correct me if i’m wrong

Disclaimer: The use of this logo is complete and utter copyright infringement.

My friend Adam Derewecki came up with a crazy idea a week-and-a-half ago: Let’s see if we can piggyback on the Superbowl by building a Web application and quickly monetizing it with ads.

The application was quick and simple. We created a mashup that contained an ESPN Superbowl widget, an incoming Twitter feed for each team, and ads. The result is TwitterBowl 2009.

Our plan was straightforward, as well. Build the application over the weekend before the Superbowl, and spend the week advertising it in every free way possible. Our opinions of the results are mixed; he feels it was sucessful, I feel it was a failure.  I’ll explain my side of the story:

1. Building the application was a nightmare. We conceived such a simple application that nothing could go wrong… right? Wrong. Adam paved the way by getting started in advance, on Thursday, a week and a half before the Superbowl. He began by hand-coding a static HTML page. No frameworks, just CSS and floating Divs. It was quick and dirty. After all, we only had to include a few third-party items on the page:

• Ultimate Twitter Search Widget from TweetGrid; a reference to a javascript file wrapped in <script></script> tags.
• Official NFL.com Widget; an embedded <object>.
• Google AdSense,
• Google Analytics,
• … and AdBrite; all small javascript snippets wrapped in <script></script> tags.

We quickly discovered that combining these separate widgets, simply put, is a headache. Our three-column layout simply did not work.  We collaborated on coding the single page, and every change we made seemed to break something new, despite the fact that we are both fairly experienced with CSS, HTML and Javascript. I have been using the three professionally for about 6 months, and for personal use for over three years. I am not a CSS guru by any stretch of the imagination, but I have a degree in Computer Science, know technology and software very well, and have gone through the painstaking processes of designing a website in pure CSS. Unfortunately, CSS is a terribly implemented lanaguage.

So, during the middle of last week, I took it upon myself to prototype our prototype (ha!) in the Google Web Toolkit (GWT). I have been using GWT to develop a side project for the last 3 months and it’s sheer power and simplicity amaze me. For the last 3 months, I have slowly forgotten the pains of CSS, browser-incompatibility issues, the brittleness of Divs, and the general pains of Web design. These all came back to smack me in the face last week. Hence…

I will never develop anything for the Web with any tool beside the Google Web Toolkit from now on.

It is, quite simply, illogical not to do so. Imagine building a chair. Chairs vary from very simple to somewhat complex. Yet, there is an upper limit to their complexity. A chair can never as complex as, for example, an automobile. An automobile requires very specialized tools and skills to build. For a chair, this is less true. I could probably go to the hardware store right now and have a crude chair built by the end of the night.

Yet, chairs do require a certain prerequisite of equipment and knowledge. For example, I could try breaking various boards over my knees to cut them and try driving in the nails with my palms. This could seem logical, even feasible, until my knees are bruised and my palms sore. Only then will I go hunting around for my hammer. I may even find my power drill to make it quicker.

Without the proper framework, Adam and I were basically driving in nails with our palms. It wasn’t until after we were tired and frustrated that we decided to experiment with a different approach. Plus, I am the only one of the two that has experience with GWT, so it was more of an experiment than an actual decision. With enough pounding and hacking, I eventually got it to work, and since it worked well, it ultimately materialized into the final product.

This wasn’t even straightforward, and resulted in a bit of a hackjob, but the results were worth it. We achieved a cross-browser (tested in IE7, FF3, Chrome and Safari) application that was accessible to anybody with a recent Web browser; particularly important when your demographic is football fans, which lends to a random distribution of technology platforms.

Lesson learned? All the people that are currently taking part in the Divs vs. Tables blogosphere debate are silly. I don’t even know how my website is laid out, because it is abstracted from me (For what it is worth, GWT code compiles most of the layout widgets into tables. It relies on CSS for styling and everything else.). In conclusion, use the right tools for the job, no matter what the scope.

