Hindsight Bias: When Stakeholders Claim that UX Work Is “Obvious”

Usability research and UI design both suffer from hindsight bias: once a result has been created, it seems obvious.

• You study users and discover X. After presenting this new X insight to stakeholders, they claim that “Obviously, X is true.” While nice to have your findings accepted, this claim of obviousness undermines the value of your work. Why pay good money for research to prove something obvious?
• You proceed through many rounds of iterative design to create a UI for something, ending up with design Y as your suggested solution to the design problem. After presenting this new Y design to stakeholders, they claim that “Obviously, Y is the way to design this. Y is clearly a great and easy UI for this feature.” While nice to have your design accepted, this claim of obviousness undermines the value of your work. Why pay good money for a designer to spend that much time, only to end up with the obvious solution?

These two bullets are parallel writing at their finest because these two sides of the coin are the same. Good UX work — whether research or design — should result in something that’s clear, easy, and (yes) obviously the right thing.

The problem is that this outcome (again, whether research findings or UI design) is only obvious after the fact. Before doing the research, we didn’t know X. And before the designer created and discarded many inferior solutions, design Y didn’t exist in the world.

With hindsight, most UX deliverables are obvious, because they are clear and easy. But only in hindsight. Without doing the UX work, we couldn’t know.

“Hindsight is 20/20” is a classic saying for a reason — where “20/20” means “perfect vision.” (Ideogram)

The conclusion that UX work has low value because it only creates obvious deliverables is hindsight bias.

Mani Pande is head of UX Research at Cisco and one of my favorite UX experts (I have quoted her before in this newsletter on benchmarking and on writing better headlines for user research findings) has an excellent article on combating hindsight bias in UX. She suggests 4 approaches:

• Reframe the conversation: Acknowledge familiar issues upfront and provide context for recurring customer pain points. This shifts focus from obvious findings to unaddressed problems. Additionally, incorporate external data points, such as competitor information or industry trends, to enhance relevance and interest for stakeholders.
• Conduct a poll: Make presentations interactive by polling stakeholders at the beginning about key questions or predictions. This approach helps demonstrate how accurate or inaccurate their assumptions were, effectively challenging the notion that the information was already known. This technique is particularly useful for both qualitative and quantitative researchers, especially those conducting regular tracking studies.
• Contextualize research: Before presenting findings, conduct a comprehensive workshop with stakeholders to review past efforts and failures in addressing similar issues. This process helps shift perception from obvious findings to appreciating actionable insights. Additionally, relate the research to the current product roadmap, distinguishing between addressable and non-addressable issues to increase its practical value and impact on priorities.
• Assumption mapping: Engage stakeholders by documenting their assumptions and confidence levels before conducting research. Create a visual grid representing these assumptions. After completing the research, revisit the initial assumptions and compare them with the findings. Group the assumptions into validated, lower confidence, and higher confidence categories based on the data. This approach effectively illustrates how research aligns with or challenges initial expectations, moving the conversation beyond “I knew this all along.”

Common across these 4 tactics is the importance of documenting the “before” state. What did people know or suspect before hearing the research results or seeing the proposed new design solution? By increasing the salience of the “before” knowledge, we increase the value of the “after” knowledge by making the delta between the two explicit.

“Installing irrigation and planting flowers have made this area more beautiful.” Stakeholders will appreciate your excellent landscaping results more if you start out by showing a photo of how the area looked before you worked on it. (Midjourney)