Exercises: Storytelling with Data

These exercises are non-programming. They build the narrative discipline that turns a collection of charts into a coherent data story. Do Part B with a stack of sticky notes or index cards — the physical rearrangement is the point.

Part A: Conceptual (7 problems)

A.1 ★☆☆ | Recall

Define "data storytelling" in your own words. Describe the three structural elements that the chapter identifies as part of the practice.

Guidance Data storytelling is the deliberate arrangement of charts, text, and transitions into a coherent narrative that walks a reader through a complete argument. The three elements are: (1) the charts themselves (meeting standards of Chapters 6-8), (2) the sequence of charts arranged to walk the reader through a logical argument, and (3) the framing around the charts (titles, subtitles, annotations, transitional text) that ties the individual charts into a single story.

A.2 ★☆☆ | Recall

Explain the difference between "data analysis" and "data storytelling" in your own words. Why does the chapter argue that they are separate skills?

Guidance Data analysis is the process of finding the story in the data — using exploration to learn what the data shows. Data storytelling is the process of telling the story to an audience — using presentation to convey the understanding to people who did not do the analysis. They require different tools and different dispositions: analysis values speed and flexibility, storytelling values clarity and intentionality. Applying analysis tools to storytelling (showing exploratory charts to executives) or storytelling tools to analysis (polishing exploratory charts excessively) both fail.

A.3 ★★☆ | Understand

Describe the three-act narrative structure as applied to data stories. For each act, explain what the chart maker is trying to accomplish and what the reader is supposed to experience.

Guidance Act 1: Context. The chart maker establishes the baseline, the background, the "normal state" of the data. The reader learns what the story is about. Act 2: Evidence. The chart maker shows the main finding — the comparison, trend, or anomaly that is the reason for the story. The reader experiences the tension of discovery. Act 3: Implications. The chart maker shows what the finding means and what the reader should do about it. The reader experiences resolution — understanding the significance of what they have seen.

A.4 ★★☆ | Understand

What is "the Big Idea" as Cole Knaflic uses the term? Why does the chapter argue that you should state the Big Idea before you design any charts?

Guidance The Big Idea is the single sentence that captures the whole point of the data story. It should be a complete declarative statement with specific content, not a vague topic label. Stating the Big Idea first forces you to be clear about what you are trying to say. It becomes the guiding principle for every subsequent decision: which charts to include (only those that support the Big Idea), which to exclude (those that do not), which to emphasize (the carrier of the main claim).

A.5 ★★☆ | Analyze

The chapter identifies four audience types: technical, executive, general, and mixed. For each type, describe one specific design decision (vocabulary, chart complexity, level of context, or something else) that would differ from the other types.

Guidance Technical audiences: use precise statistical vocabulary; include error bars and confidence intervals; assume familiarity with complex chart types. Executive audiences: short presentations with clear bottom lines; minimal methodology; headline finding plus recommendation. General audiences: plain language without jargon; simple chart types with heavy annotation; connect to reader's experience. Mixed: progressive disclosure — start with content everyone can understand, layer detail for those who want more.

A.6 ★★☆ | Understand

Explain Shneiderman's mantra ("Overview first, zoom and filter, then details on demand") and describe how it applies to a static report (not an interactive dashboard).

Guidance Overview first: the reader should get the Big Idea from the opening — headline, abstract, executive summary. Zoom and filter: the body of the document contains specific evidence and supporting charts that different readers will focus on. Details on demand: footnotes, appendices, and methodology sections contain the full detail for readers who want it. A well-structured report can be read at three different depths — headline, body, or full text with appendices — and each level provides something valuable.

A.7 ★★★ | Evaluate

The chapter identifies three "temptations" in data storytelling: cherry-picking, overstatement, and framing manipulation. For each, describe a specific scenario in your own work (or in work you have reviewed) where the temptation was present. What would the ethical response look like?

Guidance The exercise is about self-awareness. Identify real scenarios, not hypothetical ones. Cherry-picking: showing only supporting evidence and omitting complications. Overstatement: using stronger language than the data supports. Framing manipulation: choosing a frame that makes the finding seem more impressive or alarming than an equally valid alternative. The ethical response for each: add the omitted context, use accurate language, acknowledge the alternative framing.

Part B: Applied — Storyboarding and Structure (10 problems)

B.1 ★★☆ | Apply

Write the Big Idea for a hypothetical data story about the following: "Our company's customer satisfaction scores have declined slightly over the past year, from 8.2/10 to 7.9/10, with the largest decline in the enterprise customer segment." State the Big Idea as a single declarative sentence of 15-25 words.

Guidance A good Big Idea is specific and carries the finding. Example: "Customer satisfaction has declined from 8.2 to 7.9 over the past year, driven almost entirely by dissatisfaction among enterprise customers who account for 40% of revenue." Avoid vague versions like "Customer satisfaction needs attention" or "Customers are less happy."

