7 Myths About Data Analytics Career Nobody Tells You

Vague question: "Why are sales down?"

Break it down:
1. Down compared with what? (last month, last year, target)
2. Down everywhere, or in some regions / categories / channels?
3. Fewer orders, or smaller orders?
4. Fewer customers, or fewer repeat customers?
5. Any known changes? (price, stock, campaign, website issue)

Each line becomes one query or one chart.

-- Fewer orders or smaller orders? Compare two months by region
SELECT region,
       strftime('%Y-%m', order_date) AS month,
       COUNT(*)                       AS orders,
       ROUND(AVG(amount), 2)          AS avg_order_value,
       SUM(amount)                    AS revenue
FROM orders
WHERE order_date >= '2026-09-01' AND order_date < '2026-11-01'
GROUP BY region, month
ORDER BY region, month;

def insight(headline, evidence, comparison, recommendation, caveat):
    return (f"{headline}\n"
            f"- Evidence: {evidence}\n"
            f"- Compared with: {comparison}\n"
            f"- Recommendation: {recommendation}\n"
            f"- Caveat: {caveat}")

print(insight(
    "Repeat orders fell mainly in the North region.",
    "North repeat rate dropped while other regions held steady",
    "the previous two months",
    "Check delivery times in North and review recent complaints",
    "Two months of data; seasonal effects possible"))

import pandas as pd

df = pd.read_csv("orders.csv", parse_dates=["order_date"])
checks = {
    "rows": len(df),
    "duplicate order_ids": int(df["order_id"].duplicated().sum()),
    "missing amounts": int(df["amount"].isna().sum()),
    "negative amounts": int((df["amount"] < 0).sum()),
    "latest date": str(df["order_date"].max().date()),
}
for k, v in checks.items():
    print(f"{k:22} {v}")
# Run checks like these before trusting any chart.

-- Suppose an assistant suggested this to find 'average order value':
--   SELECT AVG(amount) FROM order_items;
-- Problem: order_items has one row per ITEM, not per order.

-- Correct: total each order first, then average
WITH order_totals AS (
  SELECT order_id, SUM(amount) AS order_total
  FROM order_items
  GROUP BY order_id
)
SELECT ROUND(AVG(order_total), 2) AS avg_order_value
FROM order_totals;

DOMAIN_METRICS = {
    "E-commerce": ["conversion rate", "average order value", "repeat rate", "return rate"],
    "Banking": ["active accounts", "digital adoption", "delinquency rate"],
    "EdTech": ["course completion", "weekly active learners", "renewal rate"],
    "Logistics": ["on-time delivery", "failed deliveries", "cost per shipment"],
}
for domain, metrics in DOMAIN_METRICS.items():
    print(f"{domain:11} {', '.join(metrics)}")

Thinking   Break vague questions into steps (practise weekly)
SQL        Joins, aggregates, CTEs, window functions
Excel      Lookups, pivots, data cleaning
BI tool    One of Power BI or Tableau, used well
Python     Basic pandas for repeated tasks
Writing    Five-line insight summaries
Domain     Key metrics of one target industry

Mon  One vague business question -> written breakdown
Tue  SQL practice on a public dataset
Wed  One chart + five-line insight summary
Thu  Check an AI-generated query for errors
Fri  Read about one metric in your target industry
Weekend  Project work

Data Analyst
  -> Senior Analyst (deeper analysis, mentoring)
  -> BI Developer (data models, dashboards at scale)
  -> Analytics Engineer (SQL pipelines, dbt)
  -> Data Scientist (statistics, ML)
  -> Product / Business roles (decision-making with data)