Storytelling With Data

Storytelling with Data (SWD) Skill

Apply the 6-lesson SWD framework from Cole Nussbaumer Knaflic for visual and design decisions, combined with narrative storytelling frameworks (Dykes, Duarte, Miller, Dicks, Cron, Heath) for structuring compelling stories that drive action.

When to Use This Skill

• Creating: Building new visualisations, dashboards, infographics, presentations, pitch decks, or data-driven pages
• Reviewing: Critiquing existing data communications for clarity and impact
• Improving: Performing "makeovers" on charts, dashboards, or layouts to make them more effective
• Advising: Helping choose the right chart type, structure a narrative, declutter, craft a hook, or position a message
• Storytelling: Structuring any narrative: data stories, pitches, brand positioning, conference talks, stakeholder briefings

Format-agnostic by design: This skill covers _what_ to communicate and _how_ to design it. Pair it with format-specific skills that handle output mechanics (e.g. pptx skill for slide decks, a web framework for dashboards).

---

The SWD 6-Lesson Framework

Every data communication task should be evaluated against these 6 lessons, applied in order:

Lesson 1: Understand the Context

Before touching any tool, answer three questions:

1. WHO is your audience? What do they care about? What's their relationship to you? What will motivate them? 2. WHAT do you need them to know or do? Define the single action or takeaway. 3. HOW will you communicate? Live presentation? Email? Dashboard? This changes everything about design.

Key techniques:

• The 3-Minute Story: Can you explain your entire message in 3 minutes? If not, you haven't distilled it enough.
• The Big Idea: One sentence that articulates your unique point of view, what's at stake, and what you want the audience to do. Format: [situation] + [complication] → [recommended action].
• Storyboard first: Sketch your flow on sticky notes or paper before opening any tool. Rearrange freely.
• Exploratory ≠ Explanatory: The audience sees only the _explanatory_ output - the curated insight, not the full analysis journey. Never dump exploratory analysis on your audience.

When reviewing existing work, ask:

• Is there a clear "so what?" on every chart, panel, or section?
• Could the audience state the key message after a 3-second glance?
• Is there a specific call to action?

Lesson 2: Choose an Appropriate Visual Display

Match the visual to the message. You can handle the vast majority of business communications with just a few chart types.

Read `references/chart-selection.md` for the detailed chart selection guide.

Quick decision framework:

• Comparison across categories → Horizontal bar chart (default workhorse)
• Trend over time → Line chart (continuous) or vertical bar chart (discrete periods)
• Part-to-whole → Stacked bar or waterfall (NOT pie charts)
• Two data points comparison → Slopegraph
• Single important number → Simple text (the number itself, large, with context)
• Detailed lookup → Table or heatmap

Charts to AVOID:

• Pie and doughnut charts (hard to compare areas/angles accurately)
• 3D anything (adds clutter, distorts data, makes reading values harder)
• Dual y-axes (confuse the audience about scale relationships)
• Spaghetti graphs (too many overlapping lines - filter or use small multiples instead)

Lesson 3: Eliminate Clutter

Clutter is any visual element that takes up space without adding understanding. Every element increases cognitive load - the mental effort to process information.

Identify and remove:

• Chart borders and unnecessary outlines
• Gridlines (reduce to light or remove entirely)
• Data markers on every point (only mark what matters)
• Redundant labels (if the axis says it, the data label doesn't need to)
• Bold/italic/colour on everything (when everything is emphasised, nothing is)
• Legends (can you label directly instead?)
• Rotated text (hard to read - restructure the chart instead)
• Decorative elements that don't encode data

Apply Gestalt Principles to structure what remains:

• Proximity: Group related items close together; separate unrelated ones
• Similarity: Use consistent colours/shapes for items in the same category
• Enclosure: Subtle backgrounds can group related content
• Closure: Remove unnecessary borders - the eye fills in gaps
• Continuity: Align elements so the eye flows naturally
• Connection: Lines connecting points imply sequence or relationship

Decluttering process (step by step):

1. Remove chart borders 2. Remove or lighten gridlines 3. Remove data markers (unless specific points need emphasis) 4. Clean up axis labels (round numbers, remove unnecessary precision) 5. Label data directly (eliminate legends where possible) 6. Use consistent, strategic colour

Lesson 4: Focus Your Audience's Attention

Use preattentive attributes - visual properties the brain processes before conscious thought - to guide the eye to what matters:

• Colour: The most powerful tool. Use a single accent colour to highlight the key data; push everything else to grey. Never use colour decoratively.
• Size: Larger elements draw attention first. Make the key number or data point bigger.
• Position: Items at the top-left get seen first (in left-to-right reading cultures). Place the most important information there.
• Bold/weight: Use sparingly on text to create hierarchy. Bold the key phrase, not the whole paragraph.

The Grey + One Colour rule: Default everything to grey, then use ONE strategic accent colour to highlight the story. This is the single most impactful SWD technique.