B.2 ★★☆ | Apply

Given the Big Idea from B.1, storyboard a four-chart sequence (Act 1: context; Act 2: evidence; Act 3: implications) to tell the story. For each chart, write (a) a one-line description of the chart, (b) the action title, and (c) which act of the story it belongs to.

Guidance Chart 1 (context): "Customer satisfaction over the past three years" — line chart showing the long trend. Action title: "Customer Satisfaction Was Stable Until 2024." Chart 2 (evidence): "Satisfaction by customer segment" — bar chart showing segment-level breakdown. Action title: "Enterprise Customer Satisfaction Dropped Sharply While Other Segments Held Steady." Chart 3 (evidence): "Common themes in enterprise customer complaints" — ranked bar chart or word cloud of top complaints. Action title: "Enterprise Complaints Cluster Around Onboarding and Support Response Time." Chart 4 (implications): "Projected impact of addressing the two top issues" — scenario comparison. Action title: "Fixing Onboarding Could Recover 0.2 Points of Satisfaction Within Six Months."

B.3 ★★☆ | Apply

Same data as B.1. Write three versions of the story for three different audiences: (a) a technical audience (data analysts reviewing the methodology), (b) an executive audience (the CEO and leadership team), (c) a general audience (customers reading a blog post about how you handle feedback). For each version, describe in 2-3 sentences how the story differs.

Guidance Technical version: includes methodology (how was satisfaction measured?), confidence intervals, comparison to industry benchmarks, subgroup analyses. Executive version: three slides with headline finding, segment breakdown, and recommendation. General version: plain-language article focused on what the company is doing in response to feedback; minimal charts, heavy narrative.

B.4 ★★☆ | Apply

You are designing a 10-minute executive presentation on the performance of three product lines. The CEO will see only the first two slides in detail; the rest will be background. What goes on slide 1 (the hero) and slide 2 (the primary supporting chart)? What goes on slides 3-10?

Guidance Slide 1: the main finding as a single action title and a single chart showing revenue growth (or decline) for the three product lines. Slide 2: segment or region breakdown of the main finding to show which specific areas are driving the trend. Slides 3-10: supporting details (per-product trends, cost analyses, customer segments, competitive context, risks, recommendations) that the CEO can skip but that are available for questions.

B.5 ★★☆ | Apply

Take one of the following narrative situations and sketch the storyboard as a sequence of 4-6 sticky notes (one per chart). Write out what each sticky note says.

  • (a) "Our marketing campaign in Q3 increased conversion rates by 23% but didn't increase total customers."
  • (b) "COVID booster uptake has plateaued at 45% in our country, despite 80% first-dose coverage."
  • (c) "Climate change is accelerating ocean acidification, which is causing coral bleaching."
Guidance For each, identify the three acts and sketch each chart's title. Example for (a): 1) "Total Customer Count Over 2024" (context); 2) "Conversion Rate Jumped 23% in Q3" (evidence); 3) "But Visitor Traffic Declined by 20% in Q3" (complication); 4) "Net Customer Growth Was Flat Despite the Conversion Improvement" (resolution); 5) "Recommended Focus: Driving Traffic in Q4" (call to action).

B.6 ★★☆ | Apply

The chapter describes the "grayed-out strategy" for visual emphasis. Sketch a sequence of three charts using the same underlying dataset (say, 50 U.S. states and some metric), where each chart emphasizes a different subset of states by graying out the others. For each chart, describe which states are emphasized and what story the chart tells.

Guidance Chart 1: all 50 states in gray, no emphasis. Story: "Here is the baseline for all states." Chart 2: five states with the highest rate in bright color, others in gray. Story: "These five states have the highest rates." Chart 3: one specific state (the audience's state) in bright color, others in gray. Story: "Here is where your state fits." The same dataset, three emphases, three stories.

B.7 ★★★ | Apply

Take a data story you have told recently and identify the Big Idea in a single sentence. Was this Big Idea clear in the original story? If not, how would the story change if the Big Idea had been stated explicitly at the beginning?

Guidance Most presentations do not have a clearly stated Big Idea — they have a topic and a collection of findings. Write the Big Idea retroactively for a specific story you told. Then consider how the story would have been different if the Big Idea had been the guiding principle: which charts would you have cut, which would you have added, which would you have emphasized?

B.8 ★★★ | Analyze

Find a data story from a news outlet (NYT, Washington Post, The Economist, FT) or a blog (Our World in Data, Pew Research). Analyze its narrative structure. What is the Big Idea? What is the three-act structure? What audience is the story written for? What visual emphasis techniques does it use?

