Once the team at startup studio Giant Pixel selects an idea, a few things happen.

In the first phase of design, our top-line activity is communication on the essential issue: what is the problem we are solving for users? During this clean-sheet phase, team members often struggle to express their ideas. Eliciting the best from everyone involves a lot of talking and arguing and drawing.

To help drive the conversations, our designers create visual mock ups of the product, illustrating different views and user interactions. These range from very low-resolution (whiteboard or paper sketches) to very high-fidelity (Photoshop) depending on the maturity of the concepts.

Out of this, a list of features and product areas emerges. We discuss which are required to demonstrate value to users, and which ones we can live without.

The idea gels with more rounds of discussion and mock ups. Kickoff is imminent when one of our visual designers creates detailed images that are used as part of a "blueprint" for the project lead, who owns the product. This team leader maintains the vision of what users want from the product and the business we can build around it.

We've learned that the features you don't include are just as important as the ones you do. Sobo is a broadcasting app for airing your sound bites to the world. Non-broadcasting features we considered include private messaging, reply, and chat, but we left them out...at least for now! For an experienced team that can create just about anything, the challenge is sticking to what you should do.

This keep-it-simple philosophy isn't new--Apple is famous for its minimalistic approach. But if you want to quickly build products with clarity, you must ruthlessly adhere to the minimum-viable-product principle. Come up with a first version, see how users interact with it, and then decide which changes make sense.

Sticking to priorities is critical, but you also need flexibility. We thought people could sign up for Sobo through Facebook, because this made things simple for them and us. However, early adopters of our Antenna Radio app--which curates radio content into a personalized "quick-listening" experience--objected to the Facebook-only onboarding process. Consequently, we took time to build an independent sign-on procedure, which led to a decision to let people try Sobo anonymously before providing any identity information.

For your app to resonate with a sizable market, you must keep a clear picture of what regular people--not startup people--are trying to do. Giant Pixel is very consumer oriented, even with our enterprise products. The interface should be very visual and intuitive. To optimize the user experience, we use an entertainment-industry technique and storyboard the user interaction second by second. Transitions matter.

That's phase one: Agreeing on what the product should be and visualizing it. The team now has marching orders, and begins crafting idea into product.

The Giant Pixel uses a unique approach in this second phase--our own project-management software. Now productized as Nama, the software brings a structured methodology to implementation. Tasks are prioritized and assigned to specific individuals on a rolling schedule. Discipline ships product.

Visually, Nama is a live list of who is doing what, with rich communications built in. A programmer coding a feature can open a chat with the designer and ask for a specific mock up, and the designer can attach it right in the chat window.

It's a great communication tool that also creates a detailed log. Every chat session is attached to a to-do item, preserving full context. Traditional project management tools didn't combine task management with actual collaboration. So we built one that does.

Lessons learned:

  • Set priorities and deadlines and impose them.
  • Be ruthless about the MVP rule; leave out as much as you can.
  • Keep the onboarding process both simple and acceptable.
Published on: Oct 3, 2014