Widgets, Widgets, Everywhere

Widgets are the new cool. Everyone agrees they are a big phenomenon that’s here to stay.

No, widgets are not new. They’ve been around for a very long time in multiple technology implementations. Some prior examples are in Brad Feld’s post. As for Web-based implementations, in 2001, I was on an OASIS technical committee for Web Service Interactive Applications (WSIA), which more or less was trying to build standard (unnecessarily complicated ones) about how widgets powered by Web services could interoperate on the browser page. That work eventually led to Web Services for Remote Portlets (WSRP). And what about Kevin Werbach’s 1999 Release 1.0 issue on syndication?

In their current form, widgets are the next step in the trend towards disaggregation of content at the production end and aggregation of content by the consumer. This is why they are here to stay. They will also go mobile, partly because the form factor is a fit for small screens. Most in the mobile space, from Nokia (Widsets) to Opera (Opera Widgets) are exploring the concept. There is even a W3C TR for Widgets 1.0. (I’m surprised they didn’t start with Widgets 2.0, just to stay on part with the rest of the 2.0 vintage nomenclature.) On top of this Vista, Mac OS, Google and Yahoo have their own version of widgets. Widgetbox’s directory has about 7,000 widgets. And, yes, there is volume. RockYou is pushing 100M/day.

So, in all of this, where is the money? David Cohen thinks there is money to the made. Brad Feld is skeptical. Jeremy Liew is pondering how RockYou will make money despite its volume. Fred Wilson relates widgets to feeds. Mypartner, Mike Hirshland, pushes the debate forward.

Let’s first consider some of the models for monetizing widgets:

• Several widgets can be packaged with an ad unit next to them.
• Widgets can embed advertising in their content.
• They can show promotional campaigns, competitions or other pay-for-placement content.
• Widgets can tie into affiliate networks, e.g., buy this product on Amazon.
• They can collect valuable data, e.g., MyBlogLog.

To analyze how monetization might work we have to look at the content value chain. There are widget builders. There are the page owners (think bloggers and folks who own a profile on a soc networking site). There are the publishers (site owners). There may or may not be a widget distribution/syndication network in the middle.

Widgets are content and widgets builders can extract value based on whether that content is unique, valuable and relevant. Nothing new here but the form factor. Content owners can let widgets spread in order to drive traffic back to their sites or they can decide to monetize valuable content. What’s new with widgets is the need/opportunity to syndicate at the level of the Net as opposed to through a small set of pre-negotiated relationships. This poses some distribution and measurement challenges.

For site owners, widgets offer a way to create new inventory. They also offer some very interesting targeting opportunities. Widgets let you take several bites at the same page. Managing this and targeting for maximum impact are not trivial. Certainly, most site owners won’t let others make a ton of money off of their real-estate without wanting a cut.

For most page owners, widgets are bling. Direct monetization doesn’t make sense. The average casual blogger gets 150-250 hits/month. The average “pro” blogger gets 800-1000 hits/month. The average social network profile gets less than 100 hits/month. There is no meaningful eCPM that makes direct monetization relevant for the average page owner. That’s a BIG problem for monetizing widgets–how do you make the long tail of users put monetizable widgets on their pages? Some solutions are to (a) focus on content relevant for the page owner and (b) indirect monetization, e.g., lotteries, etc.

The opportunity for widget distribution/syndication/management platforms is to help address the abovementioned problems that arise when you try to match N widget builders with M site owners and their Q millions of users, namely:

• Discovery of widgets (content) that is relevant for people with specific interests. This is not trivial as it involves not just search but also recommendation. How else can you help widget developers “move” new widgets onto pages? As the number of widgets on the Net grows, the value of these services will increase.
• Easy distribution of the widgets, from putting them on pages to enabling actions (say, working around MySpace’s Flash linking restrictions) to making sure that content is served fast. As widgets become commonplace and some standards (formal or de facto) emerge, the value-add here will decrease.
• Measurement, measurement, measurement. And analytics, which are not easy to do in a broad syndication environment. There is a lot of value in this for two reasons. First, from the standpoint of traffic rating agencies, widgets count as page views. That won’t last. Eventually, someone will realize that serving 4 square inches of content is not the same as serving 100 square inches. Second, widgets will penetrate real estate that’s not monetizable. For example, I don’t want to make money from my blog but I may put some widgets on it. From a behavioral targeting standpoint, widget distribution networks may get better data than even some of the very large ad networks.
• Monetization enablement + audit. No rocket science here but someone needs to make money move through the content value chain.
• Widget marketing services, from SEO to SEM to viral distribution enablement. A widget syndication network may have the best data to optimize these. Some type of fee or pay-for-placement structure has to be put in place amongst widget developers to address prioritization conflicts.

This is a classic aggregator/middleman play. The main reason why these types of businesses succeed is that there are economies of scale in aggregation. The two patterns of failure involve top line collapse due to big producers cutting direct distribution deals with publishers and margin collapse due to (a) the commoditization of the aggregator value or (b) the bargaining power of large producers and publishers. There are many examples of these aggregator plays succeeding (ad networks are a prime example) and many more examples of them failing.

Who knows how this will play out with widget management systems? Ideas/opinions welcome.