Choose Right Chart

Technical Explanation

Different chart types serve different purposes. Choosing the wrong chart can obscure insights or create confusion.

Chart Selection Guide

Data Type	Best Chart	Avoid
Trend over time	Line	Pie
Part to whole	Pie, Stacked Bar	Line
Comparison	Bar	Pie
Distribution	Histogram, Box	Pie
Relationship	Scatter	Bar
Single KPI	Number + Trend	Table

Common Chart Mistakes

• Using pie charts for many categories (>5)
• Using line charts for categories without time
• Using bar charts with inconsistent scales
• 3D charts (distort perception)

---

Code Examples

Using the CatCafe dataset:

-- Data for different chart types

-- LINE CHART: Revenue trend over time
SELECT
    DATE_TRUNC('month', order_date) as month,
    SUM(total_amount) as revenue
FROM orders
WHERE status = 'completed'
GROUP BY DATE_TRUNC('month', order_date)
ORDER BY month;

-- BAR CHART: Revenue by category
SELECT
    category,
    SUM(total_amount) as revenue
FROM products
GROUP BY category
ORDER BY revenue DESC;

-- PIE CHART: Order distribution (only if <5 categories)
SELECT
    status,
    COUNT(*) as orders
FROM orders
GROUP BY status;
-- Only works well for 2-4 categories

-- STACKED BAR: Revenue by segment over time
SELECT
    DATE_TRUNC('month', order_date) as month,
    segment,
    SUM(total_amount) as revenue
FROM orders o
JOIN customers c ON o.customer_id = c.id
WHERE status = 'completed'
GROUP BY DATE_TRUNC('month', order_date), segment
ORDER BY month;

-- SCATTER: Order value vs customer tenure
SELECT
    c.customer_tenure_days as x,
    o.total_amount as y
FROM orders o
JOIN customers c ON o.customer_id = c.id
WHERE status = 'completed';

-- HISTOGRAM: Distribution of order values
SELECT
    WIDTH_BUCKET(total_amount, 0, 500, 10) as bucket,
    (WIDTH_BUCKET(total_amount, 0, 500, 10) - 1) * 10 as range_start,
    WIDTH_BUCKET(total_amount, 0, 500, 10) * 10 as range_end,
    COUNT(*) as orders
FROM orders
GROUP BY WIDTH_BUCKET(total_amount, 0, 500, 10)
ORDER BY bucket;

---

The Cat Analogy

Choosing how to display cat weight data:

LINE CHART ✓ (Weight over time):

Whiskers' weight:
Jan: 10 lbs ────
Feb: 10.5 lbs ──�──
Mar: 11 lbs ─────
→ Shows trend clearly

PIE CHART ✗ (Weight vs other cats):

How much each cat eats:
Whiskers eats 30%  ────┐
Mochi eats 25%       │
Luna eats 25%        │ Don't do this!
Shadow eats 20%       │
→ Hard to compare!

BAR CHART ✓ (Compare cats):

Cats ranked by weight:
Whiskers  ████████████ 11 lbs
Mochi     ██████████   9 lbs
Luna      ████████     8 lbs
→ Easy to compare

---

Exercises

Exercise 1

What chart type would you use for:

• Monthly revenue for the past 12 months
• Revenue split between product categories
• Distribution of order values
• Relationship between customer age and spending

Exercise 2

Why should you avoid pie charts with more than 5 categories?

Exercise 3

What chart would you use to show how revenue by segment has changed over time?

---

Key Takeaways

• Line charts: Trends over time
• Bar charts: Comparisons between categories
• Pie charts: Part-to-whole (only 2-4 categories)
• Scatter plots: Relationships between variables
• Histograms: Distribution of values
• Avoid 3D charts, they distort perception