Learning by Doing: An Argument for Contextual Inquiry

You wouldn’t claim to be able to cook scallops perfectly after watching one episode of Top Chef, so why in the UX industry do we claim to understand our users’ behavior after sending out ten question surveys? In addition to being surface level, surveys run the risk of returning inaccurate data from self-reporting. Using the contextual inquiry method, you can gain more accurate data through observation and the ability to ask questions, as well as much deeper feedback.

What is “contextual inquiry”? It is observing your user in their natural environment as they use your product or service.

Imagine that you run a chain of high-end denim stores New York City, and you want to know how users pick which brand of jeans to buy. Contextual Inquiry might consist of visiting a client’s home, observing as they browse the web for different styles of jeans, and then accompanying them as they drive to your store to buy a pair.

You could conduct this research yourself, but you hire a UX consultant for their experience. They begin by finding customers willing to help (they may use a recruitment firm to find consumers in your target market), and set up a time and place to meet. The way they do this will vary, but the important result is that you observe your customer as they would shop at your store. This could include visiting their home to see how they look at your online store and research other competitors, and walking through a store with the customer to observe how they shop. It might also include accompanying the user through post-purchase activities like unwrapping, first time use, providing feedback and returning items.

According to one of the leading texts on contextual inquiry (Beyer and Holzblatt “Contextual Design: Defining Customer Centered Systems”), the ideal number of subjects is fifteen to twenty users per group. However, this is probably too many for most companies without a dedicated research and design team, so Mike Kuniavsky, in his book “Observing the User Experience” suggests observing around six to eight. Even if you can’t observe that many, one or two will be worth the effort because of the breadth and depth of the results you can uncover.

By running just a few contextual inquiries, you can gain a whole host of insights unavailable to you reading survey responses. Feedback and your ability to observe are more direct. Similarly, the information is always relevant to the users’ experiences because they are the ones showing you how & why they do things. The information you get also tends to be much more detailed, which is beneficial; you may find a variety of experiences where previously you had only seen one or two.

In addition to your ability to observe, you can also ask your users questions in their own environments (remember, contextual inquiries are not pure ethnography, questions are encouraged). As these types of interviews should take place in spaces that the user finds comfortable, they are also more likely to give honest opinions.

This method is not perfect, and there is a little learning curve. Sitting behind a computer screen is comfortable, and spending a whole day with a stranger is not everyone’s forte, however, the variety of experiences you can learn about will be far more rewarding (and profitable) than the bland answers to multiple choice surveys you are used to trawling through.

Another thing to note is that this research primarily yields qualitative data; rather than trying to find hard percentages, contextual inquiries look to uncover trends and themes. While you may find interesting statistics like “five of the fifteen of the users we observed prefer the color magenta to the color blue” often times your subject pool is too small for these results to be statistically significant. However discovering these trends and investigating them further can always be beneficial.

Most importantly, this method of research is time and resource intensive, so it can’t be done at every step of the process for fifty people. However, I would argue that the results for even a few contextual inquiries will provide user insights that may never show up in other research methods, and those may prove to be invaluable.

Contextual Inquiry is a method of user experience research which uncovers deep insights into user behavior through a mixture of observation and interviews. While it can be time and resource intensive, the data quality obtained through this method far surpasses insights that can be gained from other types of research. Contextual Inquiry should be a part of every UX researcher’s (and designer’s) toolbox as it is a fantastic method of understanding and empathizing with users and therefore improving their experience.

— Sam Selbie

Student / UX Researcher / Designer

samselbie.com

Sources:

http://www.usabilitynet.org/tools/contextualinquiry.htm

http://www.uxmatters.com/mt/archives/2012/06/why-are-contextual-inquiries-so-difficult.php

Images:

http://foodnessgracious.com/wp-content/uploads/2013/09/scallops-065.jpg

http://munfitnessblog.com/wp-content/uploads/2008/08/woman-trying-out-new-jeans.jpg