Design Thinking for the development of experiments

In this article I would like to explore why Design Thinking should be used as an overall philosophy for the design of experiments.

Initially, I have decided to apply the Design Thinking approach for the development of research stimuli for my PhD project for which the objective was to understand underlying mechanisms of unplanned and impulse purchases in e-commerce (see more details here). Later on, this experience has been combined with Google’s expertise and guidance on the creation of experiments for digital studies.

Overall, I believe that Design Thinking helps researchers achieve high quality research results.

Moreover, the approach that has been primarily used for product development enables researchers to conduct studies, which are highly explorative in nature, in the most effective and efficient way.

The implementation of the Design Thinking approach in the PhD project has allowed us to achieve a sufficient number of user cases to conduct statistical analysis on the differences in the underlying mechanisms of unplanned and impulse purchases.

In his book Norman (2013) highlights that Human-Centered Design principles represent the core of Design Thinking. Talking about Human-Centered Design, Norman mentions that it is “a process of ensuring that people’s needs are met, that the resulting product is understandable and usable, that it accomplishes the desired tasks, and that the experience of use is positive and enjoyable.”

Overall, Human-Centered Design is based on the interactive cycle, which consists of four principle stages – observations of the target population, ideas generation, prototyping and testing (Norman, 2013).

Adapted from Norman, 2013

Applying these stages of the interactive cycle to the area of experiments, we can provide the following outline.

The Observation stage primarily aims to identify meaningful questions based on previous research (that can be both secondary and primary).

In the case of the PhD project, during the observation stage we identified an existing gap in the academic literature and a research need to create an in-depth understanding of the consumer purchase decision-making process.

The main objective of the Ideas Generation stage is to build strong research hypotheses and develop ideas on the overall look & feel of the experiment stimuli while keeping a high level of creativity and not being limited by implementation constraints.

During this process, all ideas on the stimuli design should be questioned and sense checked.

In the frames of the PhD project, several ideas about site design and navigation were discussed and developed. All ideas on the site layout and triggers of unplanned and impulse purchases were questioned and sense checked according to the key usability principles and previously studied patterns of shopper behavior in e-commerce.

The Prototyping stage allows researchers to select and implement the most viable ideas from those generated during the previous stage.

During this stage we managed to develop a prototype of the e-commerce site that follows the key usability principles of the strong e-retailer site and incorporates the key triggers of unplanned and impulse purchases.

The key objective of the testing phase is to explore potential research stimuli and identify whether these stimuli meet the key study objectives and provide expected outcomes. 

While working on the PhD project, we evaluated whether the designed site prototype enables easy navigation for shoppers and provides a sufficient number of nudges for unplanned and impulse purchases.

Importantly, it’s been proven to be very beneficial to divide the respondents base into several groups with whom the continuous stimuli improvements can be tested.

Thus, for the PhD project we divided 27 respondents into groups of 5-9 shoppers that enabled us to make 4 site interactions.

In the book on Human-Centered Design, IDEO (2015) outlined the key benefit of the testing phase, which also stays relevant during the experiment development phase: “as human-centered designers, we adopt an interactive approach to solving problems because it makes feedback from the people we’re designing for a critical part of how a solution evolves. By continually iterating, refining, and improving our work, we put ourselves in a place where we’ll have more ideas, try a variety of approaches, unlock our creativity, and arrive more quickly at successful solutions”.

Talking about the implementation of Design Thinking for the development of experiments, we can conclude that interactions during the prototyping stage allow researchers to create strong research stimuli, which successfully address the key research objectives. 

Schneider (2017) highlights in his work: “Donald Norman, author of The Design of Everyday Things, describes a designer’s discontent with the first idea. Ask yourself, when was the last time that your first idea was your best idea? Meaning and new ideas emerge when we explore things”, and we as researchers should agree that we shouldn’t go to the experiment with the first designed stimuli. It will allow us to maximize research effectiveness, enable strong research results which can be translated into actionable insights.

Therefore, we can conclude that the Design Thinking approach should be researchers’ mindset to the development of experiments, applicable across different types of experiments like virtual store testing, advertising research and user experience studies. 

Sources: 1) IDEO.org (2015). The Field Guide to Human-Centered Design. DesignKit.  [online] Available at: http://www.designkit.org/resources/1 [Accessed 01 July 2018]; 2) Norman, D. (2013). The Design of Everyday Things. 2^nd New York: Basic Books; 3) Schneider, J. (2017). Understanding how Design Thinking, Lean and Agile Work Together. [online] Mind the Product. Available at: https://www.mindtheproduct.com/understanding -design-thinking-lean-agile-work-together/?ipi=urn%3Ali%3Apage%3Ad_flagship3_feed% 3BoOob2Lo%2F RKCF3KmEN7mCVw%3D%3D [Accessed 07 Feb 2020].

 

I would like to thank Naomi Grewal, Edward Gardiner and Nadia Sotiropoulou for their valuable contributions in the development of this article !