Use Charts to Answer Questions, Not Decorate Slides

(1) Bar Chart

Use it for: Comparing categories → media platforms, city audiences, ad versions. Tips:

• Go horizontal when category names are long.
• Start the axis at zero (don’t exaggerate differences).
• Keep to 10–12 bars max; group the rest as “Other.”

Examples:

• Click-through rate by headline type
• Sessions by platform (YouTube vs TikTok vs Instagram)

(2) Line Chart

Use it for: Anything that changes over time (daily, weekly, monthly). Tips:

• Smooth wild hourly spikes use weekly averages if needed.
• Label lines directly at the end; don’t make readers chase the legend.
• If your y-axis doesn’t start at zero, say so clearly.

Examples:

• News topic popularity across a campaign week
• Conversion rate trends over the last quarter

(3) Scatter Plot

Use it for: Showing relationships

Tips:

• Add a trendline to show direction, but remember: correlation ≠ causation.
• Too many dots? Use transparency, sampling, or hexbin plots.
• Add size or color for a third variable → but not all at once.

Examples:

• Video length vs. completion rate
• CPM vs. CTR across ad campaigns

(4) Stacked Bar Chart

Use it for: Showing composition (parts of a whole) over time or by category — for example, ad spend split across channels or topic mix across months. Tips:

• Keep one baseline consistent (e.g., total = 100%) so changes are easy to read.
• Limit to 3–5 segments per bar; more colors = instant chaos.
• Label percentages or totals directly — don’t make people estimate.

Examples:

• “Share of social media engagement by platform from January–June.”
• “Distribution of story topics (health, tech, politics, lifestyle) per quarter.”

(5) Bubble Chart

Use it for: Showing the relationship between three variables — where x and y represent two numeric dimensions (e.g., ad spend vs. engagement), and the size of the bubble represents a third one (e.g., audience reach or revenue). Tips:

• The area (not radius) of each bubble should represent the data — otherwise, big bubbles mislead the viewer.
• Keep the opacity slightly low so overlapping bubbles are visible.
• Limit to 8–10 bubbles to avoid clutter; label the key ones directly.

Examples:

• “Comparing marketing campaigns by spend (x), engagement (y), and impressions (bubble size).”
• “Visualizing countries by GDP per capita (x), life expectancy (y), and population size (bubble size).”

(6) Pie Chart

Use it for: Showing how a whole divides into parts — each slice represents a percentage or proportion of a total (like market share, traffic source, or budget breakdown). Tips:

• Limit to 3–5 slices — too many makes it unreadable.
• Use soft contrasting colors for clear separation.
• Order slices clockwise by size (largest starting at 12 o’clock).
• Avoid 3D or tilted pies — they distort proportions.

Examples:

• “Share of traffic from search, social, and direct visits.”
• “Budget allocation across departments: marketing, design, operations.”

(7) Histogram / Density Plot

Use it for: Showing the frequency distribution of one numeric variable — like reading time, watch duration, or post engagement. Tips:

• Pick bin sizes carefully (too few hides detail, too many adds noise).
• For smoother storytelling, overlay a density curve.
• Add annotations for key thresholds (e.g., “most users leave after 30 sec”).

Examples:

• “Distribution of video watch time.”
• “Email open rates clustered between 10–20%.”

(8) Map

Use it for: Only when geography actually matters — not just because maps look cool. Tips:

• Choose choropleth (color-filled) for regional averages, or point maps for locations.
• Avoid misleading readers with unequal area sizes — use per-capita or rate, not totals.
• Always include a legend and scale, or the map is just decoration.

Examples:

• “COVID-19 vaccination rate by state.”
• “Podcast downloads by region.”

(9) Small Multiples

Use it for: Repeating the same type of chart with identical scales — ideal for clean, side-by-side comparisons. Tips:

• Use consistent axes and colors so viewers can scan differences fast.
• Works best for 3–8 subplots; more, and the eye gives up.
• Add a short caption per chart to avoid confusion.

Examples:

• “Social media performance per country — same metrics, same scale.”
• “Engagement trend for each news section.”

(10) Gauges / Speedometers

Use it for: Dashboard metrics that show one value vs. a target (e.g., goal progress). Tips:

• Visually appealing but low information density — use only when one number really matters.
• For real insight, combine with trend indicators (like small sparklines).
• Avoid comparisons or complex data — it’s decorative, not analytical.

Examples:

• “Donation goal progress (75% reached).”
• “Monthly KPI tracker for ad impressions.”