Designing a Compelling Widget: Three Keys to Widget Engagement

This week the Widget World Expo took place in New York and Fred Wilson gave a supposedly terrific speech titled "Why Widgets Is The Wrong Word For What We're Doing". I really wish that I had been able to attend the conference - but the schedule didn't allow for it. Fred's deck in embedded below and Silicon Valley Insider gives the following recap:

To Fred, the “problem with widgets” stems back to the original separation between content and the first widgets — ads. Google was smart to put its own contextual ad widgets (AdSense) in the sidebar of its searches, being clear to its users that its content (search results) were pure and separated from their ads (paid results).

However, as widgets started to be used to display other web content -- Fred notes that his first widget was Flickr’s photostream widget, back in 2005 -- they became "relegated to the sidebar and increasingly seen as ad units and increasingly ignored.”

But they shouldn’t be ignored, Fred argued. They should be integrated into the flow and experience of the page. Developers, he said, need to put more focus on widget user experience.

This ties into a question that I get multiple times a day: what makes a great widget? The answer ties directly into Fred's point that "developers need to put more focus on widget user experience".

1. Give Users a Reason to Come Back A widget will fail if it doesn't give users a reason to engage and a compelling reason to return. When designing a widget, ask the following questions: "What's the value proposition for a user? What's their motivation to embed the widget?" You'll find that the answers are closely aligned: dynamically changing, interesting content prevents staleness and adds value to a webpage in a way that a static banner cannot. There are countless ways to do this: sales rankings, RSS updates, community activities, etc.

2. Make it Customizable Being "relegated" to the sidebar isn't a bad thing... it's consistent, highly trafficked real estate. But the publisher should (at least) be able to control the widget's size and color / theme. There are no standard widget units (or IAB sizes) and it's important to remember that your widget is a guest on other peoples' websites; consequently, you should make it as simple as possible for the publisher to create, grab and embed. Customizations can go much further: Jib Jab's Starring You widgets are great examples of full personalization.

3. Market Softly and Carefully I agree with Fred that widgets are "increasingly [being] seen as ad units and increasingly ignored.” I don't associate that, however, with placement locations (his belief) - in fact, we at Widgetbox have data showing that in-widget engagement is less a factor of location and more a factor of widget design. I believe the reason is that a lot of widgets are closer to 'portable advertisements' than branded widgets. The marketing and branding on a widget should be done gently - if it's too aggressive, publishers are turned off and unlikely to dedicate their valuable real estate to (what is essentially) unpaid ad units.

Here is an example of a highly successful widget that follows the above rules (400,000 installs and millions of views each month). The BabyTicker widget tracks the development of a baby inside the widgetized womb:

1. Engagement: users embed the widget to share progress with family and friends; meanwhile, viewers check back regularly to monitor the baby's growth. Proof that users find the BabyTicker compelling and useful: some of their most active embed locations are on start pages like Netvibes, iGoogle, PageFlakes, etc.

2. Customization: you can set the size, the mother's name, the number of babies and the due date. The widget reflects those changes and grows accordingly.

3. Branding: There is prominent, but intrusive, branding beneath the widget for