Customer research is a double-edged sword. On the good side, it informs the organization about what is valuable to their users; on the bad side, when user research is not accurate, it fills businesses with a false sense of confidence. Misleading assumptions can send them on a wild goose chase to build products that nobody needs.
I will run down the most important steps to prepare for great user interviews that are the result of proper planning—whether they are in-person or remote. In fact, the user research methodology is optimized for remote situations, and many product teams in top companies routinely conduct remote interviews in order to save time and money.
Talking vs. Listening
Henry Ford’s famous quote - “If I had listened to my customers, they would have asked for faster horses” - highlights the risk of poor customer interviews. By talking to customers, Ford may have been led down the path of horse training and steroids to create “a faster horse.”
By listening, however, and asking, “Why do you want a faster horse?” Ford would have heard things like, “I want to get from A to B faster”; “I want to travel in more comfort”; or “I want something that conveys my status.” In other words, by listening, he would have heard the early demands for sports cars, family cars, and luxury cars.
Lead with the Need
When asked about how to build and launch great products, Steve Jobs said: “Lead with the need.” Well-thought-out customer research tells a product team the user’s needs. Poor customer research, on the other hand, yields inessential data, such as what product features customers consider valuable or undesirable, but not why. This is a fine line, but a deep understanding of customers’ needs is the vital difference between research that will enhance projects and research that will lead you down the wrong path.
Test the Assumptions First
The best way to gain valuable user insights is to shift the focus from a feature toward your assumptions about users’ preferences, and test those first of all. This is hard because typically, product team members have prior assumptions about users’ preferences that they take for granted. Eliminating wrong assumptions and replacing them with authentic ones is like building the proper foundation to hold up the product. Therefore, proper user research consists of two steps: testing the assumptions about the users, and once validated, moving ahead with the product or feature testing. I usually see product teams skipping the first step and focusing on features rather than assumptions about users’ needs, and thus diminishing user research quality.
Let’s explore an example. A product team is building an app with a messaging feature. The team assumes that the product needs a messaging system and decides to test multiple versions of how it functions. As they conduct the user interviews, they determine which messaging systems are preferable and ultimately decide about the design and functionality of the messaging feature. The basic assumption that a messaging system has value for a user affects the research and leads to misleading insights. While a user may prefer a certain type of messaging system over another, there is no validation that a messaging system is required.
I have seen this scenario happen many times—a client wants to include a certain feature in the design and approaches me to optimize that solution. However, it turns out that the client has not validated the feature premise, meaning that a decision to include the feature is based on an assumption rather than the real users’ needs.
The Four-step Formula
After the product team validates its assumptions, the next step is to follow a simple four-step formula that focuses on a deep understanding of the customer. I adapted it from the five whys method, which is widely used in user research and suggests asking “why” someone said something five times in order to get the true answer. However, in my experience, this method can result in broad conversations with no meaningful user insights.
Conversely, the four-step formula limits the conversation to the product being tested and increases research efficiency. It asks users four simple questions:
- What do you think of this feature?
- Why do you say that?
- Is it a valuable feature?
- Why is it valuable or not valuable?
At a time when there was an industry trend to incorporate technology into what was previously a “dumb” product, I worked with a team launching a new retail hardware product. Every new player on the market—and there were a lot of them—was introducing products with incredibly smart features, and they were well-received by an eager early-adopter market. I can not disclose the exact product due to non-disclosure agreements, but let’s say it was a coffee machine since that makes a good example.
The company that hired me was tempted to explore the best ways to incorporate smart technology into our product. I conducted some testing to understand where we might be able to build smart features into our product. I checked our premise to make sure that I wasn’t testing which “smart features” were valuable, but rather the entire premise of whether or not smart features were desirable in a coffee machine, and why:
What do you think of this coffee machine?
When I presented some smart features to interviewees, I heard comments like “cool” and “oh wow, that’s awesome”—things were looking promising. However, I was just talking to my customers at this stage, and I needed to start listening.
Why do you say that?
When I asked: “Why did you say that it is cool, awesome, bad, or strange?” the majority of users replied: “I like it because it’s unique.”
Is it a valuable thing?
This is where things became interesting. When I asked: “Would that be a valuable feature for a coffee machine?” the most common response was: “Well, no.” These findings were very interesting: I had a full 180º flip on the perceived value of a feature. It was time to nail this interview down to understand what was happening.
Why is it valuable or not valuable?
