What is Business Analytics? A Simple Explanation for 2026

-- Business question: which regions are underperforming?
SELECT
    region,
    SUM(revenue) AS total_revenue,
    COUNT(*) AS total_orders,
    ROUND(AVG(order_value), 2) AS avg_order_value
FROM sales
WHERE order_date >= '2025-01-01'
GROUP BY region
ORDER BY total_revenue DESC;

# Python — quick KPI summary for a business report
import pandas as pd

df = pd.read_csv("sales.csv")

kpis = {
    "Total Revenue": df["revenue"].sum(),
    "Total Orders": len(df),
    "Avg Order Value": round(df["revenue"].mean(), 2),
    "Top Region": df.groupby("region")["revenue"].sum().idxmax(),
}
print(kpis)

Business Insight (example output):

"North region contributes 42% of total revenue but has
 the highest average order value (₹1,240). South region
 is underperforming at 18% with an average order value of
 ₹620. Recommendation: investigate South region's customer
 acquisition and pricing strategy."

→ This is business analytics in action: data → insight → action.

-- Descriptive: what happened last quarter?
SELECT
    quarter,
    SUM(revenue) AS revenue,
    COUNT(DISTINCT customer_id) AS customers
FROM sales
WHERE year = 2025
GROUP BY quarter
ORDER BY quarter;

-- Diagnostic: why did Q3 revenue drop?
SELECT
    channel,
    SUM(revenue) AS revenue,
    COUNT(*) AS orders
FROM sales
WHERE year = 2025 AND quarter = 3
GROUP BY channel
ORDER BY revenue;

-- Insight: paid ads revenue dropped 40% in Q3.
-- Root cause: increased ad CPC and lower conversion rate.

# Predictive: forecast next quarter's revenue
from sklearn.linear_model import LinearRegression
import numpy as np

# quarterly revenue (last 8 quarters)
X = np.arange(8).reshape(-1, 1)
y = [120, 135, 128, 150, 162, 175, 168, 185]

model = LinearRegression().fit(X, y)
next_q = model.predict([[8]])
print(f"Next quarter forecast: {next_q[0]:.0f}")

# Prescriptive: which campaign mix maximises revenue?
# (illustrative — a real model would be a constrained optimiser)

options = {
    "Email campaign":  {"cost": 5000,  "expected_revenue": 18000},
    "Paid ads":        {"cost": 20000, "expected_revenue": 42000},
    "Influencer":      {"cost": 12000, "expected_revenue": 26000},
}

# Rank by return on investment
for name, data in sorted(options.items(),
                          key=lambda x: x[1]["expected_revenue"] / x[1]["cost"],
                          reverse=True):
    roi = data["expected_revenue"] / data["cost"]
    print(f"{name}: ROI = {roi:.2f}x")

-- Monthly active users (MAU) and growth
SELECT
    DATE_TRUNC('month', login_date) AS month,
    COUNT(DISTINCT user_id) AS mau
FROM logins
GROUP BY month
ORDER BY month DESC;

# Cohort retention analysis with Pandas
import pandas as pd

df = pd.read_csv("users.csv")
df["signup_month"] = pd.to_datetime(df["signup_date"]).dt.to_period("M")

cohort = df.groupby("signup_month").agg(
    users=("user_id", "nunique"),
    retained=("last_active_days_ago", lambda x: (x < 30).sum()),
)
cohort["retention_rate"] = cohort["retained"] / cohort["users"]
print(cohort)

# A/B test: did the new landing page increase conversion?
from scipy.stats import chi2_contingency

# [converted, not_converted]
control   = [1200, 8800]   # 12% conversion
treatment = [1450, 8550]   # 14.5% conversion

chi2, p, dof, expected = chi2_contingency([control, treatment])
print(f"p-value: {p:.4f}")

if p < 0.05:
    print("Result is statistically significant — ship it!")
else:
    print("Not significant — keep testing.")

Sample Weekly Plan (10-12 hours/week):
Mon  — SQL practice (1 hour)
Tue  — Statistics / theory (1 hour)
Wed  — SQL practice (1 hour)
Thu  — Excel / Power BI (1 hour)
Fri  — Case study (1 hour)
Sat  — Project work (3 hours)
Sun  — Review + plan next week (1 hour)

Portfolio Project Ideas:

1. Sales dashboard — build KPIs and a summary for a
   fictional e-commerce business.
2. Customer churn analysis — use SQL + Python to
   identify churn drivers and recommend retention actions.
3. A/B test analysis — analyse a public A/B test dataset
   and write a recommendation memo.
4. Marketing funnel analysis — find drop-offs and suggest
   optimisation.
5. Financial performance report — quarterly variance
   analysis with a one-page executive summary.

Common Interview Topics (Business Analyst):

1. Explain a project you did — problem, approach, impact.
2. SQL: JOINs, GROUP BY, subqueries, window functions.
3. Business metrics: define MAU, DAU, ARPU, churn.
4. Case study: "How would you increase X?"
5. A/B test: p-value, sample size, statistical significance.
6. Dashboard design: what would you show an executive?
7. How do you communicate data to non-technical people?