You are viewing a public preview
Sign in to save your progress, do assignments, join discussions and get feedback.
Defining business problems, analytical questions, metrics and key messages; selecting and organizing visual evidence; structuring a narrative from context and evidence to analysis, conclusions and recommendations; titles, annotations, explanations and narrative flow. The slides walk through the data storytelling process step by step.
Learning outcomes
The lecture slides for this lesson are below.
By the end of this lesson you can:
Key idea: A data story is built, not found. Following a process from goal to decision keeps the story focused, honest and useful to the audience.
The lecture presents eight steps:
| Step | What you do |
|---|---|
| 1. Goal | Define a clear goal and a hypothesis to test; know what message or insight the data should convey |
| 2. Data collection and preparation | Collect data from sources, clean and transform it into a well-structured dataset |
| 3. Data analysis and exploration | Use statistics, hypothesis testing and exploration to find patterns, trends and relationships; identify the critical insights |
| 4. Data visualization | Create compelling visuals and choose the chart types most relevant to the data |
| 5. Audience consideration | Tailor the story to the audience with relevant messages |
| 6. Communication | Present dynamically: presentations, interactive reports and dashboards |
| 7. Feedback and iteration | Collect input, check that audience needs are met, improve, stay aligned with objectives |
| 8. Actions and decision-making | Link the story to concrete actions and decisions |
Steps 2 to 4 repeat what you did in Modules 1 to 4. What is new is that the goal (step 1) and the audience (step 5) shape everything else, and that the process does not end with the chart: it ends with a decision.
Start from a business problem, then define analytical questions and the metrics that answer them. A hypothesis keeps analysis focused: "Profit is weak because of deep discounts" tells you which data and charts to look at, and the data may confirm or reject it. After analysis, write the key message as one sentence.
Choose the smallest set of visuals that proves the message. For each visual, ask: which part of the argument does it support? If the answer is "none", remove it. Order the visuals so each one answers the question raised by the one before.
A common narrative structure:
Choose the format the audience will use: a live presentation, an interactive dashboard or a Tableau Story. After presenting, collect feedback: Did they understand the message? Did they trust the evidence? Did they have the information to decide? Revise and present again if needed. Finally, connect the story to actions: who will do what, by when, and which metric will show whether it worked.
Build a story for the regional managers of the course's office-supplies retailer, following the eight steps.
Flashcards · Building a Story from Data
8 wordsEight steps: goal, data preparation, analysis, visualization, audience, communication, feedback, action.
A testable statement that focuses the analysis; the data confirm or reject it.
The selected charts that each support one part of the argument.
The order of a story: context, evidence, analysis, conclusion, recommendation.
A chart title that states the finding rather than describing the content.
A note attached to a mark, point or area that draws attention to it.
Collecting audience input and revising the story so it meets their needs.
The final step linking the story to who does what, by when, and how it is measured.