1 00:00:00,080 --> 00:00:05,071 A duplicate order ID does not mean either row is safe to throw away. 2 00:00:05,071 --> 00:00:09,348 In this practice sheet, SO-105 was Paid and later Refunded, while 3 00:00:09,348 --> 00:00:11,843 SO-104 moved from Draft to Paid. 4 00:00:11,843 --> 00:00:15,765 We will keep each newest update by sorting complete rows first, 5 00:00:15,765 --> 00:00:17,903 then removing duplicates by ID alone. 6 00:00:17,903 --> 00:00:21,824 The check is four surviving orders with the correct current statuses, 7 00:00:21,824 --> 00:00:23,963 not just a smaller row count. 8 00:00:24,563 --> 00:00:29,169 Open the fictional CSV in an Excel for the web workbook copy. 9 00:00:29,169 --> 00:00:33,008 Read both SO-104 rows and both SO-105 rows. 10 00:00:33,008 --> 00:00:37,998 SO-104 changes from Draft on April second to Paid on April eighth. 11 00:00:37,998 --> 00:00:42,988 SO-105 changes from Paid on April third to Refunded on April ninth. 12 00:00:42,988 --> 00:00:46,826 Removing an arbitrary duplicate would retain an obsolete status, even 13 00:00:46,826 --> 00:00:49,129 though the order IDs look tidy. 14 00:00:49,729 --> 00:00:53,697 Keep the original CSV and work only in a copy. 15 00:00:53,697 --> 00:00:58,458 Select the full four-column range including its headers so a sort 16 00:00:58,458 --> 00:01:00,046 moves each complete row. 17 00:01:00,046 --> 00:01:04,807 The tested practice workbook kept Updated as fixed-width ISO text in 18 00:01:04,807 --> 00:01:10,362 YYYY-MM-DD form; that text order matches chronological order for these six values. 19 00:01:10,362 --> 00:01:14,330 Microsoft cautions that other dates stored as text may sort 20 00:01:14,330 --> 00:01:16,314 alphabetically into the wrong chronology. 21 00:01:16,314 --> 00:01:20,678 Inspect the top and bottom dates after sorting, and convert text 22 00:01:20,678 --> 00:01:24,646 to real dates if the format or order is unreliable. 23 00:01:25,246 --> 00:01:29,380 With the full four-column range selected, open Data > Custom Sort. 24 00:01:29,380 --> 00:01:34,266 Sort by Updated, Cell Values, Sort Descending; keep My data has headers selected. 25 00:01:34,266 --> 00:01:38,400 Read the top sequence: April tenth, ninth, eighth, fifth, third, second. 26 00:01:38,400 --> 00:01:42,910 Check that Order ID, Status, and Amount values moved with each date. 27 00:01:42,910 --> 00:01:47,420 The first SO-105 row should now say Refunded, while its older 28 00:01:47,420 --> 00:01:48,923 Paid row is lower. 29 00:01:49,523 --> 00:01:54,087 Select the full sorted four-column range, then use Data > Remove Duplicates. 30 00:01:54,087 --> 00:01:58,271 Leave Order ID selected as the sole comparison column and clear 31 00:01:58,271 --> 00:02:00,173 Updated, Status, and Amount. Confirm. 32 00:02:00,173 --> 00:02:04,737 Microsoft's documentation says the first occurrence of a matching key is 33 00:02:04,737 --> 00:02:08,920 kept and the other complete rows are removed; Excel for the 34 00:02:08,920 --> 00:02:10,442 web supports this control. 35 00:02:10,442 --> 00:02:14,626 Do not choose all four columns, because those six rows are 36 00:02:14,626 --> 00:02:16,527 not identical across all fields. 37 00:02:17,127 --> 00:02:20,633 Read the four surviving orders against the answer check. 38 00:02:20,633 --> 00:02:25,698 SO-105 must be Refunded dated April ninth, and SO-104 must be 39 00:02:25,698 --> 00:02:27,256 Paid dated April eighth. 40 00:02:27,256 --> 00:02:33,100 SO-107 stays Paid for two hundred dollars; SO-106 stays Pending for forty-five. 41 00:02:33,100 --> 00:02:37,385 If either older status survived, Undo in the practice copy, verify 42 00:02:37,385 --> 00:02:40,891 the date sort, then repeat the ID-only cleanup. 43 00:02:41,491 --> 00:02:45,850 If two updates have the exact same timestamp, decide a secondary 44 00:02:45,850 --> 00:02:50,606 tie-breaker before deleting either row; this practice file has no tie. 45 00:02:50,606 --> 00:02:54,965 If new history arrives later, repeat the process on a new 46 00:02:54,965 --> 00:02:57,739 copy or design a separate refreshable workflow. 47 00:02:57,739 --> 00:03:02,495 Do not describe this one-time Remove Duplicates action as live synchronization. 48 00:03:03,095 --> 00:03:05,407 The fictional source has six rows. 49 00:03:05,407 --> 00:03:10,031 After the practice cleanup, the copy should have four: SO-107 Paid, 50 00:03:10,031 --> 00:03:15,427 SO-105 Refunded, SO-104 Paid, and SO-106 Pending, each with the newest 51 00:03:15,427 --> 00:03:16,968 date for its ID. 52 00:03:16,968 --> 00:03:21,208 The two obsolete rows should be removed only from the copy. 53 00:03:21,208 --> 00:03:25,447 If your workbook keeps an older status, undo, inspect the date 54 00:03:25,447 --> 00:03:30,071 type and sort, and use only Order ID as the duplicate key. 55 00:03:30,071 --> 00:03:34,311 The written guide includes the six-row file and exact answer.