Serial list difference check
Compare the generated or assigned serials (A) with the reported or read serials (B). The tool finds those only in A, those only in B, the common ones and the duplicates in either list; it can optionally ignore differences in whitespace, case and leading zeros.
One serial number per line.
One serial number per line.
The comparison runs in your browser; your lists are neither sent anywhere nor stored.
- Only in A
- 4
- Only in B
- 3
- Common
- 3
- Duplicates in A
- 1
- Duplicates in B
- 1
- A lines
- 8
- B lines
- 7
- A unique
- 7
- B unique
- 6
Line counts: A 8 = 7 unique + 1 extra duplicates; B 7 = 6 unique + 1 extra duplicates.
Only in A (4)
Generated but not reported or read.
| Serial | A count |
|---|---|
| SN000001 | 1 |
| SN000004 | 1 |
| sn000006 | 1 |
| SN000007 | 1 |
Only in B (3)
Reported but not in A: unassigned, misread or serials from another batch.
| Serial | B count |
|---|---|
| SN000006 | 1 |
| SN000008 | 1 |
| SN000009 | 1 |
Duplicates in A (1)
Serials that appear more than once in list A.
| Serial | A count | B count |
|---|---|---|
| SN000005 | 2 | 1 |
Duplicates in B (1)
Serials that appear more than once in list B; the same item may have been read or reported more than once.
| Serial | A count | B count |
|---|---|---|
| SN000003 | 1 | 2 |
Common (3)
Common (3)
Serials present in both lists.
| Serial | A count | B count |
|---|---|---|
| SN000002 | 1 | 1 |
| SN000003 | 1 | 2 |
| SN000005 | 2 | 1 |
only A = A \ B · only B = B \ A · common = A ∩ B (unique keys, after normalisation) · duplicate = count > 1 within a list
The tool only compares text. Matching depends on the normalisation options: ignoring leading zeros treats 007 and 7 as equal, ignoring case treats SN1 and sn1 as equal; make these choices according to whether your serial numbers really are equivalent that way. The tool cannot tell the cause of a difference (read error, reprint, other batch); verify the results against the source records. This calculation is not legal or technical advice.
Let's review your serial flow so that the production, reading and reporting lists agree with each other.
Request a conversation01
How to use it
A
Paste the generated or assigned serials into list A and the reported or read serials into list B (or load them from a file).
B
Set the matching options for whitespace, case and leading zeros.
C
Review the serials only in A, only in B, common and duplicated; download the result as CSV.
02
Which differences does it look for?
In a serial number flow three lists are common: the serials assigned to the printer or label, the serials the camera read on the line, and the serials reported to a higher-level system (ERP, a government or a customer system). Comparing two of them exposes missing, extra and double records.
Serials only in A were generated but not reported or read. Serials only in B were reported but are not in the assigned list; this can be a misread, a label from another batch or a number typed by hand. Duplicated serials can show that the same item was read or reported twice.
03
Why normalisation matters
When lists come from different sources the same serial can be written differently: trailing spaces, a different case, a spreadsheet program dropping leading zeros. The right options separate real differences from false ones.
On the other hand, matching that is too lenient hides real differences. If your serial numbers are case-sensitive or leading zeros carry meaning, do not switch these options on. By default only whitespace at the start and end of a line is ignored.
04
Large lists
The comparison works by looking up a set: every serial is read once and looked up once, so lists of hundreds of thousands of lines finish in under a second or a few seconds. The lists are processed in the browser and are not sent anywhere.
For very large files use the "Load from file" button. The result table shows the first 100 rows; you can copy everything or download it as CSV.
FAQ
- Which list should be A and which B?
- The difference is found symmetrically; which one is A changes only the labels, not the result. Usually A is the expected list (generated, assigned) and B the actual one (read, reported). The "Swap A and B" button reverses them.
- How are duplicates counted?
- After normalisation, lines with the same key count as one unique serial and the number of lines is shown as its count. Duplicates are also listed separately as the serials with a count above 1 in each list; the number of extra lines is shown in the summary.
- My lists have other columns besides the serial. What should I do?
- Switch on "If a line has ; , or a tab, take only the first column". The serial number must be in the first column. If it is in another column, first rearrange the file in a spreadsheet program.
- Do hundreds of thousands of lines slow the browser down?
- The comparison runs in linear time; two lists of 500,000 lines usually finish in about a second. Typing into very large text boxes can be slow; in that case load from a file.
- How are empty lines and Windows line endings handled?
- Empty lines are skipped and their number is reported. Windows (CRLF) and Unix (LF) line endings are read the same way.
Let's make serial lists consistent on the line and in the ERP
We build the agreement of assignment, reading and reporting lists end to end; we find differences at their source and close them.