2. Programming pains had a negative impact on the product’s advertising and evangelizing. Because building TwitterBowl was a pain and relatively time-consuming, there wasn’t a whole lot of time left to devote to spreading the word about our shiny new Web application. I finished the application on Thursday night before the game, and added the finishing touches on Friday. Therefore, all that was left was Saturday and pre-game Sunday, and we both had other plans; we hoped to have spread the seeds by that point in time.

Adam did a good job in finding sources that were interested in this application and having his friends evangelize it. I was hesitant to spread it around before it wasn’t rid of its major problems, seeing as it would be a poor reflection of my work. Sure, I’m a perfectionist, but we did end up shipping this thing. Moreover, I didn’t pour myself over the minute details, but rather sought something “good enough” to proudly display. As a result, my contribution to the advertising side was relatively weak and I was honestly kind of burned out by working on it to really push it out.

Before

After

The only real selling or evangelize of anything was when I served as ACM President at my university. The officer group, which consisted of seven of us, would constantly devise ways in which we could maximize student attendance at our event and increase interest in our organization. Ultimately, the channels for a school organization are fairly limited: flyers, departmental mailing lists, facebook groups, etc., and they all serve a small subset of students that might actually be interested in the organization. A Web application, however, is much different. When you are sitting at your computer, thinking of who you should target for a Superbowl application, the world is at your fingertips, and it’s really quite daunting. 

Lessons learned? The importance of advertising and evangelizing is diminished in the tech world because programmers succeed despite obstensibly lacking these skills. It’s importance, however, is paramount. If your product is on the Web, and nobody knows it exists, is it an actual product? No. Just like the silent tree falling in the woods. In conclusion, dedicate the proper resources and people to advertising and evangelizing.

3. I don’t care about professional football. I don’t know what the authoritative media sources are, who the hot players are, how many passing yards Roethlisberger has this year, or really anything that would ultimately trickle into the product to make it better. Fortunately, Adam is a Steeler’s fan, so he carried most of the weight in this regard.

My friend in marketing, Charlie Hoehn, gave us some great suggestions for evangelization, including scouring the football-related blogosphere to contact important figures in the community about the project. However, I have no idea what they are, and given the scarcity of time on theadvertising/evangelizing side, finding the proper sources did not really happen. Our blog coverage was pretty minimal.

Lessons learned? Build something based on what you love. I have no idea how Hank Mackay enjoys selling envelopes so much. I sure wouldn’t be able to. I have written on this blog that entrepreneurs must have a mission to successfully carry out the grueling work involved, and it is perfectly evident that my mission and professional football are nowhere near intersecting.

Results

Visitors:

• 1/31: 594
• 2/1: 895
• 2/2: 712

Costs:

• TwitterBowl2009.com domain name, $10

Revenue:

• Advertisements from AdBrite, $0.27
• Advertisements from AdSense, $0.01

Delicious bookmarks: five.

Blogosphere coverage:

• ReadWriteWeb Guide to Super Bowl XLIII: Geeking Out on the Big Game
• Issac Kelly: What does my app need to be successful…
• Steelers Gab: Check Out Twitter Bowl 2009 and Have Your Say About the Game!
• Super Social Bowl 2009
• Super Bowl XLIII - Touchdown Twitter
• Here We Go Steelers - “A great way to watch a running feed of football fans on twitter, commentating and trash-talking.  Yumm-o!”

Conclusion: Most Superbowl watchers weren’t simultaneously watching the Twitter Bowl. We sure weren’t.

Related Posts: 

• Twitter Bowl 2009: A Ten Day Experiment in Gonzo (part 1) (Adam’s post on the site)

EDIT: Changed “marketing” to “evangelizing.” Our product isn’t exactly marketable; it has no unique brand, doesn’t need a sustainable position, and so on. It is simply meant to entertain multitudes of users for a short time period.

EDIT: Added three more sources of blogosphere coverage.