Memory matters:

• Iconic memory (~0.5 seconds): Preattentive attributes are processed here - that's why colour and size pop instantly
• Short-term memory (~4 items): Don't overload a single view with too many competing elements
• Long-term memory: Connect to what the audience already knows - use familiar chart types and conventions

Lesson 5: Think Like a Designer

Design serves the message. Apply these principles:

• Form follows function: Every element earns its place by serving comprehension
• Affordances: Make interactive elements look clickable; make important text look important
• Accessibility: Would this work in greyscale? For colour-blind audiences? Without a narrator explaining it?
• Aesthetics build trust: Clean, well-aligned visuals signal competence and credibility

Practical design rules:

• Use consistent alignment (create an invisible grid)
• Left-align text (easier to read than centred for body content)
• Use white space generously - it's not wasted space, it's breathing room
• Limit fonts to 1-2 families; use size and weight for hierarchy
• Action titles > descriptive titles: "Revenue grew 23% in Q3" beats "Q3 Revenue"
• Every chart, section, or panel should have a clear, declarative title that states the "so what"
• Add context with text annotations directly on the visualisation

Lesson 6: Tell a Story

Without narrative, data visualisations are just pretty pictures. Multiple frameworks exist for structuring compelling stories. Choose based on your scenario.

Read `references/narrative-frameworks.md` for detailed framework guidance and `references/hooks-and-moments.md` for hook and aha moment techniques.

Framework Selection

The Data Storytelling Arc (Brent Dykes) - Default for Data Presentations

1. Setting & Hook: Provide "just enough" context, then present a notable observation that reveals a problem or opportunity. The Hook creates an open question. 2. Rising Points: Focused supporting details that build understanding and tension. Not a data dump. 3. Aha Moment: The central insight: an unexpected shift in understanding + explicit "so what." This is the climax. Do NOT put it upfront. 4. Solution & Next Steps: Provide options and make a recommendation.

Key principle: Data storytelling = Data + Narrative + Visuals (all three required). An insight must surprise, shift understanding, AND inspire action. Reporting is not storytelling.

The Sparkline (Nancy Duarte) - For Persuasive Presentations

Alternate between "what is" (current state) and "what could be" (desired future) throughout. Contrast creates energy. End with "new bliss": the world with your idea adopted. The audience is the hero; the presenter is the mentor.

The StoryBrand SB7 (Donald Miller) - For Product/Brand Positioning

7 elements: Character (customer = hero) > Problem (external, internal, philosophical) > Guide (your brand: empathy + authority) > Plan (clear steps) > Call to Action > Failure (stakes if they don't act) > Success (transformation). Key insight: Companies sell solutions to external problems, but customers buy solutions to internal problems.

The Five-Second Moment (Matthew Dicks) - For Personal Stories

Every great story is about a single moment of transformation taking no more than five seconds. Start at the end (know your moment), begin as close to it as possible. Use the Dinner Test: if you wouldn't tell it this way to a friend, don't tell it that way at all.

Brain Science (Lisa Cron) + Sticky Messages (Heath Brothers)

Hooks work because they create knowledge gaps that trigger dopamine. Use the SUCCESs checklist: Simple, Unexpected, Concrete, Credible, Emotional, Stories. Combat the Curse of Knowledge: once you know something, you can't imagine not knowing it.

Storytelling techniques (applicable across all frameworks):

• Horizontal logic: Read just the section/slide titles in sequence. They should tell a complete story on their own.
• Vertical logic: Each individual section, panel, or slide should make sense on its own with title + content.
• Repetition: Repeat your key message at least 3 times in different ways.
• Progressive reveal: Don't show everything at once. Build up piece by piece.
• Match the medium: Live presentations = sparse visuals (you are the narrator). Read-alone formats need more text and annotation.
• Hooks within 60 seconds: Use anomaly, stakes, contrast, question, vulnerability, or in-medias-res hooks (see references/hooks-and-moments.md).
• Use "but" and "therefore" to connect story elements: "and" kills momentum (Dicks).

---

Applying SWD: Task-Specific Workflows

Creating New Data Communications

This workflow applies regardless of output format (slide deck, dashboard, infographic, static site, report).

1. Context interview - Ask the user: Who's the audience? What's the one thing they should take away? How will this be delivered? 2. Choose narrative framework - Data insight: Dykes Arc. Persuasive pitch: Duarte Sparkline. Product positioning: Miller SB7. (See references/narrative-frameworks.md) 3. Storyboard - Draft section/panel titles as a narrative arc. For Dykes: Setting > Hook > Rising Points > Aha Moment > Solution. For Duarte: What Is > What Could Be (alternating) > New Bliss. 4. Craft the Hook - Must land within 60 seconds. Choose from: anomaly, stakes, contrast, question, vulnerability, or in-medias-res. (See references/hooks-and-moments.md) 5. Chart selection - For each data point, pick the simplest effective chart (see references/chart-selection.md) 6. Build - Apply Lessons 3-5 while creating each section or view 7. Verify the Aha Moment - Does it surprise? Shift understanding? Have an explicit "so what"? Is it at the climax, not the beginning? 8. Review - Check horizontal logic (titles alone tell the story), vertical logic (each unit stands alone), and run the SUCCESs checklist

Format-Specific Considerations

Creating a Persuasive Pitch or Keynote

1. Define The Big Idea (Duarte): One sentence = situation + complication leading to recommended action 2. Map the audience as hero (Miller SB7): What do they want? What problem do they face (external, internal, philosophical)? How are you the guide? 3. Structure with Sparkline contrast: Alternate "what is" vs "what could be" throughout 4. Plant a S.T.A.R. Moment (Duarte): Something They'll Always Remember: a dramatic, tangible demonstration of the idea 5. Craft sound bites: Small, repeatable phrases for headlines and sharing 6. End with New Bliss: Paint the picture of the world with the idea adopted + clear call to action 7. Apply SWD Lessons 2-5 for any data slides within the pitch

