What content design is and why readability matters
People don't read online — they scan. Eye-tracking research consistently shows that users read only 20–28% of text on a web page. Content design is the practice of structuring and formatting content for how humans actually process information: scanning for relevance, reading selectively, and extracting meaning as efficiently as possible. This isn't dumbing content down — it's making complex information accessible by working with human cognition rather than against it.
Content design is distinct from UX writing (which focuses on interface microcopy like button labels and error messages) and from content strategy (which focuses on what content to create and how to govern it). Content design focuses on the reading experience itself: how content is structured, chunked, formatted, and sequenced to maximize comprehension and action.
Scanning patterns and content structure
Scanning Pattern Design Core Method
Use when: structuring any content that will be consumed on screen — articles, help pages, product descriptions, documentation.
Eye-tracking research reveals four dominant scanning patterns: F-pattern: On text-heavy pages without clear structure, users scan the first few lines fully, then progressively shorter portions of subsequent lines, creating an F-shape. Content in the right column of long paragraphs gets missed. Design implication: front-load important information in the first two paragraphs; use left-aligned layouts. Layer-cake pattern: On well-structured pages with clear headings, users scan the headings (the "cake layers") and read the content beneath headings that interest them. Design implication: headings must be descriptive and self-contained — they're the primary navigation mechanism for scanners. Spotted pattern: Users scan for specific visual anchors — numbers, bold text, links, images, colored elements. Design implication: use visual differentiation for key information. Commitment pattern: Users read thoroughly from top to bottom, but only when highly motivated (health decisions, financial transactions, legal text). Don't design for commitment reading as the default — it's the exception.
Plain language
Plain Language Checklist Core Method
Use when: writing any content for a general audience, or when simplifying content without losing accuracy.
Plain language isn't simple language — it's clear language that respects the reader's time. Principles: Short sentences: Average 15–20 words. Break long sentences at natural thought boundaries. One idea per sentence. Common words: "Use" not "utilize." "Start" not "commence." "Help" not "facilitate." Every uncommon word creates a micro-friction. Active voice: "We'll review your application" not "Your application will be reviewed." Active voice is shorter, clearer, and tells the reader who's doing the action. Concrete over abstract: "Upload your photo" not "Provide visual documentation." Concrete instructions tell users exactly what to do. Inverted pyramid: Lead with the conclusion, then provide supporting detail. Users who only read the first sentence should still get the key message. Avoid jargon without context: If you must use a technical term, define it on first use or link to a glossary.
Readability Scoring Technique
Use when: setting and measuring readability targets for different content types and audiences.
Readability formulas measure text difficulty. Flesch-Kincaid Grade Level: Estimates the US school grade needed to understand the text. General web content should target grades 6–8. Technical documentation can target grades 10–12. Legal or medical content aimed at patients should still target grade 8. SMOG Index: Similar to Flesch-Kincaid but focuses on polysyllabic words — better for health content. Limitations: Readability scores measure sentence length and word complexity, not concept clarity. Short sentences with jargon will score well but be unreadable. Use readability scores as a guardrail, not a target — they catch problems but don't guarantee quality.
Formatting for comprehension
Content Chunking Framework Core Method
Use when: breaking long content into scannable, digestible sections.
Chunking means organizing content into discrete, self-contained units that can be scanned independently. Heading hierarchy: Every content section gets a descriptive heading. H2 for major sections, H3 for sub-sections. Headings should be informative ("How to reset your password") not vague ("Overview"). Users scan headings to decide what to read — vague headings force them to read everything to find what they need. Paragraph length: 2–4 sentences maximum for web content. Single-sentence paragraphs are acceptable for emphasis. Wall-of-text paragraphs are never acceptable online. Lists over prose: When content is genuinely a list (steps, features, requirements), format it as a list. Don't bury a three-item list in a paragraph. Callouts and emphasis: Use bold for key terms and critical information. Use callout boxes for tips, warnings, and important notes. Use pull quotes or highlighted stats for key data points. But use sparingly — when everything is emphasized, nothing is.
Mobile Reading Optimization Technique
Use when: optimizing content for mobile consumption, where reading conditions are constrained.
Mobile readers have smaller screens, more distractions, and less patience. Optimize: shorter paragraphs (2–3 sentences max), more frequent headings (every 2–3 paragraphs), front-loaded key information (the first sentence of each section should standalone), collapsible sections for detailed content (show summaries, let users expand for detail), and optimized images (fast-loading, captioned, adding information rather than decoration). Test readability on actual mobile devices, not just responsive desktop previews.
Real-world examples
Case study
GOV.UK: content design at national scale
The UK Government Digital Service pioneered content design as a practice. Every page on GOV.UK follows strict readability standards: front-loaded key information, plain language (target age 9 reading level), descriptive headings, and ruthless elimination of unnecessary content. Pages are structured for the task users came to accomplish, not for the department that publishes them. The content design team regularly tests content with users and iterates based on task completion rates, not readability scores alone.
Why it works: Task-oriented structure, plain language standards, and measurement against user outcomes rather than content volume.
Common pitfalls
Writing for the author, not the reader
Subject matter experts naturally write at their own level of understanding — using insider terminology, assuming context, and structuring content around their mental model. Content design requires translating expert knowledge into reader-accessible format, which often means restructuring, simplifying, and adding context the expert takes for granted.