Can Legible Marketing Backfire?

Challenging design conventions by testing whether harder-to-read text and color saturation influence product memorability and consumer preference

Abstract image representing design elements and research

Project Overview

What if common design intuitions about font legibility are wrong? This research project challenges a fundamental assumption in marketing and design: that clearer, more legible fonts lead to better memorability and higher purchase intent.

Inspired by academic research suggesting that "disfluent" (harder-to-read) text can enhance learning by forcing readers to engage more deeply with content, I designed an experiment to test whether these findings translate to product marketing and consumer behavior.

The study examined two key design elements: disfluent text (using the specially designed Sans Forgetica font) and signaling (using color saturation to guide visual attention). By testing these elements in a controlled experiment, I sought to provide evidence-based insights for designers and marketers about whether making text deliberately harder to read might actually improve product memorability and consumer preference.

🎯

Goal

To determine whether disfluent text and visual signaling through color saturation affect product memorability and consumer preference in marketing contexts.

⏱️

Timeline

Fall-Winter 2022, completed in December 2022 as part of USC's Experimental Research Methods course.

đź§ 

Role

Lead Researcher: Designed experimental methodology, created survey instruments, developed visual stimuli, conducted statistical analysis, and authored the research paper.

🛠️

Tools & Technologies

Qualtrics (survey platform), Adobe Photoshop (image manipulation), Figma (website mockup), SPSS (statistical analysis), Univariate Analysis of Variance.

Challenge & Solution

The Challenge

Web designers and marketers often rely on untested assumptions about design elements—such as font legibility and color saturation—and their impact on consumer behavior. However, academic research in educational contexts has suggested that making text harder to read might actually improve retention and recall. The challenge was to determine whether these counterintuitive findings apply to marketing contexts.

  • Testing whether established design practices (prioritizing legibility) might actually be counterproductive for memorability
  • Translating academic findings about disfluent text from educational contexts to marketing applications
  • Designing an experiment that could isolate the effects of font style and color saturation while controlling for other variables
  • Creating a methodology that could produce actionable insights for designers and marketers

The Solution

I designed a 2Ă—2 between-subjects experimental study to test how font disfluency (through Sans Forgetica font) and signaling (through color saturation) affect product memorability and consumer preference. By carefully controlling the experimental conditions, I could isolate the effects of these specific design elements.

  • Created a controlled experiment with four conditions: regular font/regular saturation, regular font/high saturation, disfluent font/regular saturation, and disfluent font/high saturation
  • Developed a simulated e-commerce website mockup in Figma with consistent product displays across conditions
  • Used Adobe Photoshop to precisely increase saturation by 75% for the high-saturation conditions
  • Implemented a Qualtrics survey to collect data on product memorability and preference across the four experimental conditions
Standard Font (Roboto)
Example text written in standard Roboto font
Disfluent Font (Sans Forgetica)
Example text written in disfluent Sans Forgetica font

Comparison of regular Roboto font with disfluent Sans Forgetica font, designed to be harder to read and theoretically increase retention through deeper processing.

Process & Methodology

This experimental study followed a rigorous research methodology to test how disfluent text and color saturation affect product memorability and consumer preference in marketing contexts.

1

Literature Review & Hypothesis Development

I began by exploring contradictions between common design practices and academic research. According to Google's Material Design guidelines, designers should avoid hard-to-read fonts, yet research by Weissgerber (2016) found that disfluent text could enhance memory through deeper cognitive processing. I also reviewed studies on signaling by Lai and Zhang (2021), which found that visual cues like color could guide attention and enhance retention. Based on this research, I hypothesized that both disfluent text and increased color saturation would improve product memorability and potentially affect consumer preference.

2

Experimental Design & Stimulus Development

I designed a 2Ă—2 between-subjects experimental study with font type (regular Roboto vs. disfluent Sans Forgetica) and saturation level (0% vs. 75% increased) as independent variables. Using Figma, I created a mockup e-commerce website displaying backpacks with consistent product photography. Adobe Photoshop was used to create the saturation variations, ensuring precise 75% saturation increases for the high-saturation conditions.

