Progressive Disclosure

Show users what they need when they need it. Hide complexity until it's relevant. Let depth feel like discovery, not burden.

Evidence Tiers

[Research] — Peer-reviewed studies, controlled experiments [Expert] — Nielsen Norman Group, recognized UX authorities [Case Study] — Documented examples from major products [Convention] — Industry practice, limited formal validation

Multiple tags = stronger evidence: [Research][Expert] Mixed findings noted as: [Research — Mixed]

Research Foundation

[Research — Limited] Carroll and Rosson (1987) developed the "training wheels" approach, hiding advanced functionality to help novices succeed. However, they noted that empirical evidence for progressive disclosure effectiveness is limited.

From Carroll & Rosson (1997): "No empirical evidence exists regarding the effectiveness of progressive disclosure."

What we do know:

• [Research] Miller's Law (1956): Working memory holds ~7±2 items (some modern research suggests ~4 chunks)

• [Research] Cognitive Load Theory (Sweller, 1988): Reducing extraneous load improves learning

• [Expert] Nielsen Norman Group recommends progressive disclosure for improving learnability, efficiency, and reducing errors

Honest assessment: Progressive disclosure is widely accepted as good practice, but rigorous controlled studies are sparse. Most evidence comes from case studies and practitioner experience.

Source: Nielsen Norman - Progressive Disclosure

The Depth Hierarchy

[Expert] Structure complexity in layers:

┌─────────────────────────────────────┐ │ Layer 1: Essential (always shown) │ ← What 80% of users need ├─────────────────────────────────────┤ │ Layer 2: Common (one click away) │ ← Frequently used options ├─────────────────────────────────────┤ │ Layer 3: Advanced (discoverable) │ ← Power user features ├─────────────────────────────────────┤ │ Layer 4: Expert (documented) │ ← Edge cases, configs └─────────────────────────────────────┘

Users should be able to operate entirely at Layer 1.

Core Patterns

disclosure-1: Default to Simplicity

[Research] Supported by cognitive load theory. Fewer visible options = less extraneous load.

Overwhelming:

┌─────────────────────────────────────┐ │ Name: [] │ │ Email: [] │ │ Phone: [] │ │ Address: [] │ │ [14 more fields...] │ └─────────────────────────────────────┘

Progressive:

┌─────────────────────────────────────┐ │ Name: [] │ │ Email: [] │ │ │ │ [+ Add more details] │ └─────────────────────────────────────┘

Ask: "What's the minimum to accomplish the task?"

disclosure-2: Experts Can Skip, Beginners Can't Be Lost

[Convention] Design for two paths through the same interface:

Beginner path:

Step 1 → Step 2 → Step 3 → Done [Clear linear flow with help text]

Expert path:

[Jump directly to any step] [Keyboard shortcuts] [Collapse help text]

disclosure-3: Expand In-Place

[Convention] When revealing more, don't teleport users to new screens.

Disorienting:

[Click "More options"] [New page loads] [Click back to return]

Smooth:

[Click "More options"] [Section expands below] [Click again to collapse]

Spatial memory matters. Users remember where things are.

High-Impact Patterns

disclosure-4: Smart Defaults Reduce Choices

[Research] Related to choice overload research. Iyengar & Lepper's jam study showed too many choices can paralyze decision-making (though effect size is debated in meta-analyses).

// No default — user must choose Date format: [dropdown with 15 options]

// Smart default — user can change if needed Date format: [MM/DD/YYYY ▼] (based on locale)

disclosure-5: Contextual Revelation

[Convention] Show options when they become relevant, not before.

Too early:

[Shows "Export options" before user has created anything]

Contextual:

[User creates something] [Export options appear]

disclosure-6: Search as Escape Hatch

[Convention] For complex interfaces, search lets users skip the hierarchy.

[User knows what they want but not where] [Presses ⌘K] [Types "export pdf"] [Jumps directly to feature]

Search is disclosure for experts.

Case Study: GOV.UK Bank Holidays

[Case Study] The UK government's GOV.UK redesigned their bank holiday page based on user research:

Before: Busy page with all bank holidays, multiple regions, historical data After: Single upcoming bank holiday prominently displayed, details below

User research revealed most people wanted one thing: the next bank holiday date.

Source: GOV.UK Design System

Patterns Reference

Pattern Use When Evidence

Accordion Related sections, one open at a time [Convention]

Expandable rows Tables with detail views [Convention]

Tabs Parallel categories, similar importance [Convention]

Drawer/sidebar Secondary content needs space [Convention]

Modal Focused subtask, temporary context [Expert] NNg

Tooltip Brief explanation, no interaction [Convention]

Popover Small interactions [Convention]

Myths & Debunked Patterns

MYTH: The Three-Click Rule

Status: Debunked Origin: Jeffrey Zeldman's Taking Your Talent to the Web (2001) — no data provided

The claim: Users will abandon tasks if they can't complete them in 3 clicks.

What research shows:

• Joshua Porter (UIE, 2003): Analyzed 44 users, 620 tasks, 8,000+ clicks. Found no dropoff after 3 clicks and no decrease in satisfaction.

• Jakob Nielsen: Found a site where moving products to 4 clicks from homepage increased findability by 600% over the 3-click version.

Why it persists: It's simple, memorable, and sounds logical. But click counting ignores cognitive factors like information scent, clarity of options, and confidence in progress.

What actually matters: Each click should make users feel closer to their goal. Strong information scent beats low click counts.

Sources:

• Nielsen Norman - 3 Click Rule is False

• UX Myths - All Pages in 3 Clicks

NUANCE: Above the Fold

Status: Partially true, often overstated Origin: Newspaper terminology, adapted to web in 1990s

The claim: Users won't scroll; everything important must be above the fold.

What research shows:

• Eye tracking (NNg): 57% of viewing time is above the fold (down from 80% in 2010)

• The fold effect is real: 102% more attention to 100px above fold vs 100px below

• But people DO scroll: Chartbeat found 66% of attention on media pages is below the fold. ClickTale found 76% of pages get scrolled.

The nuance: The fold matters for initial engagement, but it's not a hard boundary. Users scroll when:

• Content above signals value below

• There's no "false bottom" (design that looks like the page ends)

• Information scent is strong

Guidance: Put your most important content and value proposition above the fold, but don't cram everything there. Design to invite scrolling.

Sources:

• Nielsen Norman - Scrolling and Attention

• Nielsen Norman - The Fold Manifesto

Key Sources

• Miller, G.A. (1956). The magical number seven, plus or minus two.

• Carroll, J.M. & Rosson, M.B. (1987). The paradox of the active user.

• Sweller, J. (1988). Cognitive load during problem solving.

• Nielsen Norman - Progressive Disclosure

• Interaction Design Foundation - Progressive Disclosure