Some UX terms have that enigmatic beauty. Bread crumbs, dark patterns, wizard… Shadowing is one of them. So when I heard it for the first time, I was instantly curious to find out what it means.

It turned out that it is not only a beautiful term, but also one of those things that make the work of product designers exciting. It even reminds me of the times when we were excited about the work of private investigators in long coats and fedora hats (it’s still quite exciting, to be honest with you).

Most of our designers don’t wear long coats and cool hats but, on the other hand, they work with some more purposeful cases than someone’s marital infidelity. But let’s see what these two professions have in common.

What is user shadowing?

Shadowing in UX research is a method of observing users as they perform their regular tasks or activities, in order to gain insight into their behaviors, needs, and pain points. The researcher, just like a private investigator, follows the user closely, either in person or remotely, and takes notes on their actions.

A private investigator, or PI for short, is like a detective that you might see in movies or TV shows. They get hired to find out information about things like cheating spouses, missing persons, or fraud. 

The difference between "shadowing" done by PI and UX researcher is that researcher doesn't have to follow people in absolute incognito. It's the opposite, a UX researcher informs them of the process and goals of the research and can ask questions. Aaand, UX researcher's job includes much less risk.

Shadowing is a qualitative research method. It must be performed in the “natural habitat” of the observed person. The researcher doesn’t invite users to the lab and ask them to perform something. Instead, the researcher goes to a place where people perform their tasks normally.

The term shadowing is also used in onboarding when new employees learn by watching seniors perform their daily tasks. These two terms have some similarities, so make sure you don’t get confused.

Although the process of user shadowing includes asking questions (or interrogation, as a true detective would say), it is different from in-depth interviews and field (or ethnographic research). Observation must be prioritized, and questions only support it.

Field research is a wider term for all kinds of research that takes place in the “field”, e.g. outside of a lab or an office.

Why and when should user shadowing be used?

Shadowing is a great UX research method because:

• It gives a UX researcher a chance to get the most objective information about users.
• The researcher can observe the real user behavior instead of playing guessing games on how they would use the solution.
• Shadowing is a great tool for developing user personas and user journeys.

Typically, shadowing would be performed at a stage when users have been using the product for a while. At this point, a researcher can come and see what their usage patterns are to confirm or disprove their hypothesis.

But, shadowing can be conducted even when the product is not yet released and nobody uses it. It is also a sign of solid design approach when you start from the problem, not from a solution or an idea. When you have defined a problem you’re trying to solve, you can go into the field to observe how people are solving it now, and what are their tasks and challenges.

Having a professional fully dedicated to UX research is a luxury that only big companies can afford. In smaller teams, the job of UX researcher is typically performed by a product designer. Since most of our clients are small-size and average-size, Eleken designers have to possess UX research skills to work on the project from A to Z.

Is it possible to do user shadowing online?

In our professional opinion, it is better to perform shadowing in real life. Still, there are situations when doing it online will also bring good results. 

If you take a look at our cases, you’ll see that most of them are related to performing regular work tasks. When we work on a tool for invoicing automatization or a CRM, online shadowing seems a logical choice. When we work on a tool for invoicing automatization or a CRM, online shadowing seems a logical choice. You can use screen sharing and record it to have materials for later study.

The danger of biases

In shadowing, the bias can be caused by both sides, researchers and users. Users can alter their behavior due to the Hawthorne effect, while researchers can get impacted by observer bias.

Hawthorne effect

The Hawthorne effect is a phenomenon when people modify their behavior because they know they are being observed, which typically results in the performance improvement. 

In the context of shadowing, it means that the presence of the shadowing researcher might alter the behavior of the person being shadowed, making it difficult to get an accurate understanding of their typical behaviors and tasks.

This principle is well-known in empirical science. When observed, people change their behavior. As it affects the results of research, scientists also invented a number of methods that help prevent bias. For example, they often use methods of “blinding” and “deception” so that participants don’t know the real purpose of the study. 

But when it comes to the UX research, it’s not always possible and sometimes the opposite technique may help, for instance, explaining the objective of the research and convincing the user that knowing their challenges are valuable to the study.

Observer bias

Observer bias refers to the tendency of the observer to see what they expect or want to see, rather than what is actually happening. This can lead to inaccurate or incomplete data and can undermine the validity and reliability of the research. In terms of UX research, when a designer that has been already working on a project, starts observing user behavior, they are more likely to see the facts that support their hypothesis, overlooking the ones that don’t.

Even seasoned scientists fall victim to this bias. There's no clear recipe for avoiding it, but being conscious is half the job. Just keep your mind open and remember that a wrong hypothesis can still bring a positive result of research as well as a confirmed hypothesis.

How to do user shadowing correctly. Tips and tricks

Biases are inevitable, and the researcher’s task is to minimize them as much as possible while making sure that the research participants are feeling comfortable and the results are meaningful. Here are some tips to do it:

• First of all, you have to make sure that the research subjects are well aware of the whole process and won’t feel awkward when being “shadowed”. Although it’s natural to feel some awkwardness… don’t let it affect the research.
• To make everyone feel comfortable, both researcher and the subject can spend some time together before the start of the research. That way they won’t feel like total strangers.
• One more danger of ruining the “shadow” effect is related to the natural desire of people to be a “good host”. As the NN group writes, it often happens when shadowing takes place at someone’s house. People would try to engage in small talk, or offer the researcher a cup of coffee… In this case, it’s better to accept this cup of coffee and then focus on research rather than ignoring the welcoming actions and insisting on starting the research right away.
• Don’t allow the focus to shift to yourself. It happens when the subject feels awkward, or is trying to help the researcher, or just wants to be nice.
• It’s OK to ask some questions during shadowing, as well as ask users to do something that you are interested in observing. But it’s also recommended to keep these interactions to a minimum and save questions for the after-session conversation.
• Think of documenting the process in advance. In most cases, taking notes is enough, though some researchers feel the need to make recordings. Be sure to do it only with the subject’s and avoid making hours-long recordings. It’s better to focus on the most important pieces.

How to analyze the results of user shadowing?

It’s important not to wait too long from observation to analysis, as some things might have become outdated over time. Even if you are the sole researcher on the project, act as if you had to present the findings to the team: make the summary and organize it in a concise way.

Get back to the answers you had prepared before the start of the research and write the answers you’ve found. If some questions remain open (or new ones pop up), it’s not a problem — shadowing is surely not the end of your research!

What’s next?

Unlike an investigator, a UX researcher’s job doesn’t finish with shadowing. Actually, it’s just the start of the exciting journey.

After analyzing the data collected during the shadowing sessions, identifying patterns and trends, and developing insights and recommendations based on these findings, you can go on with in-depth interviews. From now on, it would be easier since you already have some users to help you out with research.

Other UX research methods, such as surveys, interviews, and usability testing, can also be used to gather additional insights and validate the findings from the user shadowing sessions.

If you're interested in learning more about UX research, be sure to check out our article about our favorite research methods and how to use them to gather insights that can inform the design process. If you want to get a consultation on UX research, Eleken team will be happy to help you. Leave us a note here!

‍