One Lesson from Startup Weekend Chapel Hill

I had a good time at Startup Weekend Chapel Hill. It was exhausting but a worthwhile experience. Here’s one lesson that I learned.

Find Data then Write an Web Application For It

When we brainstorm ideas for creating web applications we think about what you can do to data. Like how to present it, manipulate it, rearrange it, etc.. That seems to be the logical way to go about it. We take for granted that there is data out there to use. But is there really? Where is it?

For example: messaging, IM and SMS, is experiencing a serious surge in popularity. Web sites like Twitter.com are gaining mass use and expectance. The mobile web is another big frontier being explored by web developers. When we think of new applications to build we base our decisions on what we’ve used and what is popular. This can be a good strategy because it positions your app in a highly visible place. (ex. Pownce got bought by Google after cloning Twitter.) Plus if one app is popular there must be a reason for it. So why not make something like it.

The problem with this approach is not the lack of originality its the direction with which we think about it. Lets think about the data first. What data will our website application use? Where will we get the data? How much data do we need? And most importantly HOW CAN WE CREATIVELY PRESENT THE DATA TO MAKE IT UNDERSTOOD AND USEFUL?

At the end of the Startup Weekend Chapel Hill we came to a realization that their wasn’t enough data. For people looking for a place to work you need data about those places. For someone who wants to advertise a place you need data about people who want it. I believe the core team who will take on WorkPerch.com will find the data and put it out there. Lack of data is why the site was released as a invitation beta. A wise move IMHO.

The spark that got me thinking about this was Jake’s comment that we should purchase some data to fill in the database to start with. I didn’t know there where companies that sold data like this. But it makes perfect since. Sadly I don’t think we can buy quality real estate and user data we need. That is up to the community who will use WorkPerch. They must provide this so it can be useful.

My suggestion to future Startup Weekends and web app developers in general is to brainstorm your app idea but then collect a bunch of data first. With so many people working on a project you could easily distribute the effort to find data. Thirty people could gather a ton in a few hours time.

Then the team could verify who owns the data. Is it in the public domain? Do we need to license it? How much will it cost? Next the data could be shared and merged. Once its in a common file format like xls or cvs the data could be put into a relational database. Then the structure of the web app could be determined. How will the user navigate this data (flow)? How will the web app logic parse this data and represent it? (graphs, print to screen) How will the web app users add to the data or manipulate it?

This way of looking at web apps isn’t new. But just having another angle to think and to apply I found really constructive. Thank you Startup Weekend Chapel Hill participants for creating an environment where we could learn so much.

Oh and one more thing. Chapel Hill Startup Weekend was in The Town of Carrboro. That is NOT Chapel Hill. No matter how you parse it. I don’t care that its a few feet away. You can not lump RTP and Carrboro together. You can not lump Chapel Hill and Carrboro together. You can not dismiss the creative vibe of this small Town. UNC may be next door but its Carrboro where cool companies like Blog Ads flock. So much more than semantics. Dig it!