Examples · Operations
Orders and refunds dashboard
One order export, the usual quirks, and numbers that still match by hand.
The decisionAre refunds creeping up, and is growth coming from more orders or bigger ones?
Sample data · not a real company
Orders report, SeptemberClick a number
Apr
May
Jun
Jul
Aug
Sep
Paid revenue by month ($k)
Receipt
Orders, September
- Definition
- Orders with a payment
- Source
- orders_q3.xlsx, 412 rows
- Left out
- 1 totals row
- Check
- Data ends Sep 28
The numbers in this example
Each one with its full receipt.
- Paid revenue$12,480
- Definition
- Paid orders, minus refunds
- Source
- orders_q3.xlsx, 1,203 rows
- Left out
- 1 totals row, 4 blank amounts
- Check
- Data ends Sep 28
- Orders412
- Definition
- Orders with a payment
- Source
- orders_q3.xlsx, 412 rows
- Left out
- 1 totals row
- Check
- Data ends Sep 28
- Average order$30.29
- Definition
- Paid revenue ÷ paid orders
- Source
- orders_q3.xlsx, 412 rows
- Check
- Matches revenue and orders
- Refund rate2.4%
- Definition
- Refunded orders ÷ paid orders
- Source
- orders_q3.xlsx, 10 refunds
- Check
- Data ends Sep 28
How it was built
- 1
The decision
Operations checks monthly whether refunds or order size need attention.
- 2
The metrics
Paid revenue, orders, average order, refund rate, and a monthly trend.
- 3
The cleaning
One totals row and four blank amounts were left out. September's data ends on the 28th, and every September number says so.
- 4
The receipts
Every number traces back to the rows of orders_q3.xlsx.