Posted on Apr 25, 2009

There’s a saying along the lines of  “the amount someone spends talking about themselves is inversely proportional to how interesting they are”. Beyond first dates and keynote presentations, this opinion directly affects social network services.

Users drop off at an accelerated rate from accessing/signing up for a new site to actually using it. Even if the sign-up process is super slick and the site is easy to use and helpful with telling users how to get started, more often than not companies forget one significant thing:

If the first impression is a ghost town until the user interacts with the service more (e.g. adds friends, follows feeds, etc.) – that will be their impression of your service and most likely their last login to it.

For Pownce, a social network I was a community manager for, this was a known weakness. I don’t have the data to show the drop-off rate from users who signed up for Pownce, but as with any social network, there’s always a large divide between active users and total users – and keeping the divide as small as possible plays a large role in the longevity of a site.

(original screenshot via Chris Messina)

Comparatively, when users sign up for Flickr, they’re shown the activity that is being created by other users on the site immediately:

(original screenshot via Chris Messina)

Additionally, Flickr always shows content other users are contributing on their service on your personal dashboard, regardless if you’ve added friends or uploaded photos of your own. In this way, Flickr is communicating that they have a live and constantly active ecosystem to participate in – making the user feel less uncertain about adding to that activity and interacting with the site.

The recommendation being – don’t use emptiness as a motivation for users to interact more with your site, even if you have super friendly instructions. Displaying example content not only shows that your site isn’t dead inside, it shows users what’s interesting without saying it.

Side note: I’ve been looking for other blog posts or links to data that discuss the ratio of sign-ups to returning users. My insight on this topic has mainly come from discussions with various social network developers. If you know of any links I should check out, please leave a link in the comments!


  1. Ryan Kuder
    April 25, 2009

    When I was at one of the big internet companies, we used to call this the “empty restaurant” problem. If you walk into a restaurant and there’s nobody eating there already, you’re not likely to hang around, no matter how good the food is. Conversely, if you walk in and the place is bustling, you’re probably going to give it a shot. It’s something we’re working on solving with our site via a combination of prompts for starter content as well as including third party content as a default.

  2. Tarek
    April 25, 2009

    You know – I should not say that but … – I have a similar problem in a location based social service I am creating called It’s still in alpha but I’ve noticed that most the people who used it got confused and didn’t really know what to do there. And yes, may be the emptiness is the reason here.

    I think one solution for that, is to have a bot or virtual user on your service and let him/her be friend of anybody joins it. And that bot can do some actions on the site every now and then in order to displaying example content to them.

  3. Ed
    April 25, 2009

    We ran into a similar problem at the last startup I worked for. Getting users to convert was always an issue, no matter how much we tweaked that first-time user experience. What really made a difference, though, was finding what the ideal use cases were and marketing to those users who might benefit from those applications. In the end, it totally changed our perspective as to what our service did.

    Unfortunately, it also told us that we might not be as mainstream as we had hoped the product would be, either.

    C’est la vie.

  4. Carl
    April 25, 2009

    I don’t have any silver bullets for solving the problem of dropoff, but this directly jives with the bit of “The Tipping Point” which I read (for the first time, yes, I’m behind this particular curve). Namely: “Stickiness”. One of the major forces in deciding whether a trend will survive / epidemic will spread, is whether the thing is “sticky”. All you need is the proper whizzbang, noticeable, memorable widget, and you’re golden. The nature of the widget, though, eludes most (all) of us. It might be the content in Flickr; or the bot that Tarek talks about, or just the name of the site, or anything, really. No-one knows. What appears to be the case though to me is: People tend to know it when they see it. And yes, emptiness: NAH.

  5. PunkToad
    June 18, 2009

    Flickr has the best interface of all time, it’s a game.

  6. angelo
    July 1, 2009

    very interesting topic, you’ve got going on here – you’re right about Flickr’s interface and its ability to attract right off the bat..i personally love it!