Guidance Pick a specific story and annotate it. Identify the Big Idea from the headline or the lead paragraph. Identify the context charts, the evidence charts, and the implications charts. Note the audience assumptions (vocabulary, context, complexity). Note the emphasis techniques (grayed-out lines, highlighted elements, annotated callouts). This exercise is about learning to see narrative structure in professional work.

B.9 ★★★ | Create

Design a 5-chart story about a dataset you have worked with recently. Be specific about: (a) the Big Idea in one sentence, (b) the target audience, (c) the three-act structure, (d) the action title for each chart, (e) the visual emphasis technique you would use in each chart.

Guidance The point is to apply the chapter's framework to real data that you know. The five charts do not all need to be different chart types — they could all be line charts, for example. What matters is that the sequence tells a coherent story with a clear Big Idea, and that each chart serves a specific role in the narrative.

B.10 ★★★ | Create

Rewrite the climate story from Section 9.8 (the five-chart sequence) for a general public audience instead of a technical one. What changes in the vocabulary, the chart types, the amount of context, and the level of statistical rigor?

Guidance For a general audience: replace "temperature anomaly" with "how much warmer it is than average"; use simpler chart types (line charts with clear annotations rather than dual time series or scatter plots); spend more time on the context chart to explain what "global temperature" means; reduce error bars and confidence intervals or explain them in plain language; frame the implications in terms the reader can relate to (their lifetime, their location, their choices).

Part C: Synthesis and Judgment (5 problems)

C.1 ★★★ | Evaluate

The chapter argues that the order in which you present charts is itself an argument. Reconstruct this argument in your own words. Then describe a specific case where you believe the order of presentation would substantially change the reader's interpretation of the same charts.

Guidance The argument: sequencing is a choice; different orders produce different impressions; there is no neutral sequence. The chart maker who pretends the sequence is arbitrary is hiding an editorial decision. Example: a story about climate that starts with the projection (Act 3) before showing the context (Act 1) will feel like unfounded alarmism; the same charts in the three-act order feel like a grounded argument. The order matters because it shapes the reader's interpretation of the same evidence.

C.2 ★★★ | Analyze

The chapter identifies cherry-picking, overstatement, and framing manipulation as the three main ethical temptations in data storytelling. Are there other ethical failures in data storytelling that the chapter does not explicitly name? Describe one and explain how it differs from the three listed.

Guidance Possible additions: omission of confidence / uncertainty, false precision (stating numbers to more decimal places than the data supports), reliance on statistical language that hides weak evidence, false attribution of causation, selective sourcing of data. For each, explain how it is similar to and different from the three listed temptations.

C.3 ★★★ | Create

Write a short "storytelling checklist" (6-10 items) that a colleague could use to evaluate a draft data story before publication. The checklist should capture the most important lessons from this chapter in a form that is scannable and actionable.

Guidance Candidate items: Is the Big Idea stated in a single sentence at the top? Is the audience identified and the complexity matched? Does the sequence follow a three-act structure? Is each chart necessary for the story? Can you defend every claim against a skeptical reader? Is the visual emphasis directing attention to the right elements? Is the level of context appropriate for the audience? Does the story end with a clear resolution or call to action? Does each individual chart meet the standards of Chapters 6-8?

C.4 ★★★ | Evaluate

The chapter argues that data storytelling is persuasion, but that ethical persuasion is possible. Do you agree? Construct an argument for or against the position that any persuasive data storytelling is, to some degree, manipulation.

Guidance This is a judgment question. The chapter's position is that persuasion aligned with the evidence is ethical; persuasion that distorts the evidence is manipulation. A counter-argument would be that any framing is manipulation because it influences the reader beyond what the "raw data" would imply. Consider the implications of each position: if all framing is manipulation, the chart maker has to produce unframed charts, which may be impossible. If framing is ethical as long as it matches the evidence, the chart maker needs the discipline to know when it does and does not match.

C.5 ★★★ | Create

Over the past eight chapters of this book, you have built a climate story from a single ugly default chart into a five-chart narrative sequence. Write a one-paragraph reflection on what you have learned about data visualization as a discipline, using the evolution of the climate story as your example. The reflection should tie together principles from multiple chapters of Part II.

Guidance A good reflection will acknowledge that the individual principles (declutter, typography, composition, narrative) are complementary rather than alternatives, and will describe how the sequence of chapters has built a complete practice. The climate story has evolved because each chapter added a layer, not because any single layer was enough on its own. Good answers will also note that the discipline is iterative — each improvement revealed the next opportunity for improvement — and that the final story is stronger than any individual chart could have been.

These exercises build the narrative discipline that separates "a collection of charts" from "a data story." Storyboarding exercises (Part B) are particularly valuable because the physical act of arranging sticky notes forces decisions that mental planning leaves unresolved. Do at least three of the Part B exercises on sticky notes or index cards before moving on to Part III.