⁵ One or the other of those things will usually stand out as the obvious best option, depending on what you’re doing – there’s no all-around winner. To replace the web with apps is harmful, and to replace apps with the web is another kind of harmful.⁶