Umar Taj

Behavioural Science is a diverse and exciting field with many career paths to explore, from shaping policy in government units to applying research in corporate settings and advancing knowledge in academia. This series features insights from professionals and academics, highlighting both the opportunities in the field and their journeys to their roles. Today, we hear from Umar Taj.

Umar is an Associate Professor of Behavioural Science. He teaches executive education at London School of Economics, University of Oxford and Warwick Business School. He is also the Co-founder/Director of Behavioural Decision Making at CogCo, an organisation helping institutions grow using behavioural science, design and data science. He previously founded the ongoing ‘Nudgeathon’ programme at Warwick Business School, which helps organisations to rapidly learn and apply concepts from the behavioural sciences to addressing their own goals.

Can you describe your work in behavioural science and decision science?

So, I'm a behavioural scientist and a decision scientist, working in two very different spaces. As a behavioural scientist I focus mostly on behavioural change, while decision science is about helping people make better big decisions, such as strategic decisions for individuals, groups, or organisations. These are the two main domains I work in. I also have a foot in academia, which I really enjoy, allowing me to do some research, though not a lot. I do a fair bit of executive education, which I enjoy a lot. Most of my time is spent putting insights from both behavioural science and decision science into practice. My passion project has been Nudgeathon, which involves helping student groups run Nudgeathon projects on university campuses whenever I get time.

What sparked your interest in behavioural science?

My interest in behavioural science was ignited during a summer school course after my first year of university. In summer schools, you've got the whole course compressed, and I think that was three weeks if I remember. That course was about decision-making for management or something similar. The content covered was more in the realm of behavioural economics. My degree already included economics, and it felt really nice because there was this psychology element. What really got me interested was the experimental bit—how some people react one way, others another, based on different interventions. The idea that you can measure things and see that an intervention has an influence systematically was fascinating. For me, it was cool because there is a science to it. That combination of factors stood out for me.

How do you apply an experimental approach to both your professional and personal life?

I think this is really important. If you want to be a good behavioural scientist, you cannot say 9 to 5, I will be a behavioural scientist, and then I will not be one. This curiosity, you can't just switch it on and off during office hours. You need to be curious all the time because that's how you start spotting things, finding opportunities, and making connections. For example, if I'm sitting in a park and see people throwing rubbish somewhere, it sparks my interest. Why are they doing that? You have to have that lens on all the time. The downside is that it can sometimes annoy people because you're always trying to assess everything or find something in it.

Liking something isn't sufficient for a career; you have to act upon things. It's important to create opportunities, which helps develop a different skill set essential for a behavioural scientist. You can't wait for things to come to you. If you have curiosity and maintain that behavioural lens all the time, you'll be more inclined to try things out.

What advice do you have for aspiring behavioural scientists regarding maintaining that lens and staying updated in the field?

At CogCo, we have an internal daily newsletter shared early in the morning with everyone, stating what papers were published yesterday. We have a list of journals we are interested in. This is very important. You can't just say, I'm done, I have a degree, I have a job, so now I can relax. In behavioural science, as in other fields, you need to be in a position where you are always up to date on what’s going on in the field. I can't teach the same tax letter trial from 10 years ago because it's outdated. That's where inspiration comes in for me. New papers are coming out, and my big advice is to go quickly to the method section and the design section. Even if the topic isn't interesting, how did they design the experiment? What was the control? How did they measure? Extract those details. I'm OK if someone doesn't read the whole paper, but you need to keep exposing yourself to this and let your brain do its stuff. It's a constant thing; you can't just say now I know and stop. It's either up or down; there isn't a flat line. My daily routine involves reading journal articles, especially those where I'm curious about the study methods and measurements. The rest of the day is mostly meetings.

Can you share an example of creating an opportunity in your career?

Nudgeathon is not experimental in nature, but it demonstrates moving from thinking to doing. When you're studying behavioural science, you're primarily learning theory without many opportunities to put it into practice. Having theoretical knowledge of behavioural science is not sufficient, you gain a lot from actually putting the insights into practice. I felt frustrated because you could complete a degree without ever applying the knowledge. When entering the job market, students often struggle to answer practical questions because they've never worked on an applied project.

So, I thought, what if we created a behaviour change challenge similar to hackathons and offered it to students? I presented this idea to Daniel Read at Warwick, who immediately supported it and said, give it a try. We had a small organising team, and we decided to go for it. This is a great example of moving from thinking about something to actually trying it.

A fun example of an experimental mindset from my university time involved toilets with light switches inside. I noticed that sometimes the light would be on, and sometimes it would be off. I wondered if entering a toilet with the light off would make someone more likely to turn it off when they left, versus if they entered with the light already on. I decided to test this by switching the light on or off before someone entered and then observing their behaviour. Though I wasn't formally recording data, this wasn’t part of any assignment, it was simply an interesting exercise in curiosity and experimentation.