The majority of answers were like: “I actually don’t want a smart feature on my coffee machine because it will become dated faster than I would expect a coffee machine to.” This was groundbreaking. Customers were telling me that they overwhelmingly did not want to introduce smart technology into the product because it would become obsolete faster, and that tradeoff would decrease the perceived value of the product.
The implications of this research were huge for the client, as the company decided not to include smart features in its product, which saved them from a potential disaster at launch. The smart features trend has come and gone for the client’s industry, and many companies that specialized in smart products in this category no longer exist, despite the significant global hype at the time. The “faster horse” for this industry field was integrated smart technology, and the product management team was able to sensibly explore this finding by asking four simple questions.
How to Conduct Remote Interviews
In my opinion, remote interviews are more informative to the researcher than face-to-face interviews. People are much better at building rapport with each other through body language, social chitchat, and validating gestures or agreement than through listening. My best research happened when I did not see the interviewee and kept the discussion limited to the project context. When you are in a room observing social norms, it is more difficult to focus on an interview, as research and rapport do not go together.
Therefore, even without the need for social distancing, I am an advocate of remote user research. The essential requirement for virtual meetings is properly functioning technology for both the interviewer and interviewee. Below, I run through the methodology and tools that I have implemented with clients all over the world to conduct effective remote user research.
The Three Crucial Tools
This is a free software tool that many people are familiar with. Interviewees can join the meeting from a browser, thus they do not need to install the software. It also allows a host to record an interview for playback later. Zoom doubles as a calendar system—once you schedule an interview, a link is automatically sent to the interviewee with instructions on how to join. I recommend scheduling an interview for 45-60 minutes while taking into account that Zoom’s free version has a limit of 40 minutes per call.
Let the interviewee know that their video should be turned off and explain the reason behind it: You will focus on listening and taking notes and will not be able to look at the interviewee’s face.
Keeping notes in Excel or Google Sheets enables you to easily follow the recommended four-step questioning methodology. The structure also mitigates bias, with its clear limitations and focus on the user’s perspective. My usual sample size for the discovery stage is 15 interviews per customer segment (user personas). Exceeding that number may result in wasted time and energy.
3. Stimuli to Test a Hypothesis
Effective remote research is centered around presenting stimuli to an interviewee. They can be renders, mockups, wireframes, concepts, or imagined scenarios that you can create yourself. Stimuli from competitors—such as competitive products, web pages, designs, and brands—are another choice. If you lack images or mockups, create a series of statements with a testing hypothesis (left image below). I usually prepare nine stimuli per interviewee. Either number or name the images so you can remember the order. Make sure that you keep the same sequence when you present stimuli to each interviewee.
Why This Drink?
A product team’s impressions about users can be deceptive. A bartender at a local bar is confident that he knows the regular visitors’ usual drinks, but does he know if it’s the taste, the price, or the social status that makes them order that drink? Tested user insights are the foundation of great products because from day one, they provide product teams with a focus on the problem—the user’s needs. By developing a step-by-step approach to conduct remote user interviews, product managers can understand the reasoning behind the problem in order to optimize a product roadmap.
Bonus tip: Recruiting interviewees for user research can result in costly overhead. I have found Facebook and LinkedIn advertising helpful and cost-effective for recruiting interviewees. I recommend checking out this source, which helped me to master the skill.
Understanding the basics
Customer interview is a structured conversation with users or customers in order to understand their needs and real-life problems. They can be conducted face to face or remotely on a video-conferencing platform. Top companies choose remote interviews as a solution to save time and money and gain valuable users’ insights.
A user interview informs product teams about what is valuable to their users and leads them to build products and features that solve actual problems. A deep understanding of customers’ needs is the vital difference between research that will enhance projects and research that will lead you down the wrong path.
User interviews fail if they are not well-prepared. The biggest portion of preparation is to conceptualize the user research. Make sure you start with testing your assumptions about the users, and only after they are validated can you move ahead with the product or feature testing. Next, you should create a hypothesis to test their premise. Finally, create a series of stimuli to show for the interviewee. For remote interviews, double-check if the interviewee is aware of the technical requirements, such as a stable internet connection.
The recommended length to interview one user is 45-60 minutes. For a remote interview, take into account that some free video-conferencing tools have certain limits for one call. For example, Zoom’s free version has a limit of 40 minutes per call.
Customer discovery interview is the process of talking to potential customers with a goal to discover if a business plan will turn into a profitable business. These interviews are conducted before developing a product or service in order to determine if there are actual customers who will use it.