Social web applications all have a similar sign-up/ramp-up flow. Attract a user to a marketing site, get them to sign up and most importantly get them to connect to other users. This last step (usually known as ramp-up) is crucial in driving engagement and growing a community. The guys behind Friendfeed (now owned by Facebook) found that once a user connected to 5 or more people, they became active and logged in reguarly. Users who didn’t cross that threshold disappeared pretty quickly. Emergent metrics such as these are very useful for learning how to increase customer activity and loyalty.
In Exceptional the big indicators for us are if a user’s application throws an exception and if a user invites their teammates to start using it. If both of these things happen, the chances are we have a loyal and active customer. So what can we do to use this data wisely? Well, we throw a test exception on any new application install, so that all users will can see and close an exception once their app is configured. Our pricing change means that adding your teammates is now free, removing the participation tax that Jason Fried wrote about. This has worked well for us as ramp up strategy, and as we learn more about our users we’ll continue to work on it. Let’s look at some other ramp-up techniques.
Getting users up and running
Instagram has a great mechanism for quickly filling up your photostream, simply provide your twitter name and with one “follow all” press you quickly see a shedload of photos. Even better is the “recommended” list where you see impressive photographs and can easily follow their owner with one tap. All of this has helped the hockey-stick growth of the application. Contrast that with Foursquare’s approach.
This is a very different approach and as you can see it’s harder to get up and running with Foursquare. Instagram is using recognition, you see people you know and tap follow. Foursquare uses recall, you must think of people you want to follow and then find and follow them. It’s similar to trying to name everyone you went to school with off-hand versus doing it when looking at a class photo.
Design influences behaviour
That’s not to say Foursquare is wrong or Instagram is right. Design influences behaviour. On Instagram I follow heaps of people I don’t know, because they post nice shots. Foursquare can’t work that way, it only makes sense if you know the people you’re connected with, otherwise you just see people you don’t know arriving places, which is boring. Facebook should in theory fall into the Foursquare category, but in reality some people use it for friends and family, some use it to track their college friends, some see it as a business marketing tool and some simply see it as a way to get laid.
Quality versus Quantity
When designing the “find your friends” screen in a social web application you need to decide whether you’re optimizing for a large number of loose connections or a small number of strong ones. This influences everything from the flow (recall versus recognition, automatic versus optional, unilateral vs bilateral relationships) to the words you use (add as a friend, start following, connect with, become a fan, request access)
The over-hyped and thus inevitably over-criticized Path.com is clearly favouring strength of connection over quantity of connections. Path limits you to only sharing with fifty people, and it only suggests you add 3 (I only have one). All of this re-enforces the idea of a personal network. This limit seems crazy in a space where it’s seemingly all about connections, but chasing the numbers is often a short sighted goal.
Take LinkedIn for example, where connecting to someone used to mean something but has now degraded into yet another game-able metric, up there with Facebook “fans” and purchased followers. Path.com may be an unclear offering, but there is something rare and valuable in a network focussing on the quality of relationships over the quantity, a network where the average person isn’t connected to 150+ others and adding buttons to their website to attract even more.
In this regard it will be interesting to see how Quora fares, its recent explosion means it now has tonnes of users and connections, but the fundamental question is “When a user logs into Quora are they interested in reading and interacting with the information in front of them, or is just another useless activity feed”. The former means Quora is going to be a success, the latter means it’s the modern day Google Wave. Right now it’s doing well for me.
Designing for engagement
Most well built web apps attract early visitors dipping their toes in, and looking around. The successful web apps manage to convert them. As a designer the challenge is to find what exactly made the long standing customers get hooked initially, and find ways to bring those features or usage patterns to all customers, whether it’s five friends, one good friend, or finding a way to help the users kick ass. The harder part is doing it without encouraging meaningless activity at the expense of your applications credibility.
The following are some good links in the are of ramp, sign-up, engagement and connections. Thanks for reading.
- Design patterns for sign up and ramp up – a worthwhile, free, report from Adaptive Path (requires newsletter sign-up)
- Metrics Driven Design [PDF] – a slide deck from Joshua Porter’s UX London presentation.
- The real life social network – a now famous set of slides from Paul Adams
Update – Jorge from Foursquare wrote a comment to say if you provide your own Twitter name as input you can select from your Twitter followers, a la Instagram. Yay for Recall
Follow @destraynor on Twitter.