Before and After: Example of a Repaired MARC Record

MARC repair can sound abstract until you see a real example. This article shows a simplified before-and-after record to illustrate the types of issues MARCReady identifies and fixes automatically, along with what remains for librarian review.

The example is intentionally small and simplified. Real records may contain many more fields, local notes, holdings, subjects, and item data.

Before repair

=LDR  00000nam a2200000 a 4500
=008  240101s        xxu           000 0 eng
=020  \$a9780060723804 paperback
=020  \$a0060723805
=100  1\$aSmith, Jane
=245  00$aThe garden handbook /$cJane Smith.
=260  \$aNew York :$bExample Press,$c2020
=300  \$a250 p. ;$c24 cm
=650  $aGardening
=650  $a
=952  \$pB12345$bMain Library$yBook

Issues in the record

This record contains several problems that MARCReady’s rule engine handles automatically, and some that benefit from librarian review.

After automated repair

The following shows the record after MARCReady’s rule engine has run. Deterministic fixes are applied automatically. Kai suggestions appear in the review interface but are not applied until the librarian accepts them.

=LDR  00000nam a2200000 a 4500
=008  240101s        xxu           000 0 eng
=020  \$a9780060723804$qpaperback
=020  \$a9780060723804$z0060723805
=100  1\$aSmith, Jane
=245  14$aThe garden handbook /$cJane Smith.
=264  
$aNew York :$bExample Press,$c2020
=300  \$a250 p. ;$c24 cm
=336  \$atext$btext$2rdacontent
=337  \$aunmediated$bn$2rdamedia
=338  \$avolume$bnc$2rdacarrier
=650  $aGardening
=952  \$pB12345$bMain Library$yBook

Fields 336, 337, and 338 are generated automatically from the Leader record type. The 300 field and 008 date remain as-is after the autofix pass — they appear in the review interface as Kai suggestions for the librarian to accept or skip.

What changed and why

020 — ISBN qualifier separated

The qualifier “paperback” was moved from the ISBN value in $a into a $q subfield. This keeps the ISBN clean for matching and validation.

Before: =020  \$a9780060723804 paperback
After:  =020  \$a9780060723804$qpaperback

020 — ISBN-10 converted to ISBN-13

The value 0060723805 was identified as a 10-digit ISBN. MARCReady converts ISBN-10 values in $a to ISBN-13 format and moves the original to $z.

Before: =020  \$a0060723805
After:  =020  \$a9780060723804$z0060723805

Note: After conversion, both 020 fields carry $a9780060723804. MARCReady flags this for librarian review. The ISBN-10 0060723805 and ISBN-13 9780060723804 may represent the same edition, or the ISBN-10 may have had an invalid check digit and resolved to the same base number. The librarian should confirm whether both fields are needed or whether one is redundant.

245 indicators corrected

The original record had 245 00 — both indicators set to zero.

• ind1 was wrong: a 100 main entry is present, so ind1 should be 1, not 0.
• ind2 was wrong: the title begins with “The “ (four characters including the space), so ind2 should be 4.

Before: =245  00$aThe garden handbook /$cJane Smith.
After:  =245  14$aThe garden handbook /$cJane Smith.

260 converted to 264

MARCReady’s RDA conversion changes tag 260 to 264 with ind2=1 (production/publication). This is part of the standard RDA treatment for legacy publication fields.

Before: =260  \$aNew York :$bExample Press,$c2020
After:  =264  
$aNew York :$bExample Press,$c2020

336/337/338 generated

Content type, media type, and carrier type fields are generated from the Leader record type (LDR/06 = a, language material). These RDA fields were absent from the original record.

=336  \$atext$btext$2rdacontent
=337  \$aunmediated$bn$2rdamedia
=338  \$avolume$bnc$2rdacarrier

Empty 650 removed

The second =650 $a had an empty $a subfield. Empty subfields are removed automatically.

What Kai may additionally suggest

After the rule-based fixes, Kai (MARCReady’s AI layer) may present additional suggestions in the review interface. These are not applied automatically — the librarian accepts or skips each one.

For this record, Kai may suggest:

• updating 300 $a250 p. to $a250 pages (RDA physical description);
• populating the blank date positions in 008 from 264 $c2020;
• adding a trailing period to 650 $aGardening.

These suggestions appear with a confidence level. Low-confidence suggestions should be reviewed carefully.

What the librarian still decides

Some changes cannot be applied safely by a rule engine alone:

• Are both 020 fields needed? After conversion, both carry the same ISBN-13. The librarian may want to keep both (to preserve the ISBN-10 in $z for reference) or review whether the original ISBN-10 was correct.
• Is the subject heading accurate? 650 $aGardening is structurally valid, but only the cataloguer knows whether it is the best subject heading for this item.
• Are the item field values correct for Koha? The 952 item field uses Main Library and Book. These must match the actual branch codes and item type codes configured in the destination Koha installation.

What if the record is too damaged to repair incrementally?

For records in worse shape than this example — garbled titles, wrong tags throughout, corrupted Leader — MARCReady’s Advanced Repair feature offers a different approach. It uses the record’s ISBN or LCCN to query the Library of Congress catalogue (via SRU) and replaces the record’s fields with the authoritative LoC version. Local holdings fields (852, 942, 9XX) are preserved.

For the example record above, MARCReady would query LoC with bath.isbn="9780060723804" and return the authoritative LC bibliographic record for that ISBN if one exists.

Try your own sample

The best way to evaluate MARCReady is to upload a sample from your own catalogue export. The free preview lets you review up to 3 records per upload at no cost (up to 15 records per month) so you can see whether the suggested repairs make sense for your data.

Related articles

• What MARCReady Can Fix — and What a Librarian Should Still Review
• Fixing ISBN Problems in MARC Field 020 Before Koha Import
• MARC Leader and 008 Explained for Koha Migrations
• How to Validate MARC Records Before Importing into Koha