Data Visualization Basics

• Quickly communicate trends, outliers, and comparisons to non-technical stakeholders.
• Support decision meetings with dashboards and simple, accurate visuals.
• Spot risks or opportunities faster (e.g., sudden dips, seasonality, anomalies).
• Build trust by avoiding misleading visuals and presenting data ethically.

Scenario: Which channel delivered the highest Q2 revenue?

Spreadsheet steps:

• Create a table with Channel and Revenue.
• Sort by Revenue descending.
• Insert a bar/column chart; ensure Y-axis starts at 0.
• Title: "Q2 revenue by channel — Online leads" (answer-focused).
• Annotate the top bar with its value.

Python (optional):

import matplotlib.pyplot as plt
channels = ["Online", "Retail", "Partners", "Direct"]
revenue = [420, 310, 190, 160]
order = sorted(range(len(revenue)), key=lambda i: revenue[i], reverse=True)
channels = [channels[i] for i in order]
revenue = [revenue[i] for i in order]
plt.figure(figsize=(6,4))
plt.bar(channels, revenue, color="#4C78A8")
plt.title("Q2 revenue by channel — Online leads")
plt.ylabel("Revenue (k)")
plt.ylim(0, max(revenue)*1.15)
for i,v in enumerate(revenue):
    plt.text(i, v+5, f"{v}", ha='center')
plt.show()

Scenario: Daily signups for 60 days. We want the trend, not noise.

Spreadsheet steps:

• Have Date in A, Signups in B.
• In C2: =AVERAGE(B2:B8) to compute 7-day moving average; fill down.
• Line chart with two series: Daily and 7-day MA; lighter color for Daily, bold for MA.
• Title: "Signups trending upward — 7-day average".

Python (optional):

import pandas as pd, numpy as np, matplotlib.pyplot as plt
np.random.seed(0)
days = pd.date_range("2024-01-01", periods=60)
signups = pd.Series(50 + np.arange(60)*0.8 + np.random.normal(0,5,60))
ma7 = signups.rolling(7).mean()
plt.figure(figsize=(7,4))
plt.plot(days, signups, color="#9ecae1", label="Daily")
plt.plot(days, ma7, color="#08519c", label="7-day MA", linewidth=2)
plt.title("Signups trending upward — 7-day average")
plt.ylabel("Signups")
plt.legend()
plt.show()

Scenario: Ticket resolution times (minutes). Are we within our target?

• Bins: choose sensible widths (e.g., 0–15, 15–30, ...). Consistent bin size helps interpretation.
• Add a vertical line for the SLA target (e.g., 45 minutes) and annotate % within target.
• Title: "Resolution times cluster at 30–45 min; 78% within SLA".

import numpy as np, matplotlib.pyplot as plt
x = np.clip(np.random.normal(40, 12, 400), 5, 120)
plt.figure(figsize=(6,4))
plt.hist(x, bins=8, color="#74c476", edgecolor="white")
plt.axvline(45, color="#d62728", linestyle="--", label="SLA 45m")
plt.title("Resolution times cluster at 30–45 min; ~78% within SLA")
plt.xlabel("Minutes")
plt.ylabel("Tickets")
plt.legend()
plt.show()

Scenario: Does discount rate correlate with conversion rate?

• Use scatter; add a trendline; label notable outliers.
• Color by segment if helpful (keep 3–5 colors max).

import numpy as np, matplotlib.pyplot as plt
np.random.seed(1)
discount = np.random.uniform(0, 30, 60)
conversion = 2 + 0.3*discount + np.random.normal(0, 2, 60)
plt.figure(figsize=(6,4))
plt.scatter(discount, conversion, color="#ff7f0e")
# Trendline
m,b = np.polyfit(discount, conversion, 1)
xs = np.array([0,30])
plt.plot(xs, m*xs + b, color="#1f77b4", linewidth=2)
plt.title("Higher discounts associate with higher conversion")
plt.xlabel("Discount (%)")
plt.ylabel("Conversion (%)")
plt.show()

Scenario: A simple performance dashboard.

• Row 1 (KPI cards): Revenue, Conversion, Avg Order Value.
• Row 2: Line chart (revenue trend) and bar chart (top categories).
• Row 3: Notes/insights with 2–3 bullet takeaways.
• Use consistent colors, aligned axes, and minimal gridlines.

Title the dashboard with the decision context, e.g., "Q2 Growth Review — Marketing focus".

• Titles answer the business question.
• Axes and units are clear; bars start at zero.
• Only one highlight color draws attention to the key insight.
• Accessible color choices; labels readable from 1–2 meters on a projector.