I refuse to choose.

The way I see it, there are two types of web projects from the design agency / product perspective, and I refuse to choose one or the other.

The first is a single, one-off project. These kinds of projects are what most people are familiar with. It’s the website for a business, a program, a nonprofit, an entity. These types of projects generally follow a certain formula: client hires agency, agency makes website, client manages website. Now, there are obviously a lot more steps in there, and also some clients prefer to keep the agency on board for regular updates, but ideally, the client is able to manage the website in-house, and the agency moves on to the next client.

The second type is a long-term project more akin to website maintenance and management. This might be web software (think Mint.com), a publisher, a social network, a search engine. These generally have a team of people in-house who work continuously on a single digital ‘product’ to make improvements, resolve user issues, and generally maintain that product.

Before starting at The Onion, I’d primarily worked on the first type of project—the one and done kind—and you’ll see a lot of those in my portfolio because they make neat and tidy case studies. It’s easy to describe the singular problem (there was no website or the existing website was in some way lacking) and to present the solution (the new website) and thought processes that went into their creation.

When I started looking for new jobs, I was really looking for a change. I wanted a change from the grind from beginning to end without much time for reflection and no opportunity to improve ideas over time. It started to feel like I was making the same thing over and over and never really learning anything from the results of what we’d created. That was no one’s fault, it’s just the nature of churning out client work and giving it back to them. Unfortunately, it means you don’t often get the data that comes later to find out what worked, what didn’t, and how it could be improved over time.

In short, I wanted to work on something in the second category, the long-term, data-driven product work.

Working on projects at The Onion gave me more of those longer term insights I was craving. I had access to all the analytics data I could dream of (and even had time to construct click tracking systems of my own device to collect meaningful, customized data), and I was able to do real-time, ongoing user testing for potential solutions when there might be more than one possible solution to a given problem. This was an incredible opportunity and I learned a lot from it, but it turns out there was a downfall there too: most projects were never really finished.

Now that I’m a free agent, so to speak, I have the privilege to pick and choose what types of projects I want to work on. I’m still freelancing for The Onion working on some longer-term maintenance for them, and I’m also currently working on a one-off project for a lab at Yale. I guess I choose…