Shadowing
Shadowing is a qualitative research technique conducted on a small scale where the researcher acts as an observer. It is best performed as a base for quantitative research to provide statistically significant insights that can be further validated with further research.
In Shadowing, design researchers observe real-life situations of a research subject or a participant for a set period of time. For this set period, the researcher does not interfere with the participant to avoid the research subject to deviating from their natural behavior under the scenario or circumstance. It can be frustrating to the researcher, as they sometimes need to probe for more detail based on what they learn in shadowing exercises. But, at the same time, Shadowing can help validate the participant’s journey. Shadowing also gives a quick insight into the design context for an audience, quickly and accurately. Shadowing can be done for as short as 30 minutes to a few weeks or months depending on what the design researcher wishes to learn from the exercise.
Shadowing is a useful behavioral observation tool where users can be observed in their natural environment. However, researchers shouldn’t make assumptions based on the findings of single isolated observations. Additionally, results should not be considered the representation of the whole population either.
Generally, there are three types of Shadowing: natural (no-interference) – where the design researcher only observes the research subject for a set period without interference; controlled – where the researcher designs a task and observes it being carried out; and participatory – where the researcher performs the activity being observed to gain a firsthand perspective. The descriptions of each of these methods are briefly explained in the following table with their purpose, advantages and disadvantages.
| Method | Purpose | Advantages | Disadvantages |
| Natural (no-interference) | To observe the natural way of performing a specific activity | Lends the users more flexibility. Chances of identifying criticalities are higher | Different research subjects can perform the same activity differently which can add complexity when consolidating different journeys into one |
| Controlled | Get Insight into a specific journey | It is an observational probing technique for the researcher as they can control the steps to follow when performing an activity | The users have lesser flexibility in performing the activity. Those users who do not perform the activity in the manner designed by the researcher may not contribute significantly to research if the outlined steps feel clumsy |
| Participatory | Gain a firsthand perspective | Gives the researcher a chance to see things from the users perspective Can lead to developing greater empathy as well as a great rapport between the researcher and participant | May affect the natural journey of the participant. |
Advantages of Shadowing
01 Real-time data collection
Shadowing is a very effective and real-time data collection method in research.
02 Rich behavioral insight
As the researcher observes user behavior firsthand or experiences it, it leads to rich and meaningful insights, which would have otherwise gone undiscovered
03 User-centered
A greater stress is laid on the user than the product or service
04 Greater Empathy
Can create a greater empathy for the user in the researcher
Disadvantages of Shadowing
01 Time-consuming
As individual participants perform or simulate a trail, the time taken for shadowing is high per user.
02 Complex Analysis
As all users may have their own way of performing an action, the data collection, organization, and specifical analysis is more complex than other methods
Think Design recommendation
Shadowing is the right method to use when you want to observe an individual and document observations in “as it was happening” format. This method is devoid of probing and you would use this when you don’t want to come in the way of the participant and his natural behavior.
Shadowing leads to a collection of information which should be further investigated and synthesized. Do not use this method when your objective is validation.