I was asked to give an impromptu tutorial in “widget advertising” last week in Bangalore. What started off as a casual request turned into a cubicle with a few people looking over my shoulder at a web browser with four tabs loaded with:
- My iGoogle page
- The new Lenovo/Microsoft Live Portal
- The plugin directory at WordPress.org
- Fred Wilson’s Blog, A VC
I have heard it said the best way to learn something is to teach it, and having never commissioned nor served a so-called “gadget ad” I found myself in the same mode I was in three years ago at CXO Media, when I conducted a series of brown bag lunches on the silent revolution occurring in the post-bubble world of web media. This is the abridged version of what I told the gang in Bangalore last week.
Traditional web advertising is defined by a call-and-serve model where a publisher fences off some turf which is occupied by a string of code which calls for an advertisement housed on an ad server. Getting that ad onto that server is the act of trafficking, and can attach conditions to when and where the ad is shown. For example, an advertiser might buy a specific section of a web site for a certain period of time, and only want an ad to be shown during certain hours, or to users from a certain geographical area, and further insure a visitor would not see the same ad more than three times. All of this is enabled between the code on the web site and the ad server.
Changing the terms — the duration of the ad campaign, the creative (say lowering the price in a banner ad in response to a competitor lowering their price), and making any modifications is a gigantic pain in the ass that requires a change order and all sorts of ad ops pain which one’s agency has to deal with. Therefore, the old model of serving ads, Dart trafficking, purchase orders, etc. , in my opinion, sucks.
Now, look at what is possible if the ad were fixed in place, like a plug-in feature embedded in the right side gutter of this blog. See the Flickr gallery? The My Blog Log thingy? Those are widgets. Apple has em. Yahoo has em. Google has em. WordPress has em. They are supposed to be all the rage and bloggers more eloquent than I have waxed eloquent about them. Some rag apparently declared this to be the year of widget. Um, okay, but how do they work for advertisers?
Let’s start with my personal Google page (my default homepage, except fricking Lenovo.com takes it over like a jealous girlfriend every other day). You understand the principle, it’s like every other portal page ever developed. MyYahoo, MyWay … pick a content module — a clock, a calculator, quote of the day from The Big Lebowski (“the Dude abides”), and place it where you want it. When you start selecting stuff, search on HP (yes, I am showing my competition some love here) and up comes this gadget.
Now, note to Eric Kintz, this thing hasn’t been updated in months, but the spirit is there and the potential is immense. So, for some reason, let’s say I am a real fan of HP cameras, I decide I need to see what they are selling every time I go to the Google portal. HP could, if they refreshed it, modify the price, change the image, push promotions, in short — this is a “perma-ad” that me, the user, can decide to place or alter.
What’s the difference between this and an ad unit? First off, it isn’t served by an ad insertion engine like DoubleClick or Atlas. Second, to modify it I don’t need to replace the entire ad, only elements inside of it. Think of this as a site within in a site. A mini-site.
Similar to this is what we’re doing with Microsoft for our homepage default in Internet Explorer. Setting the default on the browser to Lenovo.com isn’t a good idea — we get lots of garbage traffic that immediately goes elsewhere because the user can’t or can’t be bothered to change their browser’s homepage default. No, instead you want to offer the user a homepage with some modicum of utility, something that — dare I invoke the term of the year of 1997? — is sticky and will be useful. This page, a cobranded version of Microsoft Live at MSN, is an attempt to do that, and we are building a widget/gadget thingy to keep our customers informed, tell them about features and functions they may not be aware of, and let them know when we’re conducting a promotion or having a sale.
Then there are plug-ins. Plug-ins are modules a site-side publisher — like a blogger — can download and embed on their blog. Some of these plug-ins are “badges” or fixed spots that a blogger can grab as part of an affiliate program with say Amazon or Bass Pro Shops. The concept is simple — give the vendor some real estate, and any sales that result from a click from the affiliate badge will carry a unique identifier which yields the blogger/site owner a percentage of the sale.
Finally, if you want to see a site in action that uses a lot of plug-ins, widgets, gadgets, etc., you can’t do much better than Fred Wilson’s A VC. Fred is a venture capitalist at New York’s Union Square Ventures and he invests in a lot of the companies that produce these plug-ins. He’s also blogged smartly on the topic.
So, to recap:
Gidgets are the next wave in display advertising. They pose a challenge to creative teams who need to design a unit that can be easily updated, through a syndication pipe using RSS/XML principles to publish new content from a remote owner. They pose a challenge to publishers who are accustomed to a random, or roadblock model in ad tagging and serving. They probably will simplify the lives of the people running metrics as they aren’t subject to any randomness and probably will have a longer shelf life than other units served on a campaign basis. Will they perform better than banners? The Jury is out.
0 thoughts on “Widget, Gadget, Gidget”
Argh… my mind is ranging with ideas… thanks for the masterclass.
Bring on the love, David! 🙂
Let’s not forget that implied in this is that the gidget must work properly always. The customer will never forgive you if you hose their homepage, blog, etc. with rotten or unmaintained code.
And we’ve certainly seen our share of widgets around here that have hosed the code, haven’t we David? 😉
It’s not a show stopper, but I heartily recommend a thorough and rigorous qa and beta cycle.
I completely agree that the next step in display ads is rss/xml. Seen a prototype by one of our favorite Harley riding Web Producers…cool stuff. The ad can be updated hourly if you want, but changing the rss feed. Talk about maintaining relevance…