Structuring a Personal or Conference Story

1. Find the five-second moment (Dicks): What single moment of transformation is this story about? 2. Start at the end: Know where you're going, then build backward 3. Begin in the opposite: Open in the emotional/situational opposite of where the story ends 4. Apply the Dinner Test: Would you tell it this way to a friend over dinner? 5. Load engagement devices: Stakes (hopes/fears before moving forward), Surprise (build it, don't spoil it), Suspense (make them wonder what's next) 6. Slow down at the moment: Use the Hourglass (Dicks): add detail and a beat right before the climax

Positioning a Product or Brand

1. Complete the SB7 BrandScript (Miller): Character > Problem > Guide > Plan > CTA > Failure > Success 2. Define the villain: Must be a root source (not a feeling), relatable, singular, and real 3. Articulate three problem levels: External (tangible), Internal (emotional), Philosophical (why it's wrong) 4. Establish Guide credentials: Empathy ("I understand") + Authority ("I've solved this before") 5. Show the transformation: Clear before/after: who does the customer become? 6. Run SUCCESs check (Heath): Is the message Simple, Unexpected, Concrete, Credible, Emotional, Story-driven?

Reviewing / Critiquing Existing Work

Run through the SWD Review Checklist in references/review-checklist.md. For each chart, panel, section, or slide, evaluate against all 6 lessons and provide specific, actionable feedback. Additionally evaluate the narrative structure:

• Does the presentation have a clear Hook within the first 60 seconds?
• Is there a genuine Aha Moment (surprise + shift + "so what")?
• Is the insight at the climax, not the beginning?
• Does the narrative follow a coherent arc (not just a collection of charts)?
• Would it pass the Dinner Test: is it told the way a human would tell it?
• Does the closing provide clear, actionable next steps?

Chart / Visualisation Makeover

1. Identify the core message (what should the audience think/do?) 2. Choose a better chart type if needed (Lesson 2) 3. Strip all clutter (Lesson 3) 4. Apply grey + accent colour strategy (Lesson 4) 5. Add a declarative action title (Lesson 5) 6. Add text annotations to guide interpretation (Lesson 5) 7. Show before/after to demonstrate the improvement

---

Reference Files

Storytelling With Data Skill

Credit to Daniel Gormly for the original version of this skill (which was specifically for slide decks).

Chart Selection Guide

Based on the Storytelling with Data methodology. Choose the simplest chart that communicates your message effectively.

Decision Matrix

Ask: "What relationship am I showing?" Then pick from below.

---

Comparison Across Categories

Default: Horizontal Bar Chart

• The workhorse of data visualisation - familiar, easy to read, works for almost everything
• Labels read naturally left-to-right; bars extend right for easy length comparison
• Use when: comparing discrete items (products, regions, departments, survey responses)
• Order bars meaningfully: by value (largest to smallest) or by natural order (e.g., survey scale)
• Always start the value axis at zero (truncated axes distort bar comparisons)

Alternative: Vertical Bar Chart (Column Chart)

• Use when categories have a natural sequential order (months, quarters, years)
• Also works for small numbers of categories (< 8) with short labels
• Still must start at zero

Alternative: Dot Plot

• Use when you want to show values without the visual weight of bars
• Good for showing range between two values (e.g., before/after, min/max)

---

Change Over Time (Trends)

Default: Line Chart

• Best for continuous time series - the slope communicates rate of change intuitively
• Audiences naturally read left-to-right as time progression
• Limit to 3-4 lines maximum; beyond that, use small multiples or filter
• OK to not start at zero (unlike bar charts) if the focus is rate of change

Alternative: Vertical Bar Chart

• Use for discrete time periods (Q1, Q2, Q3, Q4) where continuity isn't implied
• Use when absolute magnitude matters more than the trend shape

Alternative: Slopegraph

• Use when comparing exactly two time periods (before/after, Year 1 vs Year 2)
• Excellent for showing which items increased vs decreased simultaneously
• The slope direction and steepness tells the story at a glance

Avoid: Area Chart (usually)

• Stacked area charts are hard to read - only the bottom series has a clear baseline
• Single area charts can work to emphasise volume/magnitude over time

---

Part-to-Whole (Composition)

Default: Stacked Bar Chart (Horizontal)

• Shows composition AND allows comparison across categories
• Use 100% stacked bars when proportions matter more than absolute values
• Order segments consistently across bars

Alternative: Waterfall Chart

• Shows how individual components add up to a total (or how a starting value changes)
• Great for financial walk-throughs (revenue → costs → profit)

Avoid: Pie Charts and Doughnut Charts

• Human eyes are poor at comparing angles and areas accurately
• Any data that could go in a pie chart is better served by a horizontal bar chart
• If you must show parts of a whole and your audience insists on pies, limit to 2-3 slices maximum

---

Relationship Between Two Variables

Default: Scatterplot

• Shows correlation, clusters, and outliers between two continuous variables
• Add a trend line only if it genuinely helps interpretation
• Label notable outliers directly on the chart

