How do libraries keep track of multiple copies of the same book?

Libraries record the title once and give each physical copy its own identifier - traditionally an accession number, printed inside or on the cover. Loans are recorded against the copy, not the title, so when one copy comes back the library knows exactly whose loan to close and how many copies remain on the shelf.

Can I track multiple copies of a book in a spreadsheet?

Yes, by adding a copy-number column and giving each copy its own row. It works for a small shelf, but the number only exists in the spreadsheet unless you also write it in each book, and the title's details end up duplicated across rows where they drift out of sync. Purpose-built library software keeps one title record with separate copy entries underneath.

What is the number on each library book copy called?

Traditionally it is the accession number - a unique id assigned to each physical copy as it enters the collection, usually printed as a barcode label. Modern small-library tools use the same idea with friendlier forms, such as a short readable code a volunteer can type instead of scan.

Do I need a barcode scanner to tell copies apart?

No. A scanner is one way to read a copy's identifier quickly, and it earns its keep in a library doing hundreds of checkouts a day. In a small volunteer library, a human-readable code on each copy does the same job - read it, type it, and the right copy comes up - with no hardware to buy or maintain.

How to track multiple copies of the same book

The fix is one structural rule - give every physical copy its own identity, and record loans against the copy, not the title. Everything else follows from that.

Dan Edwards

Founder

Thursday, 2 July 2026

Title card reading "How to track multiple copies of the same book" over a row of pale library books

Tracking multiple copies of the same book comes down to one rule: give every physical copy its own identity, and record loans against the copy, not the title. Record the book once, list each copy under it, and lend and return copies individually. A title is a fact about the catalogue; a copy is a thing that leaves the building.

Most small libraries discover this the hard way, at the returns desk.

Why title-level tracking breaks with two copies

With one copy of everything, tracking by title works fine. "The Hobbit is out to Margaret" is unambiguous because there is only one Hobbit to be out.

The second copy breaks it. Two members borrow The Hobbit, one copy comes back, and the record cannot tell you whose loan just closed. Guess wrong and the consequences arrive weeks later:

The overdue list chases the member who returned their copy, while the one still holding theirs hears nothing.
"How many copies are on the shelf right now?" becomes a walk to the shelf instead of a glance at the screen.
If a copy goes missing, there is no way to say which one - the hardback in good condition or the paperback with the cracked spine.

None of this is a discipline problem. The most careful volunteer cannot close the right loan when the record has no way to say which physical book came back.

The spreadsheet workaround and where it breaks

The standard fix in a spreadsheet is a copy-number column: The Hobbit copy 1, The Hobbit copy 2, each with its own row. It genuinely helps, and for a small shelf it can carry you for a while.

It breaks in two places.

First, the number only exists in the spreadsheet. Nothing on the physical book says which copy it is, so at the returns desk you are back to guessing - unless you also write the copy number inside every cover and keep the two in sync by hand, which is the actual system doing all the work.

Second, the title's details now live on several rows. Fix a misspelled author on copy 1 and copy 2 keeps the typo; sort or filter carelessly and the copies drift apart. Every duplicated row is a small integrity debt that compounds as the collection grows. The full picture of where spreadsheets strain is in whether you can run a small library on a spreadsheet.

Some library software has the same gap. TinyCat's copies field does not connect to lending, so to lend duplicates you catalogue each physical copy as a separate entry - the spreadsheet workaround rebuilt inside a catalogue, with each duplicate counting toward your plan's item ceiling. Its LibraryThing catalogue data is genuinely deep, which makes the missing copies layer easy to overlook until the second copy of something popular arrives.

The structure that works - items, copies, loans

The clean model has three layers.

The item is the title: recorded once, with the author, cover and details in one place.
The copies sit under the item: one entry per physical book, each with its own identity and condition.
The loans attach to copies: this copy, this borrower, this due date.

A return closes one copy's loan and touches nothing else. One copy of The Hobbit can be out, one on the shelf, and the record agrees with the shelf without anyone walking over to check. A donated third copy is one new line under the existing item, not a re-catalogued duplicate.

Loan-side practice - recording borrowers, due dates and overdues - is its own subject, covered in how to track who borrowed what in a small library. This piece is about the layer underneath: the copy having an identity at all.

The identity has to be on the book

A per-copy record only ends the returns-desk guessing if the physical book says which copy it is. That means each copy carries its identifier somewhere visible - which is exactly what libraries have done for a century with accession numbers, the traditional per-copy id (what an accession number is covers the concept).

The traditional form is a sequential number, usually printed as a barcode. It works, but a bare number is unreadable to a human: at the desk, "0000000341" means nothing without a scanner, and scanners bring their own failure modes - glossy tape that will not scan, alignment drift, the one volunteer who can coax the device into working.

A big-city library with hundreds of checkouts a day needs the scanner regardless; typed codes do not suit that volume, and a school running a scanner-per-desk workflow should keep it. A volunteer library lending a handful of books a session has a simpler option: an identifier a person can read aloud and type.

How Your Book Nest helps

Your Book Nest is built on the items → copies → loans structure, so multiple copies are the normal case rather than a workaround.

Every copy gets a generated three-word code, like oak-bat-tree. It is the copy's identity and the way you act on it: read it off the cover and type it - hyphens optional, capitals ignored - and the right copy comes up, no scanner needed. Because typing works in any browser, the returns desk can be a phone as easily as a computer - copies check in wherever the box of returns happens to be. Because no two copies share a code, the which-copy question cannot arise at the returns desk. Libib, which has the strongest per-copy model among the paid tools, recommends its own per-copy codes over ISBN scanning for the same reason: an ISBN matches every copy of a title, so a scan ends in a "which copy?" prompt.

Copies brought in from another system keep their old numbers too - a free-text barcode field holds whatever identifier is already stickered on the book, in any format.

The code goes on the book via the Labels page, which prints labels for your copies onto blank Avery sheets - the three-word code in large print, plus a scannable barcode if you ever add a scanner. The page remembers which copies already have a printed label, so a new box of donations means printing labels for just the new arrivals. The step-by-step is in how to print your own library barcode labels.

All of it - per-copy tracking, code generation, label printing - is included in the flat $60 a year plan, free under 100 items, and the home page is a live demo: add a book, give it two copies, and lend them to two different people with no account and no card.