The Formula For Creating The Next Killer App


Every time a friend tells me about a new app, this is how it goes:

In fact, this happened with each and every one of the following:

How is it that these leviathans of the tech world — Google, Facebook, Hacker News — are so easily replaced by young upstarts that seem to have no extra functionality? What is it about these new apps that appeals to people — and what does it take to jump on the bandwagon and build your own?

The answer is that there’s a formula for building the next killer app.

And here is that formula, in all its over-simplified glory:

I. Pick a massive tech product that’s used for loads of different purposes

II. Take one specific bit of its functionality

III. Turn that into a beautiful standalone app

This is generally known as unbundling, and it’s behind some of the most popular apps of the last 12 months.

Citymapper — the directions bit of Google Maps

People use Google Maps for a bunch of reasons. One day you’ll open it up to find the nearest coffee shop; another you’ll need to know the name of the street you’re on; another you’ll want to figure out the quickest way home.

The problem is, because there are so many things you can do with Google Maps, actually doing any one of them is trickier than it need be. So along came Citymapper with an app that’s completely focused on getting you from A to B. And, as has been well documented, because of this focus it can be infinitely simpler and infinitely better:

Getting to the Apple Store with Citymapper (10 seconds)
Getting to the Apple Store with Google Maps (20 seconds)

The result? A $10m Series A and a place on the home screen of every smartphone user in its rapidly growing list of cities.

Product Hunt — very specific up-voting

You can use sites like Reddit and Hacker News to up- and down-vote any number of things, from Bitcoin stories to Photoshop battles. And, sure enough, if you want to know what the most popular new apps out there are, there’s a subreddit for that, and a designated page on HN.

But, if all you want is a quick snapshot of what’s popular right now, it’s unbelievably cumbersome to have to navigate to Reddit, single out that particular subreddit and sift through the endless questions and news stories to find the few posts actually about new apps. Again, the reason it’s cumbersome is that it’s just one of many things you can use Reddit for.

That’s where Product Hunt comes in.

Product Hunt takes the narrow functionality of submitting new apps and up- and down-voting them and turns that into a clean, standalone app. And it does so to the tune of a place on Y Combinator and $6m in funding.

Product Hunt vs. its messy alternatives

Yo — 2014's answer to the Facebook Poke

We’ve been collectively Poking each other since Mark Zuckerberg pivoted from FaceMash in 2004. But, as the sheer number of features that Facebook offers has ballooned out to cover everything from on-site gaming to buying and sending stickers, the Poke button has receded further and further into the app, hiding away in an obscure, rarely visited sub-menu.

Cue Moshe Hogeg, who wanted a simple way of summoning his secretary. Of course, he could have used a Poke — but, in the time it took him to navigate to the right section of Facebook, find her name and confirm that he did indeed wish to Poke her, he could have just walked into the next room and tapped her on the shoulder. And so Yo was born — a list of names, a single-touch interface, and nothing else. $1.5m of investment later, and possibly the simplest app ever conceived is going strong.

Facebook Poke vs. Yo

And now the

big tech companies

are cottoning on

It’s not just young upstarts who are taking specific bits of massive, multi-purpose products and refashioning them into standalone apps — the big tech companies have started chopping up their own products, too.

Just look at what Facebook’s doing. In recent months, they’ve released Groups, Messenger, Camera and Pages, all standalone apps for things you could already do on their main site; they’ve spent billions buying Instagram and WhatsApp, third party apps that offer much the same functionality; and, for good measure, they’ve built Slingshot, Hyperlapse, Paper and Rooms, each of which offers an entirely new way of creating, sharing and consuming content.

The upshot of this?

Where Facebook used to be a website, it’s now an app company.

What’s even more crazy is that apps that were already unbundled versions of other apps are now unbundling themselves. Take, for example, Cluster. They’d taken one element of Facebook — sharing photos privately with specific groups of people — and unbundled that into a standalone app. But that wasn’t enough for them — they found that people were still using their app for purposes that could be called different. Perhaps you want to share photos of your baby with your family. Perhaps you’re a teacher sharing photos of your class with their parents. Well, in Cluster’s eyes, that’s functionality that can be unbundled even further. So they created a bunch of new apps for every one of these usage patterns that sprung up.

So why is unbundling so effective?

Startups that follow this rule are raising huge rounds; established companies who do the same are pulling in reams of new users. So what is it about unbundling that works so well?

I reckon it has three things going for it: simplicity, ubiquity and the death of the logout.

I. Simplicity

When your app only needs to do one thing, it can do it really well. You can have big, clear buttons for the things people use it for —because there’s no need to clutter up the screen with all those other options you’ve got rid of. This makes your app unbelievably easy to use — and that’s what smartphone users want.

II. Ubiquity

The phrase There’s an app for that has never been more true. People are getting more and more used to having a different app for everything they do — I have one app for checking train times, another for communicating with my colleagues, and another for finding the nearest Starbucks. In a world in which people are more accustomed to opening up an app in order to perform a very specific task, it makes sense to give every task its own app.

III. The death of the logout

Using lots of different services used to be a hassle — every time you switched between services, you’d have to enter your username and password (or at least go through a login screen, even if it did remember your details). But with mobile apps, login has all but been removed; you login once, and can then promptly forget your password like it’s the 14 times table. Switching between apps is as easy as swiping between photos. Your phone already knows the person using it is you — so there’s no need to incessantly ask you your mother’s maiden name or the make of your first car.

When it’s so easy to switch between apps, and you’ve got two things your service needs to provide — say, a news feed and private messaging — why not split it up into two separate apps?

Of course, the truly innovative apps aren’t the ones that simply move existing functionality out of a complex system — the real innovators are the people doing something new. But that doesn’t mean unbundling is in any way second-rate app-making — far from it. If there’s a task a huge number of people perform, and someone comes along and makes that task easier and quicker to accomplish, the amount of time saved by humanity as a whole is incalculable. That in itself is reason enough to unbundle.

So, if you’re weighing up what your next app should be, have a look at vast, sprawling services like Google, Facebook and Amazon, find something they do that takes a couple too many clicks to accomplish, and turn it into a standalone app. You never know — it might just be the next Yo.

(My vote is for birthdays. Would someone please make a decent birthdays app. Thanks.)


My own foray into the world of tech has been in building Jukedeck, a system that composes original music and gives you unique, bespoke music for your videos. I’d love to know what you think!