---

Single Key Number

Default: Simple Text

• If your key message is one number, make it BIG
• Show it as a large formatted number with brief context: "Revenue grew 23% year-over-year, reaching $4.2M"
• No chart needed - a chart with one data point is wasteful

---

Detailed Lookup / Reference Data

Default: Table

• When the audience needs to look up specific values, a well-formatted table beats any chart
• Use light formatting: remove heavy borders, alternate subtle row shading, align numbers right
• Bold or colour-highlight the row/column you want to draw attention to

Alternative: Heatmap

• A table with colour-encoded cell values to show magnitude patterns
• Good for spotting patterns in matrix data (time × category, feature × product)
• Use a single colour gradient (light → dark) rather than a diverging palette unless there's a meaningful midpoint

---

When to Combine Visuals

• Chart + Simple Text callout: Lead with the headline number in large text, support with the chart below
• Small multiples: When you have too many series for one chart, repeat the same chart type for each category - same scale, same format, arranged in a grid
• Annotation layers: Add text callouts, reference lines, and shaded regions directly on charts to guide interpretation

---

Universal Rules

1. Every chart needs a declarative action title - not "Q3 Sales by Region" but "Southeast region drove 60% of Q3 growth" 2. Label data directly when possible - eliminate legends by placing labels next to the data they describe 3. Use consistent scales when comparing multiple charts - don't change axis ranges between views 4. Round numbers - 23% not 23.47% unless precision genuinely matters 5. Bar charts must start at zero - line charts don't have to 6. Horizontal bars > vertical bars in most cases (labels are easier to read)

Colour & Emphasis Strategy

Colour is the single most powerful tool in data visualisation - and the most frequently misused. These guidelines ensure colour serves the story, not decoration.

---

The Core Principle: Grey + One Accent Colour

The most impactful SWD technique:

1. Default everything to grey - all bars, lines, labels, and supporting elements 2. Apply ONE accent colour to the specific data point, series, or element that IS the story 3. The accent element instantly draws the eye because of contrast with the grey surroundings

This works because of preattentive processing - the brain detects colour differences before conscious thought. A single blue bar among grey bars is impossible to miss.

When to use more than one accent colour

• Comparing exactly 2-3 things (e.g., "us vs competitor" - use brand colour vs grey)
• Showing positive vs negative (green vs red, or blue vs orange for accessibility)
• Categorical data where the categories ARE the story - but still limit to 3-4 colours max

---

Colour Selection Rules

Do

• Use colour meaningfully - each colour should encode information, not just look pretty
• Be consistent - same colour = same meaning throughout the entire piece
• Use saturation for emphasis - muted/desaturated for background, fully saturated for focus
• Test in greyscale - if the chart loses its message in greyscale, the colour strategy isn't strong enough
• Use sequential palettes for magnitude (light → dark of one hue)
• Use diverging palettes only when there's a meaningful midpoint (e.g., positive/negative, above/below target)

Don't

• Don't use rainbow palettes - they have no natural order and create visual chaos
• Don't use red + green as the only differentiator - ~8% of men are red-green colour blind
• Don't use bright colours for large areas - saturated colours on big chart elements are overwhelming; reserve them for small accents
• Don't use colour to differentiate 7+ categories - if you need that many, restructure the visual (small multiples, filtering, or direct labels)
• Don't use coloured backgrounds unless there's a specific design reason - white or very light grey is almost always better for data readability

---

Emphasis Hierarchy

Layer these preattentive attributes for progressive emphasis:

---

Colour for Specific Chart Types

Bar Charts

• Grey all bars, accent the one(s) that tell the story
• Or: grey all bars, use a gradient of the accent colour to show magnitude
• Never use a different colour for every bar unless the categories ARE the point

Line Charts

• Grey all lines, accent the one series the audience should track
• Use line weight (thickness) as secondary emphasis - thicker = more important
• Dashed lines for projections/forecasts to distinguish from actual data

Tables & Heatmaps

• Use subtle alternating row shading (very light grey, not colour)
• Use a single-hue sequential palette for heatmap values
• Bold + accent colour for the specific cells you want to highlight

Text-Heavy Content (Slides, Reports, Infographic Panels)

• Title hierarchy: dark colour for main title, medium grey for subtitles
• Use accent colour in text sparingly - highlight only the key phrase or number
• Never use more than 2 text colours in a single view or section (excluding grey for de-emphasis)

---

Recommended Palettes

Safe Defaults (accessible, professional)

• Accent: Steel blue (#4472C4) or Teal (#008080) - professional, colour-blind safe
• De-emphasis: Medium grey (#A6A6A6) for supporting data
• Background elements: Light grey (#D9D9D9) for gridlines and borders
• Positive/Negative: Blue (#4472C4) / Orange (#ED7D31) - avoids red/green issues

For Dark Backgrounds

• If the user insists on dark backgrounds, invert: use lighter greys for de-emphasis and brighter accent colours
• Test carefully - dark backgrounds reduce readability and are rarely worth the aesthetic trade-off

---

Applying Emphasis to Non-Chart Content

The same principles apply to text-heavy sections in any format (slide panels, dashboard cards, infographic sections, report pages):

• Bold the key phrase, not the entire sentence
• Increase font size for the single most important number or statement
• Grey out supporting context - it's there if they need it but doesn't compete for attention
• Use colour in text minimally - one accent colour for the critical insight
• Progressive disclosure: In presentations, use animation to reveal key insights progressively. In interactive formats (dashboards, web pages), use overview-then-detail patterns, expandable sections, or scroll-triggered reveals. In static formats (infographics, reports), control the reveal through reading order and visual hierarchy

Hooks & Aha Moments: The Engine of Story

Hooks open attention loops. Aha Moments close them. Together they power every compelling narrative: data presentation, pitch, or conference talk.

