An understanding acquired over time that design is common sense. That design is everywhere. That we are constantly designing whether we realise it or not.
I can point to two major sources that may have garnered this view. A lecture during my Bachelors brought up examples that a mother designs experience of a family (eg: food choices) and we design ourselves in what we wear. This lecture being a part of an introductory course to HCI, coming from a lecturer who we looked up to, I took this view for granted. Later on, Don Norman's claim that we're all designers (Norman 2005) further cemented this understanding.
This view was so ingrained that, I delivered this message in a presentation to 100+ employees of the product company I was working with (Dec 2016).
"If everyone is a designer because they change the color of their walls, then everyone is a mathematician because they count change at the grocery store.
The above comment by Bill Buxton (Buxton 2007) countering Don Norman was much brought up in HCI articles, probing the question, what is a designer?. These articles ask for rigour, design process replicability and address the need to define design. This view is something I'm coming to appreciate the more experience I gain with research and completely align with at the moment (it's November of 2025).
The field does believe that everyone is creative though (Sanders and Stappers, 2008). Hence their involvement in participatory design approaches.
RCTs and meta analyses are the gold-standard in any research for evaluating outcomes for humans. What can be more telling than for a factual finding that something works along with its extent and confidence?
Not for HCI. When in comes to evaluation of early-stage design interventions for behaviour-change, which is typical of HCI works we need to know why a design works and how can we do better (Klasnja et al, 2011). RCTs and singling out independent variables are much narrowed in scope and won't reveal much. More important contributions include understanding use, missing features, bugs, likes and frustrations of a system, their reasons etc.
RCTs can be the gold standard for lab studies focusing on button-level interactions and usability but for HCI interests as of 2025 as I understand it (i.e., creating useful technologies for human wellbeing) RCTs just can't offer much. They are expensive, require skill, and narrowed in scope albeit out of necessity.
Klasnja, P., Consolvo, S., & Pratt, W. (2011). How to evaluate technologies for health behavior change in HCI research. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '11).
Wondering...
Still wondering...