The Human Element: Using Research And Psychology To Elevate Data Storytelling — Smashing Magazine

Why Data Stories Fail — and What To Do About It

Data storytelling is a powerful tool for UX professionals, blending analysis with narrative to explain design decisions, advocate for user-centered improvements, and create persuasive presentations. Done well, it simplifies complex information and compels action, translating into stronger buy-in for research initiatives and greater alignment across teams.

Examples like The New York Times’ Snow Fall and The Guardian’s The Counted demonstrate how data stories can leave lasting impressions by humanizing statistics through interactive visuals and compelling narrative.

What Makes Data Storytelling Effective

At its core, data storytelling serves three primary functions:

• Simplifies complexity — Makes data understandable and actionable.
• Engages and persuades — Creates both emotional and cognitive engagement so audiences feel compelled to act.
• Bridges gaps — Connects raw information with human experience, making data relevant and relatable.

Familiar narrative frameworks provide a useful starting point. The hero's journey (as outlined in Vogler's 1992 text) or the Freytag pyramid give structure, helping storytellers shape a beginning, rising action, climax, falling action, and resolution. Data visualization techniques — interactive charts, maps, and infographics, as covered by Cairo’s The Truthful Art (2016) — are equally critical for transforming raw numbers into digestible visual forms.

Going Beyond the Basics

More sophisticated narrative techniques can deepen the impact:

• The three-act structure — Divides the story into setup, confrontation, and resolution, helping build context, present the problem, and offer a solution (Few, 2005).
• The hero’s journey (data edition) — Frames the audience or decision-maker as the hero who must overcome a problem. The data itself becomes the journey, revealing challenges and insights before arriving at resolution.

Example: Presenting declining user engagement data could follow the hero's journey. The "call to adventure" is the declining engagement; "challenges" emerge from data showing drop-off points; "insights" come from analysis that reveals root causes; the "resolution" is the data-backed solution the audience, as hero, can implement.

The Limits of Traditional Models

Many widely used models for data storrytelling follow a linear path: selecting data, tailoring to audience, storyboarding visuals, and issuing a call to action. While structured and technically correct, these models often fall short because they prioritize presentation over experience, neglecting the fact that different people perceive and process information differently. This gap weakens their overall effectiveness.

Key problems with these approaches include:

• Cognitive overload — Providing too much data without context or narrative confuses rather than enlightens. This can be especially difficult for individuals who need information in smaller, more digestible portions.
• Emotional disconnect — Heavy data without an emotional hook fails to drive engagement. People remember information that resonates with their feelings and values.
• Lack of personalization — A one-size-fits-all story misses its mark. What works for a CEO is unlikely to persuade frontline staff.
• Over-reliance on visuals — Charts and graphs can’t stand alone. Without narrative context, visuals are insufficient and may not be accessible to all.

Two Critical Fixes

Traditional models improve significantly by prioritizing two components:

Audience understanding. Too little focus is placed on who the audience is, what they need, and how they process information. Without this, data stories become irrelevant or even misleading. Effective storytelling requires knowledge of the audience's demographics, psychographics, prior knowledge, beliefs, attitudes, and motivations.

Psychological principles. Insights from psychology clarify how people process information and make decisions. Without these, even a beautiful design can fall flat. When audiences feel understood, they are more likely to be persuaded; data-driven stories that speak to the heart and mind of the listener are more likely to drive action.

Applying the Theory of Planned Behavior

All storytelling is persuasion, intended or not. To be convincing, data storytellers need a theoretical framework to identify and measure the psychological factors that govern their audience's response.

Drawn from research by Ajzen (1991), the Theory of Planned Behavior (TPB) is widely cited for predicting and explaining human behavior. It consists of three elements:

1. Attitude — A person’s favorable or unfavorable evaluation of a behavior. Someone who believes exercise benefits their health will hold a favorable attitude toward working out regularly.
2. Subjective norms — Perceived social pressure to perform, or refrain from, a behavior. In the exercise example, that means what people think their peers, family, online communities, or wider society would say about the importance of physical activity.
3. Perceived behavioral control — How easy or difficult the behavior appears. This is about whether someone believes they have the time, equipment, physical ability, and support necessary to act.

These three components combine to form behavioral intentions, which serve as a proxy for actual behavior based on what can be realistically measured in a research context.

UX researchers and data storytellers should build a working grasp of TPB, or a comparable theory, before attempting to assess audience attitudes, norms, and perceived control. Doing so creates a sound foundation for narrative strategies that are grounded in how people actually decide and act.

From Audience Insights to Persuasive Narratives

Understanding your audience and applying psychological principles are the twin foundations of effective data storytelling. The question is how to put them into practice. The Audience Research Informed Data Storytelling Model (ARIDSM) offers a structured five-step approach for integrating UX research and psychology into narrative creation.

Step 1: Define Clear Objectives

Before working with data, decide what the story should accomplish. Are you informing, persuading, or driving action? What specific takeaway matters most?

Clear objectives act as a roadmap, keeping data, narrative, and visuals aligned. Without them, a story becomes unfocused. Frame objectives with action verbs and measurable outcomes. Instead of "raise awareness about climate change," aim to "persuade 20% of the audience to adopt one sustainable practice." For example, a data story about employee burnout might target convincing management to implement work-life balance policies, with a goal of reducing reported cases by 15% within six months.