---

What Makes a Hook Work (Neuroscience)

The brain is a prediction machine. A hook works by creating a knowledge gap: a space between what the audience knows and what they want to know. This triggers dopamine release (Lisa Cron), which says "pay attention, this matters."

A hook must do two things: 1. Signal that something unexpected, important, or threatening is happening 2. Create an open question the audience needs answered

Why hooks fail:

• No stakes: "Here's our Q3 data" (who cares?)
• No gap: The answer is obvious or already known
• Too abstract: The brain needs specifics to engage (Cron: "we think in images, not abstractions")
• Buried: Setup goes on too long before the hook lands

---

Hook Types for Data Storytelling

1. The Anomaly Hook

Present a data point that breaks an expected pattern. The audience's brain screams "that doesn't fit."

Pattern: "Everything looked normal... until [unexpected observation]."

• "Revenue grew 12% last quarter across every region, except Southeast, which dropped 23%."
• "Customer satisfaction scores have been stable for 18 months. Then in March, they fell off a cliff."

Why it works: The brain hates unexplained pattern breaks. It demands resolution. This is Dykes' Hook in the Data Storytelling Arc.

2. The Stakes Hook

Open with what's at risk: financial, operational, reputational, or human.

Pattern: "If we don't [act], [consequence]."

• "At current churn rates, we'll lose $4.2M in recurring revenue by Q4."
• "Every week we delay, 300 more records are at risk of permanent loss."

Why it works: The brain prioritises survival-relevant information (Cron). Stakes activate the amygdala.

3. The Contrast Hook (Duarte)

Juxtapose the current state against a possible future to create tension.

Pattern: "Right now [what is]. But imagine [what could be]."

• "Right now, our reviewers spend 6 hours per document. With the right system, that could be 45 minutes."
• "Today, these sites are documented by hand. What if we could map an entire area in 20 minutes?"

Why it works: Contrast creates cognitive dissonance. The gap between "what is" and "what could be" generates energy and desire for resolution.

4. The Question Hook

Pose a question the audience can't help but try to answer.

Pattern: "What if...?" or "Why does...?" or "Did you know...?"

• "Why do 40% of our highest-value customers leave within 90 days of onboarding?"
• "What would happen if we could predict failures 500 metres before they occur?"

Why it works: The brain is goal-directed (Cron). A question gives it a goal. It can't not try to answer.

5. The Confession/Vulnerability Hook (Dicks)

Start with a personal moment of failure, surprise, or realisation.

Pattern: "I was wrong about..." or "I didn't expect..."

• "When I first saw the feedback data, I thought the system was failing. I was wrong."
• "Three months ago, I would have told you this approach couldn't match manual accuracy."

Why it works: Vulnerability builds trust instantly. The audience leans in because authenticity is rare. The brain responds to characters who have something at stake.

6. The "In Medias Res" Hook

Start in the middle of the action. Skip all preamble.

Pattern: Drop the audience into a specific, concrete moment.

• "The alert fired at 2:47am. The system was about to breach its safety threshold."
• "Slide 47. That's where the presentation fell apart in front of the executive team."

Why it works: Specificity creates mental imagery (Cron: "the brain thinks in images"). Action creates momentum. The audience is instantly "in the scene."

---

What Makes an Aha Moment Work

Definition (Dykes)

An Aha Moment combines: 1. An insight: an unexpected shift in understanding 2. A "so what": why the audience should care

An insight is NOT just an interesting observation. It must challenge existing assumptions and shift perspectives.

The Insight Test

Ask: Does this finding...

• Surprise? Is it unexpected given what we believed? (If the dog bites the mailman, it's not a story. If the mailman bites the dog, it is.)
• Shift understanding? Does it change how the audience thinks about the problem?
• Inspire action? Does it create urgency to do something differently?

If it doesn't pass all three, it's an observation, not an insight.

Placement Principles

Do NOT put the Aha Moment first (even though "executive summary" culture wants you to):

• Upfront disclosure kills suspense and emotional power
• Reporting is not storytelling: know which mode is appropriate
• Exception: If the audience explicitly demands bottom-line-up-front, give a one-sentence preview then tell the story ("Here's the headline, now let me show you why")

Build to it progressively: 1. Setting establishes context 2. Hook opens the question 3. Rising Points build evidence and tension 4. Aha Moment lands at the peak: maximum emotional and intellectual impact 5. Solution channels the energy into action

Making the Aha Moment Land

Techniques from multiple frameworks:

• The Hourglass (Dicks): Slow down right before the moment. Add sensory detail, a pause, a beat. This signals "this is important" and lets the audience catch up emotionally.
• Contrast framing (Duarte): State the old belief explicitly before revealing the new understanding. "We assumed X. The data shows Y."
• Concrete specificity (Heath): The Aha Moment must be tangible, not abstract. Not "customer behaviour is changing" but "customers who contact support in the first 48 hours are 3x more likely to churn within 90 days."
• Emotional connection (Cron): Connect the insight to someone the audience cares about: real people, not averages.
• The "so what" bridge (Dykes): Immediately follow the insight with why it matters. Don't let the audience guess: tell them explicitly.

