Two weeks ago, I finished my four month internship in the Microsoft Garage program. The unique format of this internship found teams of six interns — 5 developers and 1 UX Designer — working on an experimental project for a Microsoft sponsor, like Azure, Hololens, or Cortana. Over the course of 4 months, we were responsible for scoping our project, building it from scratch, and having fun along the way. That last part was probably the easiest one…

I had a great time with this team of geeks.

This internship had its fair share of successes, but I found myself learning more from the challenges and failures along the way. In this article, I’ll go over the mistakes I made and the lessons I learned throughout our product development cycle to hopefully save you some headache in the future. Quick disclaimer: due to confidentiality, I can’t speak about my product in detail but hope the lessons are nonetheless valuable.

Designing for new paradigms is hard — but very rewarding

In the Garage, there are many opportunities to work with cutting edge technologies — like mixed reality, artificial intelligence, or the Internet of Things. Unlike well-documented platforms like mobile or web, there are no established ways of creating a product for these new platforms so you have the freedom, and challenge, of doing that yourself.

My team was building a conversational, artificial intelligence product and it was unlike any design challenge I’ve worked on before. There are no design standards for conversation, and as such I fumbled my way through the initial iterations of the user flow. As I dived deeper into the project, I began to love the challenge of creating something new and I learned so much more than I would have if I was working within my comfort zone.

Similarly, designing for new paradigms gives you the opportunity to create an industry leading product. In other words, creating a product for mixed reality has a much higher chance of garnering attention than another mobile app in the App Store. This is a crucial consideration when launching a product or choosing a team to work with.

Identify all options and commit to one

The combination of creative freedom and tight deadlines forced our team to answer broad questions like:

  • What are we going to build to address the sponsor’s challenge?
  • Who are our target users?
  • Which features will we commit to?

There were multiple correct answers to any of these questions and that was the hard part. Using an aforementioned example, I was responsible for committing to one central conversation flow for our project. Given multiple seemingly great options for a conversation structure, I struggled to commit to one for a large portion of the internship. This soon caught up to me as our initial designs felt disjointed and unclear.

Initial product options — there’s over 100 here!

I learned from this mistake when I needed to design a logo for our product in two days. I started my ideation broad, sketching up dozens of concepts, showing them to anyone who walked by, and using their input to create more ideas. I then committed to the concept that had the most promise and spent hours going deep into the design and perfecting the details. This approach is difficult, but absolutely necessary in product development.

Listen to your user, but remember your constraints

“You should add X feature!”

Anyone who has developed a new app has heard this advice countless times. While well intentioned, it is crucial that you balance user feedback with your team’s constraints, which in our case was always time. While we would love to work on every feature that was beneficial to the user, we didn’t have the capacity in our short internship. We prioritized our features based on user needs and we passed feature requests onto our handoff team who would maintain the project when we left.

User testing was important, but so was knowing our limits.

Avoid the sunk cost fallacy

The sunk cost fallacy is a common misconception that your past investments (be they financial, emotional or otherwise) should affect your future decision making. In terms of product development, it pops up when someone says,

“But we’ve already been working on this for months, we can’t just switch now!”

Midway through our internship, we were met with a unique opportunity to work with technologies developed by a team within Bing’s research division. Working with them would mean that we scrap a large portion of the codebase and design we had been working on for months. After several discussions with our team and stakeholders, we took the plunge and agreed to work with the team, not letting sunk costs deter us from a good opportunity.

Don’t let past investment deter you from a new opportunity — this is how bad products are made!

Thanks for reading.

  • Learn more about The Garage at Microsoft here.
  • Apply for the Garage Internship here
  • Reach out here if you have any questions or just want to chat!