Nov 28, 2021
 min read

How to identify the needs of people in panic?

If you are given a project to solve problems that take place in a disastrous situation, how would you do user research? Mainly in a case of an unpredictable disaster such as fire, flood, typhoon, or earthquake that you’ve rarely seen in your entire life. It requires different approaches to identify customers and opportunities.

I would like to introduce an interesting example where I took part in an industry project as a design researcher in 2018. This project was initiated by a multidisciplinary consortium consisting of a manufacturing company, software company, design agency, university, and public design institute based in South Korea. And it was funded by the R&D programme for a Korean governmental institution for 3 years (2018–2020).

To put it shortly about the brief of the project goal, this project set out to develop a new product-service system (PSS) equipped with IoT-based technology to help people rapidly and safely escape high-rise buildings in case of a fire. In the first year, we as a design team were charged with user research tasks such as justifying the problems, discovering people’s needs, and then creating new opportunities by designing service scenarios until the second year of prototyping.

The primary goal of user research was to understand the needs of people who were stuck in the event of a fire in high-rise buildings.

We just simply came across several challenges soon after planning user research. The challenges made us question, as follows:

1. Where to find relatable interviewees?

In the first place, we tried to map out potential customer segments within the context and look for suitable people for interviews. We defined potential interviewees with several requirements. For example, they should’ve had previous experience in the event of a fire so that we could collect practical data from those experienced people.

But imagine if you were a victim of a fire in the past and someone tried to remind you of the worst experience you’ve ever had. You know, this experience is so catastrophic that you may die or injure seriously. The more damaged you are from the disaster experience, the more likely you will refuse the interview.

Ironically, we don’t know when this exceptional experience will happen, but we should always be ready for the moment that may occur anytime.

People who had severe trauma from the disaster didn’t want to talk about their stories. In addition, it was difficult to find people who less suffered from a fire previously for interviews since the event of a fire isn’t a common accident.

2. How to characterise people in panic?

The biggest challenge was to understand what sorts of issues and needs people have within the context apart from looking at the relevant literature. In a fire situation, people tend to stick to an unexpected point, ignore the given guidelines and can’t make the right decision. All their actions and behaviours are unpredictable, which is called “Panic”.

Imagine that you are standing on the 19th floor in the 63rd-floor building. Are you confident that you won’t get panicked?

“Panicking” is one of the significant hindrances to characterising a persona in a way that many barely experience panics or shocks in a lifetime, too for me. Panic is a sudden feeling of anxiety and fear, which is strong enough to dominate or prevent reason and logical thinking. This phenomenon can be seen easily in the case that people are located in burning buildings.

Hence, we couldn’t specify those panicked people with representative characteristics without interviews with experienced people that brought about chaos for us as researchers.

How we did

We virtually failed to have interviews with people, which caused us to create personas without strong evidence. This kind of user research wasn’t familiar to us. And this kind of experience in the disaster situation contains different natures in comparison with the standard experience we normally get as follows:

  1. This experience won’t be welcome to people.
  2. Everyone is the subject of this experience.
  3. No one thinks they’ll ever have this experience. (A lack of awareness)
  4. You don’t know when this experience will occur. (Even it may not happen for the rest of your life.)
  5. The more we don’t experience this, the safer our society can be.

According to these natures of the disaster experience, we tried to break through it using different approaches.

1. More explore relevant literature and experts

Instead of interviews with customers, we more of tried to listen to professional stories by studying literature and having interviews with experts. There were some relevant papers, articles, books and blogs written by professionals who are working closely on the disaster. We concentrated on studying them for months and collecting relevant data to build a sensible persona.

Experts from different fields such as psychologists, firefighters, building managers and safety guards were subjected to interviewees, and they gave us very insightful information. The firefighter interviewee had been experiencing many accidents, major or minor. The psychologist who studies behaviour psychology related to disasters spoke for panicked people. These responses allowed us to clarify the target people closely.

2. Experiment producing a hypothetical fire situation

We planned an experiment to observe people’s emotions and behaviour by establishing a virtual fire environment. A building of 13th floors located in Busan was chosen as a place of experiment, and we visited there in advance to structure the experiment plan. We designed a scenario with eight sequential situations for subjects and prepared action cards at each location of the situations. As we studied, the eight situations were carefully selected and located depending on turning points where people’s decisions may be made. For instance, a situation when you encounter forking paths critically put you in decision making.

We attempted to plan it as real as possible despite some limitations to establish a suitable environment. We couldn’t get permission from the building manager to use all the floors for the experiment, but only the 4th floor. Also, it was very demanding to set the similar environment and situations to the real fire one. For example, the real environment in the event of a fire may be noisy, out of electricity, crowded, smoky or even collapsed. But these elements were a little beyond the tasks we could do.

It was impossible to reproduce the environment of the fire situation as it was unless we set fire to the building.
The virtual scenario with 8 sequential situations to evacuate the building

Before the experiment started, we drew some guidelines on the ground to follow the given paths and situated a decision-making board where participants could attach the action cards depending on the priority of behaviour in each situation.

We also aligned roles to each researcher; I was charged with a moderator who led the participants with our scenario, and one of us took notes by observing its circumstance, one recorded a video, and one arranged the boards and cards for the next participant after card sorting for each.

Finally, 10 subjects from different backgrounds were invited to the experiment, and the experiment was conducted one by one. In the first place, we described the given situation to each participant that they are in the event of a fire while you were working in the building and the first card sorting was begun after the alarming the bell.

Participants put cards on the board depending on their priority of actions
The experiment was taken place even in the emergency stairs

After the care sorting per each, we brought them to a private room for the in-depth interview. First, the interviewer and participant reviewed the recorded video together with the participant to remind their action, and the interviewer asked specific questions regarding their decision on card sorting for 8 different situations, such as reasons why they decided to stop moving on the way. This interview session after the experiment was helpful to dive into understanding participant’s thoughts and to feel deeply.

Of course, it wasn’t possible to set the real like environment to focus on the experiments for participants. But this experiment was meaningful to inch closer to the needs of our target people for the project. The thing is that it was our last resort to discover our target’s thoughts at this moment.


In many industries and academic fields, modelling ‘Persona’ is all the rage to shape target customers’ representing their identity. Typically, this representative persona is formed by conducting user research using different methods, such as interviews. Researchers ask interviewees to explain their previous or current experience with the service they received. These answers are used for generating customer profiles. However, as I introduced in the example above, disaster situations are cases that are difficult to identify personas in conventionally so it would be a worthwhile topic to discuss “what would you do if you are given this sort of mission?”

Subscribe to my weekly newsletter

Lorem ipsum dolor sit amet, consectetur adipiscing elit.

Thanks for joining our newsletter.
Oops! Something went wrong.