5 cognitive traps to avoid in Discovery

As more of our projects move through Discovery and beyond, we’re gaining valuable knowledge that we’re looking forward to sharing with other government agencies. Our intention is that our experiences will benefit other agencies as they move through the same processes, making the path easier and more efficient for all. Fabio Pereira, a Transformation Manager at the DTO, explains why we need to be aware of our own biases and how they can affect research.

Five coloured cards of cognitive effects. Confirmation Bias, Anchoring Bias, Overconfidence bias, Social desirability and Blind-spot bias.

Leisa Reichelt, Head of Service Design at DTO, pointed out in her ‘What is Discovery?’ post that ‘the Discovery stage is one of the most important – but probably the least understood’ in our four-stage service design and delivery process.

We know that Discovery is for discovering, not validating. This means we have to be authentic in our approach to learning through Discovery, not just look to have our existing ideas confirmed.

Part of my role as a Transformation Manager is to raise awareness of these pre-existing ideas or cognitive biases that can impact on our results.

In order to steer teams more effectively towards building simpler, clearer, faster public services, we need to be aware of these biases and how to overcome them.

So, to begin with, what are cognitive biases?

As human beings, we think we make rational decisions every day, but in fact, we’re all seeing the world under a set of behavioral illusions that can really muck up our decision making. These are called cognitive biases. They are sort of like optical illusions, but instead of interfering with the way we see things, they get in the way of our behavior when we make decisions and solve problems. Being aware of these biases can help us make better decisions.

Below, I’ve put together a list of some examples of cognitive biases, that can affect the Discovery phase:

Confirmation Bias

What is it?

Our tendency to focus on information that confirms our assumptions, rather than letting the data in front of us speak for itself.


During research, we might inadvertently ignore important facts and ‘pain points’ that users are raising because they don’t fit with our existing assumptions.

How to avoid

Be very careful not to build preconceptions of what users need, before you see the research. Listen to them with an open mind for anything possible. Don’t validate, discover!


What is it?

Our tendency to rely too heavily, or ‘anchor’ our thinking to, the first thing that we hear and, therefore, from that point on, give less importance to what comes next.


Information we receive before going out and doing user research can anchor us, which might lead to confirmation bias during the research insights consolidation phase.

How to avoid

Be especially mindful that things you learn before, during and after user research should have equal weight, especially the things mentioned by the users, who we’re here to serve.

Overconfidence effect

What is it?

Our tendency to, sometimes, be over convinced that “we’re right”.


This bias can affect users who are being interviewed. During Discovery, when playing the user research sport, the team can be led to believe that one thing a user said is the most important thing and applies to everyone. The team can also be too confident about Discovery. Can we really interview 100 people in 1 week?

How to avoid

Confirming information with multiple users; not fully relying on one person’s answers when they claim that something is their biggest pain point and they are completely sure about it.

Social Desirability Bias

What is it?

Our tendency to make more ’socially acceptable’ decisions when in the presence of someone else, which differ from the decisions we might make when left alone or acting independently.


Be aware that users being interviewed in your research might not behave the same way that they would otherwise, if they were elsewhere, not being observed or not in a group situation.

How to avoid

As much as possible, observe users in their real environments, under the real conditions they will be using a service. This maximises the likelihood that they will behave how they usually do when not being ‘researched’.

Blind-spot bias

What is it?

Our tendency to lack awareness of our own biases, and think, ‘these biases affect other people, not me’. If this thought has crossed your mind while reading this, chances are, you’ve just experienced the blind-spot bias.


Not being aware of our own biases might lead us to fall into a few traps, including those outlined here in this article.

The key is understanding

Ultimately, the key is to understand the cognitive biases that might be at play in the Discovery phase; just being aware of these - and taking practical steps to address them - can help us make better decisions.

Cognitive biases are often studied in psychology and behavioral economics. Other areas of Government are also using behavioural economics. The federal government, for example, has recently established a new behavioural economics team in the Department of the Prime Minister and Cabinet.

Image credits: Image icons created by iconoci, Yuvika Koul, Shreya Chakravarty, Eric Bird, Dmitry Baranovskiy from Noun Project.