---

The Transformation Arc

Every great story, at every scale, follows the same shape:

Old Understanding  ->  Tension/Evidence  ->  New Understanding  ->  New Action

In data storytelling (Dykes):

Setting  ->  Hook  ->  Rising Points  ->  Aha Moment  ->  Solution

In presentations (Duarte):

What Is  ->  What Could Be (alternating)  ->  New Bliss

In personal stories (Dicks):

Beginning (opposite of ending)  ->  Events  ->  Five-Second Moment

In brand narrative (Miller):

Hero Wants  ->  Has Problem  ->  Meets Guide  ->  Gets Plan  ->  Takes Action  ->  Avoids Failure  ->  Achieves Success

They are all the same shape. The audience starts in one place and ends in another. The hook opens the loop, the aha moment closes it. Everything else is the bridge between.

---

Quick Reference: Hook + Aha Moment Checklist

Before you present, verify:

Hook:

• [ ] Does it land within the first 60 seconds?
• [ ] Does it create a specific, unanswered question?
• [ ] Is it concrete, not abstract?
• [ ] Does it signal that something is at stake?
• [ ] Would it pass the Dinner Test (Dicks): would you open with this if telling a friend?

Aha Moment:

• [ ] Is it genuinely surprising? (Challenges existing assumptions)
• [ ] Does it shift understanding? (Not just an interesting fact)
• [ ] Is the "so what" explicit? (Don't make the audience guess why it matters)
• [ ] Is it placed at the climax, not the beginning?
• [ ] Is it concrete and specific, not abstract?
• [ ] Does it connect to people the audience cares about?
• [ ] Does it naturally lead to a clear action/recommendation?

Narrative Frameworks for Storytelling with Data

A synthesis of frameworks from respected storytelling thought leaders, adapted for data communication. Use this reference when structuring any narrative: presentations, pitches, reports, or data stories.

---

1. The Data Storytelling Arc (Brent Dykes)

The most directly applicable framework for data presentations. Based on Freytag's Pyramid, optimised for fact-based stories.

Structure: 4 Sections

Setting & Hook

• Setting: Provide "just enough" context: area of focus, timeframe, who the data represents. Don't summarise the analysis process.
• Hook: A notable observation that reveals a potential problem or opportunity. It captures attention and creates an open question.

Rising Points

• Focused supporting details that build toward the climax. Not a data dump.
• Each point should advance the audience's understanding.
• Number of rising points scales with topic complexity.

Aha Moment (Climax)

• The central insight: an unexpected shift in understanding + explanation of why the audience should care (the "so what").
• Must motivate the audience to address a problem or pursue an opportunity.
• An insight challenges existing assumptions and shifts perspectives. Without it, your data story is "trivial or boring" (Dykes).
• Placement matters: Don't put the insight upfront (executive summary style). That removes the suspense essential to engagement. Reporting is not storytelling.

Solution & Next Steps

• Provide options and make a recommendation.
• Without clear next steps, audiences become paralysed by the information.
• The arc should exit at a higher position than it started: each data story elevates the audience's domain knowledge.

When to use: Any data-driven presentation, insight report, or stakeholder briefing. This is the default narrative model for data communication.

Key Dykes principles:

• Data storytelling = Data + Narrative + Visuals (three pillars, all required)
• Eye-appealing charts without strong narrative = weak data stories
• The insight is the source of surprise: it invites the audience to replace well-established explanations with new, unexpected ones
• Data stories should "induce a change in the audience through learning"

---

2. The Sparkline / Presentation Form (Nancy Duarte)

The go-to framework for persuasive presentations. Discovered by reverse-engineering great talks (Steve Jobs iPhone launch, MLK "I Have a Dream").

Core Structure: "What Is" vs "What Could Be"

A great presentation oscillates between the current state and the desired future:

What Is -> What Could Be -> What Is -> What Could Be -> ... -> New Bliss

Beginning

• Establish common ground with the audience by describing "what is": the current reality they recognise.
• Turning Point 1 (Call to Adventure): Create imbalance by stating "what could be" for the first time. The gap between current and future must be clear and explicit.

Middle

• Alternate between "what is" and "what could be" to create contrast and tension.
• Three types of contrast:
• Content contrast: Your views vs audience's views, current state vs future state
• Emotion contrast: Analytical content vs emotional content
• Delivery contrast: Traditional vs nontraditional delivery methods

End (New Bliss)

• Paint a picture of the world with your idea adopted.
• End with a clear call to action.
• The audience should leave with the "new bliss" as the last thing in their minds.

Duarte's Key Principles:

• The audience is the hero, not the presenter. The presenter is the mentor/guide.
• The Big Idea: One sentence that captures situation + complication, leading to what could be. Must convey your unique point of view.
• Contrast = engagement: Moving between opposing poles creates energy and interest.
• S.T.A.R. Moments: Something They'll Always Remember: a dramatic, significant, sincere moment that magnifies the Big Idea. (e.g., Jobs pulling MacBook Air from an envelope)
• Sound bites: Small, repeatable phrases that feed headlines, social media, and rally cries.

