Wisdom is the action-oriented peak of the data, information, knowledge, wisdom (DIKW) pyramid. Data becomes information when it is organized and meaningful; information becomes knowledge when interpreted in context; knowledge becomes wisdom when it is enriched with ethics and leads the audience toward action. A data story at this level does more than display a finding: it states what should happen next. section 1.4 section 1.5
A call to action is the immediate expression of wisdom. It can ask a question, propose a solution, invite discussion, or direct the audience to take a concrete step. Next steps make that invitation more complete by identifying alternatives, actions, or resources justified by the analysis rather than by the chart alone. section 1.4.3 section 3.3.3 section 9.1.1
Sound wisdom is not produced automatically by transforming knowledge. It combines knowledge synthesis with experience and good judgment: weighing risks and benefits, considering multiple perspectives, and anticipating possible outcomes. This makes wisdom a decision-making layer, not merely a persuasive title or annotation. section 9.0 section 9.1
Useful forms of next steps include:
- Ask for support, such as inviting participation in a survey or drawing on audience expertise. section 9.1.1
- Provide different options so the audience can compare possible ways forward. section 9.1.1
- Enable free interaction through an interactive chart, allowing the audience to explore data and form conclusions. section 9.1.1
- Encourage further learning by linking to resources, studies, or references. section 9.1.1
From pet counts to a campaign decision
Suppose event data shows that 9 of 10 participating pets are cats and only 1 is a dog. That count is data; the proportion is information. Context explains that dog owners were poorly reached because dog-focused advertising was weak. Wisdom changes a descriptive title such as “9 pets out of 10 are cats” into an actionable recommendation: increase the advertising campaign on dog-related websites. section 1.4.3
Add a recommendation to an Altair story
For a chart showing growing tourist arrivals in Portugal, a textual annotation can answer a decision-maker’s question directly: consider building a new swimming pool. The recommendation should remain proportionate to the evidence: a simplified chart may omit countries and years needed for a major investment decision. In a fuller design, place a companion bar chart beneath the main visualization to compare strategies and their proposed success percentages. section 2.3.3 section 3.3.3
Listing 2.16 Adding the call to action to the chart
text_cta = f"""Since the number of tourists arriving in Portugal is constantly expanding,
you can think about building a new swimming pool!"""
df_cta = pd.DataFrame([{'text' : text_cta}])
cta = alt.Chart(df_cta
).mark_text(
lineBreak='\n',
align='left',
fontSize=20,
y=0,
x=0,
color='#81c01e'
).encode(
text='text:N'
).properties(
width=200
)
final_chart = ((commentary & airports) | ((chart + annotation) & cta)
)
final_chart.save('chart.html')
Use the listing as a compact pattern for turning a recommendation into a separate Altair text chart, then composing it with the evidence chart. section 2.3.3
Generative AI can help draft engaging calls to action, but the analytical responsibility remains with the storyteller. For a fake-news story, a prompt can turn a warning into an audience-directed title; review the result so that its action, framing, and claims fit the evidence and audience. section 5.5.3 section 1.5
