For me, it all started in 2007. I was still pursuing my BFA in New Media Design at Rochester Institute of Technology. My friend Adam Barnum had just gotten a designer position at Thumbplay. As it turned out, they were making a huge marketing push to sell mobile content, their major product at the time being Ringtones and Wallpapers (their current services now push Full-Track MP3’s and soon, streaming music, but I digress). They had designers, but they needed a Flash person to help implement their designs. I agreed to help, and we were soon on our way.
I was doing tons of Flash banners, from regular website banner sizes (728×90, 160×600, 300×250) to even the banners that were displayed on the top of AOL Instant Messenger buddy lists. Each banner had around 4 pieces of content. The process was simple, create a template file in all shapes and sizes using their creatives, and cut and paste all the new content in. The issue here was that all of the advertisement companies enforced ridiculously strict file-size requirements (think 15-20k, really ridiculous, considering the amount of content that was in these banners), and so I would be optimizing all aspects of the banner until the banner’s file size was under the threshold. SWF file size optimization almost became a science – recreating as many of the design assets in Flash (because more images mean a larger file size, and re-drawing designs in Flash are vectors, which are math-based and have a much smaller file size), compressing images as small as they could be without the loss of quality, and using as few embedded font faces as possible.The audio previews were pulled dynamically from Thumbplay’s CDN servers, so there was no issue there. I became a machine, generating so many Flash banners over such a short period of time that my classmates had started calling me the “banner guy”, and they came to me when they needed optimization help on freelance projects that dealt with banners.
One of my contract projects shortly after that was a Flash module for a landing page, codenamed “Lefty” (above). It was a pretty simple setup, an XML-driven Flash module that displayed album art and played previews. There were next and previous buttons that would let the user cycle through content. The Flash module was placed into an image of a phone on the landing page, and the creative/marketing team at Thumbplay were extremely happy with the results. The benefit of having dynamic content powered by an external XML file was a huge time and money saver, and they were able to update the landing pages internally with new content, whenever and with whatever content they wanted.
Because of the amazing portability of this setup, they became XML-hungry addicts. XML this, XML that! Pretty soon all of the Flash banners I created for them had the great power of XML driven content behind them, and they were able to maximize impressions, which led to more subscriptions. Best of all, there was only one .SWF file per size, per design, and they were able to create banners by simply creating a new XML file and changing the XML URL flashvar that was passed into the banner.
I graduated in May of 2008, and they scooped me up to bring me in house. Most successful of all the banners was Thumbplay’s Google Gadget, a Flash widget that was served on Google’s Gadget Ad Network. The biggest advantage to this network was that Thumbplay’s marketing team could have the widget content match the webpage’s content around it, for example, a Kanye West Widget on Kanye’s Flashing Lights YouTube music video page, the popular Daft Punk soundclip site iDaft(above), or a Britney Spears widget on a popular Britney news blog(the first photo in this entry). Being able to customize an ad based on the content displayed on the page made it more likely for people to want to click it, and increased click-through rates severalfold.
For even more exposure, the folks in the business development department decided to take these ads to MySpace for Thumbplay Open, an affiliate program where artists and other partners could make referral money when people subscribed through their links. We ran into a roadblock here, because MySpace doesn’t allow link-outs from Flash modules embedded on MySpace. We managed a work around, which was just to include an html footer that linked to a general landing page. That way we could still have the customized content while being able to make our affiliates money. We were able to create the Widgetron (above), which allowed our partners to easily create customized widgets with personalized music lists, colors, and sizes. Our Affiliates would then be able to copy and paste the embed code on their sites; the whole process could take less than five minutes. Today the widgets are embedded on hundreds, if not thousands of artist MySpace sites.
The widgets were huge. At one point we were acquiring many thousands of subscription sign-ups a day! Everyone loved them, even our competitors. Over half a year later, we saw many attempts to imitate our success, with similar gadgets popping up around the web, made by rivals Jamster, Sendme Mobile, and other rival companies as well (below).
Imitation is the best form of flattery, isn’t it?
The views expressed in the above entry are my own and do not necessarily reflect those of Thumbplay, Inc.