Backpack image with regular color saturation
Regular Saturation (0%)
Backpack image with high color saturation
High Saturation (75%)
3

Survey Implementation & Data Collection

I developed a Qualtrics survey that randomly assigned participants to one of the four experimental conditions. Participants were shown screen recordings of interactions with the e-commerce website mockup, with the version determined by their assigned condition. After viewing the recording, participants were asked to identify their favorite product and recall specific product details (such as price or color), allowing me to measure both preference and memorability. I recruited 48 participants, primarily USC students and personal connections, across the four experimental conditions.

4

Data Analysis & Results Interpretation

I analyzed the data using Univariate Analysis of Variance with α=.05 to test for main effects of font type and saturation level, as well as their interaction. I removed one outlier due to nonsensical data. The analysis focused on identifying whether disfluent text and/or increased saturation significantly affected product memorability and consumer preference. I also analyzed effect sizes to understand the practical significance of any statistical findings.

Results & Impact

The study produced surprising results that challenge both intuitive design assumptions and some academic research findings, offering valuable insights for designers and marketers.

F(1, 43) = .07
Disfluent Text Effect
No significant main effect for disfluent text on product memorability or preference (p=.56, η²=.002).
F(1, 43) = .33
Signaling Effect
No significant main effect for color saturation (signaling) on memorability or preference (p=.79, η²=.008).
F(1, 43) = .08
Interaction Effect
No significant interaction effect between disfluent text and signaling (p=.77, η²=.002).

Key Findings

The study's results provide nuanced insights about design elements in marketing contexts:

  • Challenging Academic Findings: The results counter some academic literature on disfluent text, suggesting that benefits observed in educational contexts may not transfer to marketing and product presentation.
  • Challenging Design Intuitions: The results also counter common design intuitions that emphasize font clarity, as no evidence was found supporting either approach for marketing effectiveness.
  • Signaling Results: Surprisingly, even signaling through color saturation—which has more consistent support in the literature—showed negligible effects in this marketing context.

"This study set out to explore the intuitive conceptions designers have regarding fluent fonts. The present research found no evidence against these intuitions, but it also found nothing in favor of them. So there is still a lot to explore in this area, and future research is needed to uncover more results."

— From the research paper conclusion

Reflection & Learnings

This research project provided valuable experience in experimental design and challenged me to think critically about established design practices. While the results didn't support the hypotheses, the process of testing these assumptions generated important insights and avenues for future exploration.

What Worked Well

  • The 2Ă—2 experimental design effectively isolated the variables of interest (font type and saturation)
  • Using e-commerce mockups created a realistic context for examining consumer behavior
  • The methodology balanced scientific rigor with practical resource constraints

Challenges & Solutions

  • Color disparities among participants' screens were acknowledged as a limitation for saturation testing
  • Limited product variety (focusing only on backpacks) may have affected generalizability
  • Video presentation inconsistencies were identified as a potential confounding factor

Future Considerations

  • Develop custom-coded websites to track mouse movements and user interactions for more precise behavior measurement
  • Test a wider variety of product types to improve generalizability
  • Explore alternatives to disfluent text that might increase memorability without sacrificing aesthetics

Personal Takeaway

This project taught me the importance of testing design assumptions rather than relying on intuition or conventional wisdom. Though the results didn't support the hypotheses, the research process itself was valuable—demonstrating that null results can be just as informative as positive findings. I gained hands-on experience with experimental design, statistical analysis, and the complexities of studying human behavior in digital environments. The project reinforced my commitment to evidence-based design approaches and sparked my curiosity about other untested assumptions in the design and marketing world.

Download Research Paper

Interested in the complete research findings? Download the full research paper that details the methodology, results, and implications of this study.

Download Full Research Paper (PDF)