List ReconcileCompare two lists and keep original rows in the result
List Reconcile ·
To compare two lists and keep original rows, load the complete tables and select one comparison column from each. A missing ID is much easier to investigate when its location, description, and source line are still attached. Copying only the ID column into a comparison tool can leave you with another lookup task as soon as the results arrive.
An Excel user comparing lists with multiple columns wanted fruit names and their associated cities returned together. The suggested formula compared fruit-city combinations, which is a broader task than this tool supports. This guide covers keeping related fields when one selected key is enough to decide which records to review.
Work backward from the person receiving the result
Imagine you need to send a warehouse supervisor the stock codes that do not appear in a catalog. A list of codes identifies the issue, but the supervisor may need the bin location to check the shelf. A product description helps them recognize the item. The original record gives them that context without another export or manual lookup.
Decide which fields the recipient needs before loading the data. Keeping context does not mean forwarding every column indiscriminately. If the source contains private notes or personal data unrelated to the task, create a suitable export with the required fields. List Reconcile retains the columns you give it in the source views and richer downloads.
A source row is more than its comparison key
Consider this fictional A file:
sku,bin,description
00123,Rack-1,"Bracket, small"
00123,Rack-8,Bracket spare
00456,Rack-2,CableB contains:
sku,status
00123,ListedChoose sku on both sides in Compare two lists. The summary contains two distinct values. 00123 has counts two and one. 00456 has counts one and zero. Selecting Only A returns 00456; opening its source rows shows Rack-2 and Cable, from line 4 of A.
Selecting Different counts includes both values, because 00123 also has unequal counts. Opening that value shows both A records and the single B record. The tool does not select Rack-1 as the “correct” match and silently discard Rack-8. A person can inspect all three records before deciding what the repeated key means.
Distinguish a grouped report from a joined table
A grouped comparison has one summary row per distinct key and keeps references to every source occurrence. A join can produce a combined row for each pairing between two tables. With repeated keys, those pairings require care: two A records and three B records may yield several combinations, depending on the operation.
List Reconcile gives you a grouped report. It does not promise a one-to-one match between repeated records, and it does not attach an arbitrary first B row to each A row. This makes the counts visible, but it also leaves a boundary: the report cannot decide which warehouse entry corresponds to a particular catalog event.
Use a stable record-level identifier if the next step requires individual pairing. If you only have a shared product code, treat the attached rows as evidence to examine. The duplicate-count guide shows why equal membership can still hide a frequency difference.
Understand the source line numbers
Source line numbers refer to physical lines in the original input, including the header and blank lines. The first data row in an ordinary CSV with a header is usually line 2. If a quoted description spans two lines, one record can have a line range instead of a single number.
This matters when someone opens the file in a text editor. A displayed spreadsheet row and a physical text-file line are not always the same location. Sorting a spreadsheet or deleting blank rows creates another copy whose row numbers can differ again. Preserve the file used for comparison if another person needs to follow the references.
For a large input, the main text area shows a short preview while the comparison uses the entire input. The source dialog and download use the parsed records from that full input. Do not interpret the visible preview length as the number of rows processed; check the comparison status and result counts.
Choose a download for the handoff
The Excel report contains a Summary sheet plus Source A, Source B, and Rules sheets. Source values are written as text, which helps retain identifiers such as 00123. Apply the desired filter before downloading; the source worksheets contain the records associated with the selected results.
The JSON report nests the relevant source records under each comparison key, with counts and original line references. That layout is useful when a reviewer wants to inspect a single discrepancy together with its evidence. It is also suitable for a script that reads strings and record arrays without guessing spreadsheet types.
The CSV report contains the grouped summary and line references. It is not a full-row export of both source tables. Likewise, Copy values copies only the comparison keys. Choose Excel or JSON when keeping the surrounding columns is part of the task, rather than assuming every output format contains the same information.
Check the filter before you send the file
The download uses all results in the current filter and search, including pages beyond the visible fifty rows. If the search box contains 001, the report will exclude otherwise relevant keys that do not contain that text. Clear search for a complete “Only A” report, or describe the narrower selection in your handoff.
Open the downloaded file and check one known result against the source. In the example, find 00456, confirm its A count is one, and check that the associated bin is Rack-2. That small verification catches a wrong filter or wrong column before someone starts reviewing hundreds of records.
Send the report with the source filenames and the question being investigated. “These keys occur in the warehouse export but not the catalog export” is a reproducible observation. “Delete these items” is a separate business decision that requires context beyond membership in two files.
For another practice file, load the duplicate-count example. Its Repeated IDs filter includes both a 4/2 difference and an equal 2/2 repeat, with every original occurrence available through View rows.