Top 10 Business Analytics Interview Questions with Answers

Q2. Four types of analytics (with one e-commerce example)

Descriptive  - What happened?
               Sales fell 8% last month.
Diagnostic   - Why did it happen?
               Drop came from mobile users after an app update.
Predictive   - What is likely to happen?
               If unchanged, next month may fall further.
Prescriptive - What should we do?
               Fix the checkout bug; test a recovery offer.

(Numbers are illustrative in an answer - say so.)

Q4. Choosing KPIs for a new food delivery feature

1. Goal: increase repeat orders from new users
2. North-star metric: % of new users ordering again in 30 days
3. Supporting metrics:
   - Time to first order
   - Order completion rate
   - Average delivery time
4. Guardrail metrics (must not get worse):
   - Cancellation rate
   - Customer complaints
5. Review: weekly dashboard, monthly deep-dive

-- Q5: Top 3 products by revenue in each region
WITH product_rev AS (
  SELECT region, product, SUM(amount) AS revenue
  FROM sales
  GROUP BY region, product
)
SELECT region, product, revenue
FROM (
  SELECT *, RANK() OVER (PARTITION BY region ORDER BY revenue DESC) AS rnk
  FROM product_rev
) ranked
WHERE rnk <= 3;

-- Q6: Month-over-month revenue growth (PostgreSQL)
WITH monthly AS (
  SELECT DATE_TRUNC('month', sold_on) AS month, SUM(amount) AS revenue
  FROM sales
  GROUP BY 1
)
SELECT month, revenue,
  ROUND(100.0 * (revenue - LAG(revenue) OVER (ORDER BY month))
        / NULLIF(LAG(revenue) OVER (ORDER BY month), 0), 1) AS mom_growth_pct
FROM monthly
ORDER BY month;

Q7: Bring customer city from the 'Customers' sheet
=XLOOKUP(A2, Customers!A:A, Customers!C:C, "Not found")

Older Excel versions (INDEX-MATCH):
=IFERROR(INDEX(Customers!C:C, MATCH(A2, Customers!A:A, 0)), "Not found")

Q8: Pivot table steps
1. Clean data: remove blanks and duplicates, fix types
2. Select the table -> Insert -> PivotTable
3. Rows: Region   Columns: Month   Values: Sum of Amount
4. Group dates by month if needed
5. Add a slicer for Product category
6. Format numbers and add a short insight below

Q9. Sales dropped last month - investigation plan

1. Check the data: is the drop real? (tracking, late data)
2. Size it: how big, which dates did it start?
3. Break down:
   - Region / city
   - Channel (app, web, offline)
   - Customer type (new vs returning)
   - Product category
   - Price and discount changes
4. External factors: season, holidays, competitor moves
5. Find the biggest contributor -> form a hypothesis
6. Recommend action + how to measure recovery

# Q10: Is the new checkout page's conversion rate better?
# pip install statsmodels
from statsmodels.stats.proportion import proportions_ztest

conversions = [530, 584]      # control, variant (example data)
visitors    = [10000, 10000]

stat, p_value = proportions_ztest(conversions, visitors)
print(f"control {conversions[0]/visitors[0]:.2%}, "
      f"variant {conversions[1]/visitors[1]:.2%}, p = {p_value:.3f}")

# Also check: test ran full weeks, users randomly assigned,
# no change in guardrails such as refund rate.

Framework for any analytics question

1. Clarify:  restate the question, ask about goal and data
2. Structure: say your approach in 2-3 steps
3. Analyse:  walk through metrics, SQL or logic
4. Insight:  state what the data says in one sentence
5. Act:      recommend a next step and how to measure it

Example opener:
"To make sure I understand - we want to know why
repeat orders fell, using order and customer data?"

Prepare 3 stories using STAR:

Situation: ______________________
Task:      ______________________
Action:    (your analysis, tools, steps) ______
Result:    (what changed - be honest, no invented numbers)
           ______________________

Story 1: an analysis project
Story 2: a mistake you caught in data
Story 3: explaining results to a non-technical person

Question | Can answer in 2 min? | Example ready? | Practised aloud?
---------|----------------------|----------------|-----------------
Q1       |                      |                |
Q2       |                      |                |
Q3       |                      |                |
...      |                      |                |
Q10      |                      |                |

Goal: all 10 ticked before the interview day.