When Facebook opened its platform to third-party developers in late May, mom-and-pop shop widget makers from California to Turkey cropped up everywhere to capitalize on Facebook's fast-growing user base of 52 million people. Now, only six months later, there are 7,000 applications in the Facebook directory, and more than 100 are added each day.
Initially, it was easy to dismiss this widget economy as a microcosm for another Web bubble. College kids, armed only with a widget and a prayer, developed one-hit wonder applications that spread across Facebook like viruses. Their instant success sometimes led to multimillion dollar offers from overzealous buyers hoping to scarf down the remaining pieces of the Web 2.0 pie. But in reality, widgets can be innovative, user-friendly products that can be a powerful addition to a Web campaign.
Read on to learn the lessons of the widget makers.
1. Go Ugly Early
If widget makers on Facebook ever have regrets, it's that they delayed an hour or two to polish their application before releasing it to the masses. Gone are the days of brooding over pieces of code or taking extra time for testing. Since most widget makers rely on advertising for their revenue (not complicated software licenses that tie up buyers for years), they accept the fact that a widget won't be perfect on its first release. Instead, they get it out as quickly as possible, hope it catches on and then make adjustments as the user community aggressively leaves messages on feedback loops.
Take Jesse and Joe. They're widget guys. Jesse Tevelow, 24, and Joe Aigboboh, 22, formed J-Squared Media in late May, which coincided with Facebook opening its APIs to third-party developers. Aigboboh wasted no time. He developed what would become the partnership's flagship widget, Sticky Notes, in less than a week. The widget-essentially an electronic Post-It note for friends to leave each other messages on their Facebook pages-attracted one million installs in a month time and now claims around 200,000 active users.
Yet not everyone was enamored with the product from the beginning. The widget development model allows for dissent in its feedback loops, which are akin to online discussion boards where users make suggestions (or demands) on the developer. Under a suggestion board aptly named "Suggestions to make Sticky Notes app better," for instance, users have suggested tweaks ranging from a desire to change sticky note colors to a wish to override the current limit of five sticky notes on a user's board at one time.
If Aigboboh had taken an extra week to tweak the application before releasing it, he might have been scooped by another developer -- and if that had happened, it wouldn't be him and Tevelow entertaining offers between $3 million to $5 million for their company (which they've rejected, with the goal of creating more widgets and further growing the business).
There's not a good payoff for perfection," says Jim McGee, an independent consultant who advises companies on new methods of innovation. "Being second with the perfect widget isn't nearly as good as being first with an OK widget."
2. Take Advantage of the Billboard Theory
It possible to compare the widget market to the record business: You sign many bands, see who is a hit, and dispose of the duds at your earliest convenience. The successful widget companies have diversified their offerings to fit this model. "They have a stable of apps and have built a portfolio," says Gartner Valdes.
RockYou's offerings, for instance, run the gambit, from its picture apps to its messaging Glitter Text widget. As its talented developers create more widgets, analysts like Valdes wonder if they'll actually create barriers to entry in the widget market. It's pretty hard for the little widget maker to compete with RockYou or Slide," he says. RockYou serves a wide variety of social networks, including Facebook, MySpace, Bebo, Friendster and Tagged. All told, they receive about 150 million "widget views" per day. That's a lot of eyeballs-eyeballs that can be monetized.
"What we've done has never been about one product," RockYou's CTO Shen says. "We have fourteen widgets on MySpace and Bebo, for instance. We typically focus on the one-to-many model. A lot of people use the buzzword virality-we really mean it."
3. Learn to Scale -- Fast
Ask J-Squared's Aigboboh the proverbial "what have you learned" question, and his response is curt: "I got a crash-course in scaling."
The instant popularity of his Sticky Notes app was his first case study, but now he's got even bigger scaling issues to worry about with the emergence of OpenSocial, the Google-led open standard for widget development that most of the social networking heavyweights (except Facebook) recently adopted. With MySpace's recent announcement of OpenSocial support, a popular widget could suddenly need to scale to tens or even hundreds of millions of users-quickly.
That kind of growth poses serious challenges in the widget software model. On Facebook, for instance, developers are usually responsible for hosting the back end for their widgets on their own servers. "If you have a million users the first week, you have to rent a U-Haul truck and try borrowing servers from your friends," says Gartner's Valdes, referencing what a start-up company named iLike was forced to do after one of its widgets hit it big. (Google claims that OpenSocial "gadgets" won't require developers to supply servers -- unless they want to-which could be a big draw for the platform.
Widget makers on Facebook have little choice but to provide the space. Their users won't help them. Download has become an ugly word for the modern user. According Oliver Young, an analyst at Forrester Research who researches Web 2.0 technologies, consumers want as little a footprint as possible on their computers. "Convincing a consumer to download something is as easy as pulling teeth," he says.
4. Get in the Flow
Modern day users don't like seeking out applications. Instead, the application must flow to them by being pushed into their chosen site at the moment there's something interesting to see. This flies in the face of enterprise models designed during the "Days of Desktop," where users have to find various applications by clicking on title bars, shuffling windows or alt-tabbing their way from task to task. For widgets to be effective, they must operate on a push model, where the users preselect what they want to see and the software provider ensures that it gets pushed to that user's chosen platform when needed. "You don't want to have to log on to a 16 different places," Boyd says. "I, as a user, expect things to flow to me, not the other way around."
Getting into the flow also means making sure that your widget works the way users want it to-and quickly adjusting the interface as necessary. Tools have emerged to help the cause. In October of 2006, Goowy Media launched yourminis, which allow developers to create, track and analyze a widget. One nice feature of the site is a "heat map" that shows which portion of a widget receives the most user attention. Using the information, developers can then make tweaks to improve the experience (or scrap the widget and move on). "You want it to be engaging, otherwise users won't interact with it," says Gary Benitt, a cofounder and COO of Goowy Media.
Beyond just the analytics, J-Squared Media's Tevelow says he has another secret weapon for getting in touch with his users: He talks to them. "My two younger sisters are good test cases," he says. "And Joe and I are young, so we have an easy time thinking about what going to work and what won't I think a lot of big companies are having problems with that right now."
5. Ready or Not, Here They Come
Facebook launched in February of 2004. This spring, the first set of graduates who had it for their entire college careers will hit the workforce. And corporate IT departments that heed the lessons from the hardworking widget makers will be ahead of the game, says Forrester's Young. "Enterprise architects might look long and hard at something like Facebook's APIs and figure out what services they're exposing, how they manage them and try to replicate that from within the enterprise through a similar service oriented architecture (SOA) or web-oriented architecture," he says.
Some vendors have already braced for this paradigm shift. While it was MySpace's involvement in the OpenSocial announcement that dominated news reports, the buried lead story might have been that Salesforce.com<, whose Force.com platform allows the developer community to build business applications, also adopted the standard. "OpenSocial and the Force.com platform will allow developers to build a new class of powerful, socially aware applications that support business processes," says Adam Gross, a vice president at Salesforce.com. "Because Force.com itself provides an open API, developers can use it to connect to other internet social networks to build apps."
The development of these business applications will be welcomed by a generation of workers who might want, for instance, their Salesforce.com CRM data pushed as a widget to their Facebook page. For now, these users wait downstream for such information to flow to them. Upstream, the IT departments and the CIOs leading them will be in the control room alongside the dam separating the consumer and enterprise pools. They will have to decide whether to open up the flood gates all at once or just let it out a little at a time. Either way, the water level will continue to rise, and the question is, as they always say, will they be ready?