How to organize a product design process the PhD MIT Media Lab way

This is a guest post from Drew Harry, one of our instructors and PhD at MIT Media Lab.


I’ve taught the introductory design workshops at the Startup Institute Boston for the first two sessions. Because design is more about practice than about learning particular content, we focus on going through a group design process, going through understanding users, building personas, idea generation, idea selection, and concept development. Since it’s a whirlwind tour, we focus on doing more than teaching the “right” way to go about these processes. But I want to share with you some of the resources that we share with students after the introductory class as a sort of reading list for people who want to learn more about the space, and see how design professionals talk about these tasks.

Keep in mind that none of this is about right or wrong ways to run a design process. It depends a lot on what your team is trying to accomplish and how much time and money you have to spend on the process. There’s no particular reason to be dogmatic about any of this. Just be aware of the diversity of points of view and tools for doing engaging and innovative design work! Keep an eye out for opportunities to try new techniques and see if they work for you.

Methodology Overview

There are many points of view about how to organize a design process. These links try to cover some of the major ones. Nominal group technique isn’t design specific, but it’s a great framework for thinking about how to make effective decisions as a group. The Quora post summarizes some other broad methodologies that you should be aware of.


Perhaps the most obvious way to understand a user group is to conduct interviews with members of that group. Interviewing draws from historically anthropological methods of inquiry that focused on the researcher as a participant-observer who embedded with a culture and sought to build relationships with informants who could help be guides to the nature of life as part of that culture. Although companies rarely expected their designers to truly embed and become participant observers, it’s nevertheless gospel that conducting interviews with users is a central part of almost all design processes.

  • The author of this article is sort of a notorious pain, but he’s got some nice advice about what not to expect from interviews and how to get as much as you can out of them. 
  • Here’s a good example of a more elaborate interview protocol when you have specific questions you want to answer across a larger range of people
  • This is a nice interview that covers persona construction + interviews, and here are some general guidelines particularly for situations where you can talk to someone in the place they live/work.
  • Another nice breakdown of some of the different parts of context that are relevant while interviewing

Cultural Probes

Cultural probes are a way of understanding a user population. In contrast to interviewing, it is a way to empower users to capture details about their daily life that are relevant to a design process. Typically designers put together a kit with tools to help someone capture or represents their experiences in a way that they can then transfer or explain to the designer. Probes are particularly valuable because they focus less on people trying to self-summarize and represent past experiences, and more on capturing experiences in the moment. They are often a complement to interviews, and can help ground a subsequent interview with recordings of in-the-moment experiences.

Affinity Diagramming

It’s easy to rapidly generate a ton of qualitative data, but much harder to make sense of it. Affinity diagramming is the quintessential group design activity — generating a ton of post-its and shuffling them around into conceptual groups. When you see pictures of the work spaces of designers, affinity diagrams are usually the most eye-grabbing component. The core idea is to create artifacts that represent different points of view, ideas, values, or attitudes and try to find relationships between them. It’s an iterative process, and helps you understand the broader data set, identify gaps, and build a common sense of what you’ve learned.

  • Introduction, focused on managing ideas but similar approaches work with interview data
  • Photos of affinity diagrams in progress


The outputs of early work to understand a user group can be a little ephemeral. Personas are a kind of design artifact that tries to embody those lessons in representative users with specific, detailed accounts of their experiences. Personas can be a great way to ground your decision making through a design process in concrete people, not abstract senses of what a generic user might care about.


One of the paradoxes of brainstorming is that all the research says it it’s incredibly inefficient. A group of four will come up with a higher quantity and higher quality of ideas if they ideate separately rather than as a group. So why do we brainstorming? These articles suggest some different hypotheses. The reality is, most design teams ideate as a group, so it’s worth bringing to the table some techniques to help mitigate the inefficiency that brainstorming can often suffer from.

Apply to Startup Institute in New York and Boston. Summer session starts in June. The sooner you apply the better your chances of getting in.