Retail and e-commerce companies generate huge amounts of data from orders, visits, deliveries and returns, and they make daily decisions on pricing, stock and marketing. That makes it one of the most common domains for analyst jobs.
Core metrics
| Metric | Meaning |
|---|---|
| GMV | Gross merchandise value: total value of orders placed |
| Net revenue | Revenue after cancellations, returns and discounts |
| Orders | Number of orders placed |
| AOV | Average order value = revenue รท orders |
| Conversion rate | Share of visitors or sessions that place an order |
| Repeat rate | Share of customers who order again within a period |
| CAC | Customer acquisition cost |
| Return rate | Returned orders รท delivered orders |
| RTO | Return to origin: orders that could not be delivered and went back, common with cash on delivery |
| Basket size | Items per order |
Funnel analysis
The purchase journey has steps: visit โ product view โ add to cart โ checkout โ payment โ order. Measuring the share that moves from each step to the next shows where customers drop off.
SELECT
COUNT(DISTINCT CASE WHEN event = 'view' THEN session_id END) AS viewed,
COUNT(DISTINCT CASE WHEN event = 'cart' THEN session_id END) AS carted,
COUNT(DISTINCT CASE WHEN event = 'checkout' THEN session_id END) AS checked_out,
COUNT(DISTINCT CASE WHEN event = 'order' THEN session_id END) AS ordered
FROM events
WHERE event_date >= '2026-10-01' AND event_date < '2026-11-01';A sudden drop at payment suggests a gateway problem; a drop at cart may be about delivery fees shown late.
Cohort retention
Group customers by the month of their first order and track what share orders again in later months. Cohorts show whether retention is improving for newer customers, which overall repeat rates can hide.
Returns and RTO analysis
Returns cost money twice: reverse shipping and lost sales. Typical questions:
- Which categories, sizes or sellers have the highest return rates?
- Are return reasons ("size issue", "not as described") concentrated in certain products?
- Which PIN codes have high RTO, especially for cash-on-delivery orders?
Typical analysis questions
- Why did repeat orders fall for customers acquired during a sale?
- Which discounts bring profitable customers, and which only bring one-time buyers?
- Which products often run out of stock, and how much demand is lost?
- Which customer segments should get which offers?
Techniques
SQL for funnels and cohorts, pandas for cohort tables, Power BI dashboards for daily tracking, and clustering such as K-Means for customer segments.
Practise with real-style data
Our retail and e-commerce projects include a quick commerce retention investigation and a fashion returns and RTO cost analysis, each with a business brief and messy data.
Practise what you just read
Reading helps; writing the code yourself is what sticks. Start here: