Testing Viral Products, Agile Style

Blog-Blank

We’re building a pretty cool new application that has the potential to be viral.  Although I can’t tell you what the actual app is about (because its still a secret) I want to tell you how we are using agile techniques to test the concept and gather data from real users in order to validate how viral it might be, before actually building the product.

Viral applications are those that are spread from user to user in the natural usage flow of the app.  The interesting aspect that not many people know, is the notion of the “virality coefficient”.   This concept boils down to mathematics and user flows:

Hang in there with me, because this math is really simple and very powerful.  “Exposures” is the number of friends that each shares the application with.  “% Conversion” is the percentage of those friends who then become users of the system.  What’s interesting is that if these multiply together and result in a value greater than 1.0, the application will have exponential growth and a good chance of reaching literally millions of users very quickly.  Back at Snapvine (acquired by WhitePages in June), our first viral app exceeded a Virality Coefficient of 1.0 and took off like a rocket, racking up over 1 million users in about 6 weeks.  Peter (our ops guy) and I have the unique experience of frantically driving a truck around over a weekend collecting servers from friends so we could build out the data center and keep up with traffic.  A nice problem to have!

Anyways, back to the story!  The new product concept the team is developing is potentially very viral.  However, rather than spend months building the full product and launching it before knowing if the app can reach the exponentially viral “> 1.0″ point, I pushed the team to create a very simple test app of the concept — just the viral touch points — and launch that with real users in less than a week.  I don’t know if this test will show that our concept is truly viral, but I do know that this agile product testing methodology is already yielding useful data.  In the first 24 hours, our starting group of 200 users have already sent over 3600 emails to their friends and some of those are starting to convert in to the app.

What’s your own experience with agile product concept testing or designing viral applications?  Please leave a comment!

jheitzeberg@whitepages.com'

by Joe Heitzeberg

One thought on “Testing Viral Products, Agile Style

  1. matt@surffc.com'
    Tech Blogger says:

    I wrote an app for Facebook which was designed to be viral, and whilst I did see a little success with it, I wouldn’t exactly say it was an instant success. I think the trick is to ensure that your product is hot and unique – then you can expect some impressive results.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>