When to use: Persuasive presentations, keynotes, pitch decks, change management comms, any talk where you need to move people to action.

---

3. The StoryBrand SB7 Framework (Donald Miller)

A 7-step framework for brand/business storytelling. Positions the customer as the hero and your brand as the guide.

The 7 Elements:

1. A Character (Hero = your customer): Define what they want: one clear desire. Opens a "story gap" between their present state and desired state.

2. Has a Problem: Three levels of conflict the villain causes:

• External: The tangible problem (e.g., inefficient process)
• Internal: The emotional frustration (e.g., feeling overwhelmed)
• Philosophical: Why it's fundamentally wrong (e.g., "it shouldn't be this hard")
• Key insight: Companies sell solutions to external problems, but customers buy solutions to internal problems.

3. And Meets a Guide (your brand): Demonstrate two things:

• Empathy: "I understand your pain"
• Authority: "I have the experience to help" (testimonials, results, credentials)

4. Who Has a Plan: Give them a clear path (3-4 steps max). Reduces perceived risk and builds confidence.

5. And Calls Them to Action: Direct ("Buy Now") and transitional ("Download the free guide"). Without a clear CTA, nothing happens.

6. That Helps Them Avoid Failure: Show the negative stakes: what happens if they don't act. Most businesses don't bring up the negative stakes enough.

7. And Ends in Success: Paint the transformation: the "before and after." People are drawn to transformation.

When to use: Product or service positioning, website copy, pitch decks, sales conversations, brand messaging, consulting proposals.

---

4. The Five-Second Moment (Matthew Dicks)

A performance storytelling framework from a 61-time Moth StorySLAM champion. Applicable to any personal or business story.

Core Concept: Every great story is about a single moment of transformation: a shift that takes no more than five seconds. Without it, you don't have a story; you have an anecdote.

Key Techniques:

Homework for Life

• Daily practice: At the end of each day, write 1-2 sentences about the most storyworthy moment. Date + moment, nothing more.
• Develops a "storytelling lens": you start seeing meaningful moments everywhere.

Story Construction Rules

• Start at the end: Know your five-second moment first, then build backward.
• Begin as close to the end as possible: Don't waste time on setup that doesn't serve the moment.
• The Dinner Test: If you wouldn't tell the story this way to a friend at dinner, don't tell it that way to anyone. Storytelling is not theatre.
• Stories can never be about two things: One five-second moment per story.
• Present tense: Makes the audience feel they're with you in the moment.

Engagement Devices

• Stakes: Load the audience with your hopes and fears before moving forward. Plans that don't work > plans that do.
• Surprise: Build it in, don't ruin it with thesis statements.
• Suspense: Hint at future events but reveal only enough to keep guessing.
• Backpacks: Load the audience with information that creates anticipation.
• Hourglasses: Slow down just before the moment the audience has been waiting for.
• Breadcrumbs: Drop hints that pay off later.
• Crystal Balls: False predictions to keep the audience guessing.
• Use "but" and "therefore" to connect story elements: "and" provides no momentum.
• The negative is almost always better than the positive for storytelling.
• Laughter is the best camouflage for hiding important information.

When to use: Personal anecdotes in presentations, conference talks, team meetings, pitch openers.

---

5. Brain Science of Story (Lisa Cron, "Wired for Story")

Why stories work at a neurological level.

Core Cognitive-Story Pairs:

Key Cron Principle: The brain's hardwired desire to learn "what happens next" creates attention. This triggers a dopamine rush. Without it, even perfect charts won't hold interest.

Application to data storytelling: Your audience's brain is asking "what's in it for me?" from the first slide. If you don't establish stakes and a clear goal immediately, you lose them: regardless of how clean your charts are.

When to use: As a mental checklist when reviewing any story or presentation for engagement. Useful for understanding why SWD techniques work.

---

6. SUCCESs Framework (Chip & Dan Heath, "Made to Stick")

Why some ideas stick and others don't. A diagnostic checklist for any message.

The Curse of Knowledge: The single biggest barrier to sticky messages. Once you know something, you can't imagine not knowing it. This is why experts produce terrible presentations: they skip context their audience needs. Combat it by using concrete language and the Dinner Test.

When to use: As a final review checklist for any presentation, pitch, or report. Run your key message through each letter of SUCCESs.

---

Framework Selection Guide

---

Recommended Reading (Prioritised)

1. Effective Data Storytelling - Brent Dykes (Data + Narrative + Visuals; the Data Storytelling Arc) 2. Resonate - Nancy Duarte (Sparkline; "what is" vs "what could be"; audience as hero) 3. Building a StoryBrand - Donald Miller (SB7; customer as hero; brand as guide) 4. Storyworthy - Matthew Dicks (Five-second moment; Homework for Life; performance storytelling) 5. Made to Stick - Chip & Dan Heath (SUCCESs framework; Curse of Knowledge) 6. Wired for Story - Lisa Cron (Brain science of hooks; cognitive-story pairs) 7. DataStory - Nancy Duarte (Data-specific sequel to Resonate) 8. Stories That Stick - Kindra Hall (4 business story types: value, founder, purpose, customer)

SWD Review Checklist

Use this checklist when reviewing or critiquing any data communication: charts, dashboards, infographics, slide decks, static sites, or reports. Evaluate each item and provide specific, actionable feedback.

