Why Retail Companies Are Hiring More Data Analysts in 2026

-- One view of sales across channels
SELECT channel,
       DATE_TRUNC('week', sale_date) AS week,
       COUNT(DISTINCT bill_id)        AS bills,
       SUM(net_amount)                AS net_sales
FROM (
    SELECT 'store' AS channel, sale_date, bill_id, net_amount FROM store_sales
    UNION ALL
    SELECT 'app'   AS channel, sale_date, bill_id, net_amount FROM app_sales
    UNION ALL
    SELECT 'web'   AS channel, sale_date, bill_id, net_amount FROM web_sales
) s
GROUP BY channel, DATE_TRUNC('week', sale_date)
ORDER BY week, channel;

Power Query steps (Excel)
1. Data > Get Data > From Folder  (one CSV per channel)
2. Combine files, add a "channel" column from file name
3. Change types: sale_date = Date, net_amount = Decimal
4. Close & Load to Data Model
5. Insert PivotTable:
   Rows    = week (grouped from sale_date)
   Columns = channel
   Values  = Sum of net_amount

-- Sell-through rate by product for the season
SELECT p.sku,
       p.category,
       SUM(s.qty)                               AS units_sold,
       MAX(r.qty_received)                      AS units_received,
       ROUND(100.0 * SUM(s.qty) / NULLIF(MAX(r.qty_received), 0), 1) AS sell_through_pct
FROM products p
JOIN receipts r ON r.sku = p.sku
LEFT JOIN sales s ON s.sku = p.sku
GROUP BY p.sku, p.category
ORDER BY sell_through_pct ASC;

DAX measures (Power BI)
Promo Sales =
CALCULATE ( SUM ( Sales[net_amount] ), Sales[is_promo_week] = TRUE () )

Normal Weekly Sales =
DIVIDE (
    CALCULATE ( SUM ( Sales[net_amount] ), Sales[is_promo_week] = FALSE () ),
    CALCULATE ( DISTINCTCOUNT ( 'Date'[Week] ), Sales[is_promo_week] = FALSE () )
)

Average bill value   = net sales / number of bills
Units per bill       = units sold / number of bills
Sell-through %       = units sold / units received x 100
Like-for-like growth = sales growth from stores open in
                       both periods
Markdown %           = discount value / original price value
Repeat rate          = customers with 2+ bills / all customers

Build a practice dataset (synthetic):
stores(store_id, city, opened_on)
products(sku, category, mrp)
sales(bill_id, store_id, sku, qty, net_amount, sale_date)
receipts(sku, store_id, qty_received, received_on)
customers(customer_id, joined_on)

Generate 6 months of rows with Python, then answer:
- Which category has the lowest sell-through?
- Which store has the highest average bill value?

Business question:
Who asked and what decision depends on it:
Data needed (tables, date range):
Metrics:
Comparison (vs last week / last year / control):
Output format (dashboard, 1-page note):
Deadline:

Weekly Sales Note — Week __
Headline (2 lines):
Top 3 movements:
  1.
  2.
  3.
Stock alerts (top sellers below cover):
Promotions running and early read:
Questions for next week:

Project: Omnichannel retail dashboard (synthetic data)
Page 1: Sales by channel and week
Page 2: Category sell-through and stock alerts
Page 3: Promotion uplift and margin
Page 4: Loyalty customers — repeat rate by month
Write-up: 5 insights and 3 recommendations