Jānis Ķesa

It is not just service designers who are prone to confirmation design, the majority of humans are. Well, at least to some degree. When reading this post it is likely you will have a chance to recall a case or few where your own thinking and therefore decisions have fallen victim to the confirmation bias. 

Confirmation bias is probably one that is easiest to explain and comes first to mind when talking about cognitive “errors”. Not a surprise as it is well-researched and several popular books are touching on the subject.

What is it then? The tendency to search for, interpret, favour, and recall information in a way that confirms one’s preexisting beliefs or hypotheses. This simple mistake we make innocently almost all the time can lead service designers to overlook contradictory information or fail to explore alternative perspectives that in order may turn out to be crucial for the best design outcome.

When looking at Service Design flow through the lens of “Double Diamond”, confirmation bias is the most dangerous at the Problem part of the process, when research is done (discovery) and the problem is defined. 

Most of the time we have our set of beliefs ready to build a case on them, and even if we don’t have preexisting ones beliefs can be formed quickly which is sort of a natural response of our brain to help us be more efficient and make decisions faster. Service Design is more about accuracy than speed unless we are looking at some of its methods taken out of context (like sprints and other co-creation tools). To stay as accurate and true to the problem it is important to stay as objective and neutral as possible.

Imagine being a part of a service design team that is hired to work on the improvement of the public transport system in some city (hypothetical example). You and your peers all use the transport system and for you, it is clear that the most important part of that service is reliability in terms of timing. You believe that this is the main metric to evaluate the system as all of you have experienced how quickly your day is ruined if you have to reschedule your calendar for the sake of a late commute. The whole research process is designed accordingly. You carry out interviews and launch surveys where a disproportionate amount of questions are asked about late buses, effectiveness of schedules, time saving etc., while barely touching on other potentially important aspects of public transport. Your desk research is mostly about time-saving and reliability of schedules. And… you have not noticed that there is also a concern about safety on the public transport rising. After all, you have never felt unsafe during your trips, just as you have not encountered people with disabilities on your morning commute – another issue you are not paying enough attention to, as you are already under the spell of confirmation bias, receiving answers that confirm your beliefs, being busy finding more evidence that this is the real problem. At the end of the project an app that predicts arrival times of the buses is launched, it is cool, and it serves people like you. However.. the overall quality of service has not improved very much as you have neglected the need for security and accessibility already at the discovery phase. You have fallen for the confirmation bias!

Confirmation bias is often triggered when an article or another authoritative publication by a star designer or researcher is supporting your beliefs thus reinforcing them. This is extremely easy to follow such a path as you are not making any mistakes, you can always quote the publication of that other expert.

What can we do about it? How to mitigate the risks? Surely enough it is easier said than done but here’s my advice (re-read the paragraph above). Develop critical thinking skills and try looking at the issue from as many viewpoints as possible. Gather feedback, and listen carefully! Another approach – develop a system for data analysis that can minimise your own errors caused by confirmation bias and apply this same data evaluation method for all your projects to identify suspicious mismatching patterns.

In conclusion: We all make mistakes but understanding and addressing confirmation bias in service design is crucial for creating services that genuinely meet users’ needs and foster innovation. By recognizing and actively minimising risks of these biases, designers can ensure their solutions are both user-centred and grounded in reality.

P.S. There are several forms and shapes of Confirmation bias that I am hoping to address in some later articles.