---

1. Context & Message

• [ ] Clear audience: Is it obvious who this is for? Is the complexity/detail level appropriate for them?
• [ ] Single key message: Can you identify the ONE thing the audience should take away from each chart, panel, or section?
• [ ] Action title: Does each title state the "so what" - not just describe the data? (e.g., "Churn increased 15% after price change" vs "Monthly Churn Rate")
• [ ] Call to action: Does the communication end with a clear recommendation or next step?
• [ ] Explanatory not exploratory: Is this showing curated insights, or dumping raw analysis?

Common problems:

• Generic titles like "Overview", "Results", "Data"
• No clear recommendation - ends with data and no interpretation
• Too much data shown "for completeness" rather than for the story

---

2. Chart Type Selection

• [ ] Right chart for the message: Does the visual type match what you're trying to show? (comparison → bars, trend → lines, composition → stacked bars)
• [ ] Simple over complex: Could a simpler chart type work? (e.g., bar chart instead of radar chart)
• [ ] No pies or donuts: If present, recommend replacing with horizontal bars
• [ ] No 3D effects: If present, remove immediately - they distort data
• [ ] No dual y-axes: If present, split into two separate charts or use indexing
• [ ] No spaghetti: If >4 overlapping lines, recommend small multiples, filtering, or highlighting one series

Common problems:

• Pie chart with 8+ slices where a bar chart would be much clearer
• Line chart for categorical data that has no natural order
• Overly complex or exotic chart types when simple bars/lines would work

---

3. Clutter Audit

• [ ] Chart borders removed: No unnecessary boxes around charts
• [ ] Gridlines minimal: Light grey or removed entirely
• [ ] No redundant labels: If axis labels exist, individual data labels are usually unnecessary (and vice versa)
• [ ] No rotated text: If axis labels are rotated, the chart should be restructured (usually flip to horizontal bars)
• [ ] Legend placement: Can the legend be replaced by direct labelling on the data?
• [ ] Data-ink ratio: Is most of the "ink" (pixels) used to represent actual data, not decoration?
• [ ] White space: Is there enough breathing room, or is everything crammed together?
• [ ] Decimal precision appropriate: Are numbers rounded to the level of precision that matters?

Common problems:

• Every data point labelled AND axis ticks AND gridlines - pick one approach
• Bright coloured backgrounds or borders that compete with data
• Logos, clip art, or decorative images that add no informational value

---

4. Attention & Emphasis

• [ ] Strategic use of colour: Is colour used to highlight the key message, or is it decorative/random?
• [ ] Grey + accent pattern: Are non-essential elements pushed to grey with one strategic accent colour?
• [ ] Visual hierarchy: Can you tell what's most important within 3 seconds?
• [ ] Preattentive attributes used: Is size, colour, or position guiding the eye to the key insight?
• [ ] Not everything is bold: Bold/colour/size emphasis is reserved for the key elements only

Common problems:

• Rainbow colour palettes where every category gets a bright colour - nothing stands out
• Everything is the same visual weight - no hierarchy
• Red/green colour coding that's inaccessible to colour-blind viewers

---

5. Design Quality

• [ ] Consistent alignment: Are elements aligned to an invisible grid? Check left edges, spacing, chart sizes
• [ ] Text is left-aligned: Body text and labels should be left-aligned (not centred) for readability
• [ ] Font consistency: Maximum 2 font families; hierarchy through size and weight only
• [ ] Annotations present: Are there text annotations on charts explaining key data points or changes?
• [ ] Accessible in greyscale: Would the key message still come through without colour?
• [ ] Standalone clarity: If someone sees this without a presenter, would they understand it?

Common problems:

• Centred text blocks that are hard to scan
• Inconsistent spacing between sections or chart elements
• Charts that are meaningless without someone talking over them

---

6. Narrative Structure

• [ ] Sequential logic: Reading just the section/panel/slide titles in order tells a coherent story
• [ ] Self-contained units: Each individual section, panel, or slide makes sense with just its title + content
• [ ] Narrative arc: The communication follows Beginning (setup) → Middle (tension/insight) → End (recommendation)
• [ ] Repetition of key message: The core takeaway appears at least 3 times (intro, body, conclusion)
• [ ] Progressive disclosure: Is information built up logically, or is the audience hit with everything at once? (In interactive formats: is there an overview with drill-down?)

Common problems:

• Titles that are just category labels ("Finance", "Operations") rather than insights
• No clear narrative thread - feels like a collection of charts, not a story
• The conclusion introduces new data instead of synthesising and recommending

---

Review Output Format

When providing a review, structure feedback as:

1. Overall assessment - 2-3 sentence summary of the biggest opportunities for improvement 2. Top 3 priority changes - The changes that would have the most impact, ranked 3. Section-by-section / chart-by-chart feedback - Specific, actionable items with clear recommendations 4. What's working well - Reinforce the things that are already effective

Always suggest _what to do instead_, not just what's wrong. Where possible, describe or sketch the improved version.