---
title: Ethnographic Research Exercise
teaser: Practice ethnographic research with grocery shoppers.
tags: design,user research,product design
author: Jeff Smith
published_on: 2015-01-15
---

At thoughtbot, we've used product design sprints for over a year now to kick off
client engagements. They work well.

However, one thing we've found is that the sprint relies heavily on the client's
understanding of the user we’re building around. This can pose a bit of a
dilemma; not every client or stakeholder is a user of the product they envision
bringing into the world. And, more importantly, they're never unbiased. This
isn't a surprise: rarely do our clients have the full picture of their users.
Sometimes they really don’t know who their users really are.

To compensate for this and assumptions in general, we bake a copious amount of
validation into design sprints: the entire last day is dedicated to [user testing](https://thoughtbot.com/blog/the-product-design-sprint).
Likewise, throughout our post-sprint design process we often run weekly user tests.

Nevertheless, a gap still exists: we don't adequately immerse ourselves in the
design challenge before starting a sprint.

## Ethnographic Research

Internally, we've been playing around with a Phase 0 in our [design sprints](http://playbook.thoughtbot.com/#product-design-sprint)
to inform our designs before starting a sprint. We're sharpening specific design
tools for the earliest parts of our client engagements. Ethnographic research is
one of those tools and a place where we're actively growing.

According to [Charles Pearson](https://twitter.com/mediauras), an ethnographer
currently working with Adobe Photoshop and who we’ve worked with in the past,
ethnographic research is a method pioneered by cultural anthropology. Traditional
ethnographic principles commonly focus on immersion and a long-term engagement
with a topic or people. Often, it entailed living for roughly a year with
another people group, experiencing all four seasons with them. The goal was, and
is, to relate and engage; to develop a very deep empathy for whoever it is
you’re trying to understand. It seeks out those in-between moments, those
unscripted happenings and observations. As opposed to typical behavioral science
methodologies which might depend on self-report, ethnography can bring to the
fore truths that many might not know about themselves. This is the kind of data
that has clear value to innovative design.

A [great primer on ethnographic research](http://www.aiga.org/ethnography-primer/)
was put together by AIGA if you’re interested in learning more, and
["Just Enough Research"](http://www.amazon.com/Just-Enough-Research-Erika-Hall/dp/1937557103)
by Erika Hall is a good resource for a longer dive into user research, including
ethnographic studies.

Charles helped us think through our research methodology. His insights
significantly influenced our exercise and how we thought about ethnography as a
whole:

- Have conversations with at least five different people. This allows for enough
  data to begin developing insights.
- Talk to people casually and engage on a human level – it doesn’t need to be
  sterile. Being casual and yourself produces better results everywhere:
  relationships, friendships, and, yes, user research.
- Follow the conversation wherever it leads because that’s often where the
  nuggets are. Sometimes you want to stick to the script. Often leaving it
  behind is where the most interesting insights are found.
- Try to construct portraits of real people instead of using an abstract
  persona. Personae are great. Concrete people with the granular details that
  come from living life are even better.
- Avoid pointed questions. For instance, instead of asking "How do you store
  your photos?", consider asking them to show you how they store photos.

## Grocery Shopping

Following Charles Pearson's advice, most of the San Francisco design team took a
day to get offsite and practice.

There were several different locations and users we thought about visiting and
we used two criteria to shape where the environment we chose to work in:

- Is it easy to access? Can we go there without much preparation?
- Is it a context outside of our own where we have limited familiarity? We
  wanted to limit the ease of making assumptions about the natives we talked to.

![Location options](https://images.thoughtbot.com/ethnographic-research-exercise/options.png)

We settled on shoppers in local grocery stores, ranging from budget grocers to
the bourgeois to the granola. We went out in pairs with a single goal: have a
conversation with 5 people about their grocery shopping experience (the common
opener: "Hey, I'm a designer trying to build an app to help people better shop.
Mind if I ask some questions about your grocery shopping experience?"). How do
they shop? What do they use to keep track of what they by? Do they collaborate
during the process?

After a couple hours, we reassambled and in the office and aggregated our
insights on the whiteboard. Many were observations. Most grocery shoppers
assumed a particular posture: hunched over their cart as the clutch a slip of
paper or balanced a smartphone percariously. Other insights came from
conversation. We would never have expected one shopper to use a Dropbox file she
assembled over the week for keeping track of her groceries. We wouldn't have
considered putting that into a persona before the exercise.

## Synthesis

The process of observing and having open-ended conversations with people in
their native environment is incredible and a valuable technique.

If we actually had been creating a tool to improve the shopping experience, we
would have validated and invalidated many of our assumptions. Ethnography can be
limited to a couple hours. Try it on your next project.