Essence of viral apps

So there are a few things necessary for apps to be viral (some not necessary but at least very helpful).

Social in Nature: The first is that the app has to be social in nature.  That means an app that you naturally use with your friends, where the functionality increase as the number of friends you have using it increases, or where the functionality of the app revolves around interacting (socializing/exchanging) with friends.  This is slightly different from (a subset of) total network effects, where your utility of the app increases as the as the total population using the app increases (e.g. gmail spam filtering).  For example, chat clients – the more friends you have on a particular chat client, the more useful it is to you as an indivudual because you can communicate with these friends.  So you would want them all on the chat client, hence invite them all on.

Looking at some of the more facebook apps, this same principle applies. Superwall/Funwall – what use would this be if no one can see what you have posted for people? They have to have the app also to see this.  All the gifting apps: you have to give the gift to someone!!!  Sell your friend (human pets, friends for sale, and all the other clones) – make your friends do things.

Invite to Use: A second is forced virality (depending on how you execute this, it may violate the do no evil principle).  Force people to invite other people before allowing them to use the app, or certain functionalities of it.  Alternatively, you can bait people to invite more friends by offering points/premium functionality.  Texas holdem – get points for every friend you invite.  All those stupid apps where you have to invite 20 friends to see the answer, or even to see the app itself (which may be a fake app) (note this is not a sustainable as either an app or a business – and very bad karma if you care about that).

Viral Vectors to Spread: You have to employ as many vectors to allow people to notify other people of the service.  In facebook world, this includes the standard invites, notifications, news feeds, minifeeds, presence on the profile, etc.  Each of these has to have a call to action, and a compelling reason for the reader to take that action.  In non-FB world, this includes an easy way to email all their friends (e.g. plaxo), and adding to social bookmarking services (digg, delicious, stumble upon, etc) so strangers can discover it.  The more of these vectors you put in, the easier it is to spread.

Easy signup: It must be very easy for new users to start using the app.  There should be little to no registration process (in FB world, all this is/should be automated)

Metrics Tracking and Continuous Testing: You have to have mechanisms to track where your users are coming from, who is sending invites and why they are sending it, and why people sign up or don’t sign up – and where they are dropping off in the process.  This means building in tracking metrics right from the start into the apps that you build, and building A/B testing mechanisms to test alternatives on what is more efficient in driving the behavior that you want.  A difference of 1% growth (especially on a daily or weekly basis), or of an extra 5% of invited people signing up for your service is huge.  You won’t know how to get that extra 5% (or even a lot more) without knowing what works and what doesn’t.

Useful App: One last thing that I have to mention is that the app you are building has to be fun/useful/interesting to being with.  If you built a piece of crap, of course people won’t tell their friends or invite them to use it   So that means at the end of the day, you still need to build a great application, especially if you want to retain the users that you’ve managed to get to visit your site or use your application – and keep using it.