IoT Needs Qualitative User Research

There is little debate whether or not IoT is a hot and emerging field. According to General Electric, IoT will add $10 to $15 trillion to worldwide GDP growth by 2030—the equivalent of China’s entire current economy. However, due to the rapid emergence of IoT, we have to be cognizant to ensure that we avoid failed IoT applications.

While there are various causes of failures in IoT applications, the specific cause I want to discuss is not conducting qualitative user research. Not conducting qualitative user research typically results in over-engineering a product, being solutions-focused, and generally making something that nobody wants.

For example, Griffin unveiled a $100 toaster this year, the Connected Toaster, that sends you a notification when your toast is done. 

There are many things wrong with this.

The Connected Toaster seems like a classic example of slapping IoT on an application without considering user needs. The product can save your “optimal” toasting settings and notify you when your toast is done.

However, on most toasters, the dial already saves past settings and a sound notifies you of completion. People do not have the problems that this toaster wants to solve and it is apparent that Griffin did not conduct adequate user research while designing their toaster. Griffin could have simply talked to people and learned that most people do not need a “smart” toaster.

Why Should We Do Qualitative User Research?

I ask myself this question a lot. It is easy to see that qualitative user research is important from the example above, but it can be expensive to implement. I've heard all sorts of things about conducting user research:

• “People don’t know what they want – why should we even talk to them?”
• “Individual user interviews are not useful. Hard numbers and lots of data are always better.”
• “User research is a waste of time and takes way too long.”

These concerns are valid, however I propose three reasons for conducting user research:

1. Problem-focused rather than solutions-focused

It’s easy to slap IoT on any application. Seriously, just add Bluetooth and make an app. However, there is an important difference between IoT additions and legitimate IoT solutions to user problems. People may not know what the solution is, but they have firsthand experience with their problems.

2. Quantitative research is not good enough

According to Wired’s 7 Things Your Company Gets Wrong About User Research, "Quantitative methods let us test our assumptions, and qualitative research lets us find out what we don't know.” While numbers are great, they really don’t tell the whole story.

3. Long-term savings

While qualitative user research takes some upfront cost, there are many benefits in the long term because features are streamlined to only the ones users want. This results in less overall development time, higher quality features, and better alignment with user needs.

How to Conduct Qualitative User Research

There are numerous ways to conduct qualitative user research, ranging from a simple chat to a PhD level ethnographic study. I'll focus on a practical qualitative research method process based on design thinking methodologies.

Before I jump into the process, I want to stress the importance of ethics. Make sure to tell users you are using their input to develop a product and ask if they can interviewed (and recorded if you decide to do this).

The first step is simple – just talk to people.

Talking to Users

The key here is to have an open-ended interview. Do not enter an interview with any biases or leading questions. It's not your place to tell your users what their problems are with leading questions.

It is important to both prepare questions as well as leave time for open-ended discussion. For a thirty minute interview, I like to have roughly fifteen to twenty minutes of prepared questions and leave the rest of the time to really listen to what the person has to say.

I would try to really get to know the personal stories of your users. Get to know their background, what do they like to do in their free time, what their routine is, etc. Remember that in the end you are designing solutions for people, not numbers, so treat them like people. My deliverable for this section is four to five major insights about my users. Afterwards, I create a journey map.

Making a Journey Map

A journey map is one of the most powerful qualitative user research tools, and the one I find myself using most frequently. A journey map shows the status quo of a user’s process. This artifact can be as simple as a bubbles and arrows, and can be constructed from user interviews and user observations.

Most importantly the journey map shows pain points in the user journey. These pain points represent opportunities where you can design solutions. From my user insights and journey map, I write down some problem statements.

Defining a problem statement

After you have identified user pain points, it is important to write at least one problem statements to focus your solution on. An example of a problem statement is “How can we keep track of our health and exercise accurately?”. If the solution to the problem statements are IoT applications then go ahead with IoT. If it is not IoT, then you should go ahead and solve the problem with the most appropriate technology.

Not all problems require IoT solutions. However, by conducting your user research, you can create meaningful IoT solutions. Remember, qualitative research can be as easy as talking to someone, and it's definitely worth it.

There is a lot to learn in terms of qualitative research, so I have listed some useful articles and guides below in no particular order:

1. Stanford HCI Class Needfinding Tools
2. A Practical Guide to User Needs
3. Ideo.org Design Kit (Inspiration section)
4. 7 Things Your Company Gets Wrong about User Research
5. When and How to Create Customer Journey Maps
6. UX Design for IoT - 5 Important UX Design Decisions