Case Study: MakeFive Concept, Online Community, and Web App Development
Ever find yourself comparing lists with friends? Us too! We love sharing our favorite movies, music, restaurants, and almost anything else. Combine our propensity for list-making with the fact that we build web apps and you’re left with our own—sort of addictive—online community.
MakeFive is a discussion platform that’s made manageable by lists. In it people skim topics, make their picks, and then discuss them. It was built on the notion of creating “top 5” lists for anything and everything. Through this simple concept, users connect with each other over shared interests, extending their network beyond friends and acquaintances, to like-minded people.
We’d be lying if we said that we really understood what we were getting ourselves into with the design, development, and promotion of our own social network. Sure, we’d built substantial web applications before, but never to this extent. Our first weeks were comprised of an intensive process of exploring potential functionality, establishing classification systems for content, and (what felt like endless) sketches of the potential user experience design.
Our initial prototype went live within 6 months of the project start date. Given the depth of the system, and the small team working on it, we were particularly happy with this turnaround. Getting it out quickly provided insight into how people used the system; what we learned allowed us to rapidly iterate and make the web application better. An example of this was in the list building interface we initially developed. It was elegant, using simple drag-and-drop in the browser window. The problem was that most users didn’t know they could do such things within this context.
As much as we liked UI functions like this one, we reshaped them however we needed to. A design solution that people don’t use is quite likely flawed; therefore, we held nothing sacred. We continued working like this for many months: listening to the community, using the site for ourselves, carefully examining functions, and even ripping things apart when we needed to. Sometimes this meant simplifying a submission form; other times it involved implementing third-party authentication protocols. Whatever it took, we just concentrated on making the thing work better.
With MakeFive, users can vote on a topic with a single click, plus add videos, photos, descriptions, and comments to their selections. Additionally, they can create profiles, share lists with friends, and communicate with other members. It’s a fully featured social network, with custom algorithms developed for everything from calculating points for users (the most active users earn prizes and status), to determining the ranking of different items in a list (not as easy as it sounds), and finally integrating moderation systems for limiting spam and system abuse.
In addition to the requirements of the community itself, we wanted to explore methods of “seeding” the site in other parts of the web. This involved a number of different approaches including the development of a Facebook application, a widget that allows for adding dynamic lists to other websites and blogs, and integration with individual users’ Twitter accounts. Each of these technologies involved its own unique set of challenges, from third-party APIs that are in a constant state of flux, to limits imposed by the multitude of browsers and platforms in use.
Many site owners dream of getting a big surge of traffic—perhaps by making the homepage of a social sharing site like Digg. What few consider are the logistics related to such traffic surges. We faced this early on, when a topic did just that. Suddenly, the site was receiving hundreds of visits per second —a nice problem to have, but a problem nevertheless. We quickly adapted our underlying technology stack and relocated the site to Amazon’s EC2 cloud computing environment in order to meet these demands. The site now utilizes an entire suite of cloud services to improve access times and seamlessly adapt to varying processing requirements.
Aside from those “lucky” moments, be they a feature on Digg, or the occasional reference on StumbleUpon, we faced some very real challenges in marketing the site to new groups. Sure, the property worked, but community building is a bit of a Catch-22. Put plainly: until it reaches a critical mass, most people won’t give it the time of day; but, how do you actually get your community to that point? We looked at a variety of methods for building site activity, in order to kick-start the community.
Initially, we focused on getting the word out to other popular destination websites. As some of them made note of the property, traffic increased. From there we worked to entice the users through contests with weekly iPod and cash giveaways, and community rankings that some users took great pains to achieve. Like the design, this was an iterative process and we learned as we went. On one day we’d see people log-in and use the site for 8 hours straight. On other days, we had to deal with users trying to “hack” the system to win the contest.
MakeFive is not a household name like some other social networks out there. That’s a pity, as it is awfully entertaining. What it has accomplished is notable, though: a steady increase in traffic that is becoming rather substantial. The site currently sees around 12,000 visitors each day, which puts it on track to achieve over four million visits this year.
“The concept is simple: you make a list of anything you want—the worst typefaces of all time, the best tech sites, or things to do after the economy collapses—and vote up the items you think should make it to the top five. The site is well-designed and fun.”
—Erick Schonfeld, Co-Editor
TechCrunch