Step 2: Conduct UX Research

Audience research gathers insights into demographics, needs, motivations, pain points, and information consumption preferences. Understanding these factors lets you tailor the narrative to capture attention and ensure comprehension. Methods include surveys, interviews, persona development, and message testing with potential audience members. If your story encourages healthy eating among college students, survey students to identify prevailing attitudes toward specific healthy foods and use those findings in your narrative.

Step 3: Analyze and Select Relevant Data

Data analysis bridges raw information and meaningful insight. Exploring patterns and trends grounds the story in evidence, adding credibility and persuasive weight. Clean and organize data first, then identify key variables and metrics. The psychological principle guiding your research determines which variables matter. With the Theory of Planned Behavior (TPB), examine how you measured social norms to understand audience perceptions, then frame calls to action around those norms. Statistical methods like factor analysis can group respondents with similar traits; t-tests can compare group averages; correlations can reveal directional relationships between survey items.

Step 4: Apply the Theory of Planned Behavior

The Theory of Planned Behavior posits that intentions — the strongest predictors of action — are shaped by three components: attitudes, subjective norms, and perceived behavioral control. Integrating these into a narrative enhances its persuasive power.

• Influence attitudes: Present evidence highlighting positive consequences of the desired behavior, framed as beneficial and aligned with audience values. Knowing the audience already has a positive attitude toward exercise lets you emphasize holistic life improvements and identity rather than basic physical benefits.
• Shape subjective norms: Demonstrate that admired or similar others accept and practice the behavior. If research shows the audience does not view exercise as normative among peers, search for examples of similar groups who do exercise — such as age-matched cohorts with high participation rates.
• Enhance perceived behavioral control: Acknowledge barriers like time, resources, or capability, and provide practical solutions. This step increases the audience's confidence in their ability to act.

Step 5: Craft a Balanced and Persuasive Narrative

This synthesis step combines data, audience insights, psychology, and storytelling techniques into a coherent whole. Structure the story with a clear arc: establish context, introduce the problem, present data-driven insights supporting your objectives, and conclude with a call to action aligned with the attitudes, norms, and perceived control you have cultivated.

Consider a city planning office with data showing bike owners rarely cycle. A narrative might open with a relatable story about inactivity's impact, then present benefits of cycling on health, social connection, and well-being (attitudes). Showcase commuters who successfully integrated cycling (subjective norms), provide tips on safety, route planning, and affordable bikes (perceived behavioral control), and use infographics comparing commute times and costs. The call to action encourages trying cycling for one week, with links to bike-share programs and local cycling communities.

Testing the Method

Validating the claim that audience research and psychology improve data storytelling requires experimentation. Preliminary research using climate change messaging offers some support, though full results await publication in academic venues. The study design closely mirrors the model itself, with added steps for comparing tailored messages against a control and measuring likely actions.

The research proceeded through seven steps:

1. Choose a topic and dataset. Climate change was selected deliberately as a polarizing subject where shifting opinions is difficult.
2. Identify the audience and take measurements. The audience comprised general public members not working with climate data. A survey measuring TPB components related to climate change was distributed via Google Forms, social media, and online message boards.
3. Analyze data and form groups. Factor analysis on survey responses revealed two distinct audience groups, with notable differences such as Group 1 showing higher positive attitudes toward climate action.
4. Incorporate TPB into data analysis. Group segmentation informed narrative framing — a group with positive attitudes might need less convincing about the problem and more specifics on actionable steps.
5. Create tailored and control stories. Multiple data stories aligned with each group were developed alongside a control message lacking substantial directional framing.
6. Release and measure. Participants rated their likelihood of clicking to "LEARN MORE." The hypothesis predicted higher likelihood scores for the message tailored to their group.
7. Analyze differences. Preliminary findings show small directional differences favoring tailored messages over the control, though larger sample sizes are needed for statistical substantiation.

Survey items measured each TPB factor with multiple questions, combined into a mean score per component. Note that even the theory's creator acknowledges no standardized validated questions exist for measuring TPB concepts across different topics; guidance for generating items is available from academic sources.

Future research may refine what makes messages impactful, and other psychological models may prove more suitable depending on audience and topic. Maslow's hierarchy of needs, for instance, could frame a data story by measuring audience needs and showing how a decision helps meet them.

Beyond the Dashboard: Making Data Stories Resonate

Conventional data storytelling frameworks tend to prioritize the structure of the narrative over the people receiving it. The result is often content that is technically sound but fails to persuade or engage. The missing ingredients are typically a rigorous understanding of the audience and a deliberate application of psychological triggers.

A practical alternative is a five-step model: define specific objectives, conduct UX research, analyze the data, apply psychological principles, and finally, construct a balanced narrative. This structure serves as a roadmap for transforming raw numbers into stories that appeal to both logic and emotion. The goal is not just to inform but to create a memorable experience that inspires a specific action or shift in perspective. By focusing on the human element, data storytellers can move beyond simple presentation and unlock the narrative’s full potential.

The effectiveness of a story is not a matter of guesswork. The same user research methods used to form a hypothesis can be deployed to test the final output. While this requires an investment of time, the payoff is measurable. A/B testing a psychologist-informed story against a control message or other drafts provides empirical evidence of its strength. If the impact is significantly greater, the initial resource outlay is recovered through a stronger, more persuasive connection with the audience.

Practitioners are encouraged to adopt this method and share their experiences to refine the approach further.