The key takeaway is to just do it. Often, we worry too much about the details. Just give it a try and don't worry about it too much.

What skills are key for a behavioural scientist?

I often ask students what they think are the most important competencies or skills to be a good behavioural scientist. Most often, I get responses centred around knowledge, such as learning new theories and techniques. I don't think that's enough. One of the most important skills is the ability to sell behavioural science. This includes project management and stakeholder management. Beyond knowledge, these skills are crucial.

If you join an organisation, you'll probably gain the necessary knowledge, but being able to manage projects and stakeholders effectively is what sets you apart. Running a behavioural science project is not just about knowing models or theories; you have to actually sell behavioural science. This is why I encourage students to start doing something while they are studying. For example, if they come up with an idea to reduce late submission of books in a library, they need to sell it to the library administration to get permission to run an intervention. Going through this process is very important because they have to explain what behavioural science is, convince others of its value, and then set up and manage the project.

These experiences are crucial because when you enter the applied space, you still need to convince people of the value of behavioural science. Developing these skills beforehand is very helpful. When you take initiative and try things on your own, you develop these essential skills, which a degree alone doesn't provide.

What advice do you have for aspiring behavioural scientists in terms of gaining practical experience?

My strong advice is to actively look for opportunities where you can introduce some form of intervention. Go to your corner shop, have a chat with them, and ask if there is something they want their customers to do that isn't happening. Figure out what they are interested in and see if you can help. It's fine not to have a perfect sample or be very robust about it. The whole idea is to get someone to share a behavioural challenge with you. If you simply ask whether they have any behavioural changes they want in their customers, you won't get a good response. You have to extract a behavioural challenge from them.

If you're a student, consider the area from where you live to your campus. There will be opportunities along the way. Think about those and start making small changes to see their effects. This approach is very important.

Quite often, you'll be surprised by the proportion of positive responses you get because you're helping someone achieve something they also want. This increases the chances of someone saying, "Yes, I'll give it a try. Tell me what to do." Then you can suggest something you're working on as well, which is a very useful skill.

What are some of the best ways to make a difference within or using behavioural science?

I do feel quite strongly about this. There's enough evidence to show the positive impact of behavioural science. The positive impact, if you measure it, is always from where you were to where you got after your intervention. In many instances, in places outside of the WEIRD countries, the baseline is actually really low, which means the opportunity to create an impact is really high.

One way to create impact is to look for opportunities where the status quo or control is really low, giving you a much bigger room for improvement.

It's natural for us to go for something easy, like working on problems with already good data available because it's less challenging. For example, working on communications is easier because making one change can have widespread effects. However, I encourage people at the start of their careers to get into spaces that are not easy.

I remember working on a vaccination project where the first step was to create a data pipeline to record vaccinations. We started with pen and paper, needing a way to get all those records in one place and then transcribed. This isn't typically the role of a behavioural scientist, but in that environment, if an intervention made a difference, the impact would be much greater due to the low control.

Think of your role as more than just a behavioural scientist, but as a problem solver. I ended up solving the problem of data collection, even though it wasn't strictly behavioural science. Having a mindset focused on addressing problems is beneficial, especially early in your career. If you are impact-driven and want to improve people's lives, this approach helps you identify the spaces to work in.

Many "wicked problems" have structural challenges, but definitely also have behavioural challenges. We often shy away from these because they are harder and require stepping outside the typical behavioural scientist role to do other things as well.

If you could work on any project, what would it be?

I think it's much easier to answer now because I've been grappling with this. As I mentioned earlier, I work in two spaces: behavioural science and decision science. One focuses on behaviour change, and the other on big decision making. I believe that helping people make better big decisions can have greater impact than behaviour change work. If I could do anything, I would like to work on helping people make better big decisions.

These big decisions include career choices, having children, deciding where to live—strategic decisions that people make. I'm baffled by the way people make these big decisions. If people want to improve the quality of their decision making, that's the space I would like to work in—focusing on big decisions, which are not many in one's life, but have significant impact, and helping the masses with them.

If there’s one piece of advice you could give to people in their early careers today, what would it be?

Maybe I'll go with a couple. I think I shared already: one, I don't think you can be a 9 to 5 behavioural scientist. If you really want to excel, you can't just treat behavioural science as a job; it's a way of thinking, a way of solving problems. You can't just switch it on and off. You have to have that curiosity all the time and keep looking for inspirations. So that's one piece of advice: don't think of being a behavioural scientist as a 9 to 5 profession. It's about how we should be solving problems.

The second piece of advice, which I've mentioned before, is to just try experimenting. Give it a try, find opportunities yourself, and create them. Don't worry about the robustness of things; the skill you’re learning is the creation of opportunities.

Thank you so much, Umar, for sharing your insights with us today