ISBN Metadata & Covers

Look up ISBNs or search by book title and author to get book metadata, edition details and cover image URLs. Build bookstore listings, fill gaps in a library catalog, or add book covers to a reading app, then export the results as CSV, Excel or JSON.

USD 0.79 per 1,000 delivered books on every Apify plan, plus the small Actor Start event described below. No book-data API key, separate metadata subscription or per-export fee.

Use cases

• Bookstores and inventory teams: match ISBNs to titles, authors, publishers, publication dates and edition details.
• Library catalogs: retrieve source-backed descriptions, languages, subjects, page counts and Open Library IDs.
• Reading apps: find cover image URLs and optionally use a labelled work-level cover when an edition has none.
• Lists without ISBNs: search by title or author and retrieve matching ISBN-bearing editions.
• Catalog imports: download flat ISBN, description, edition and cover columns to join to your own inventory.

This Actor packages Open Library's freely available data into a managed lookup, search and export workflow. You can also access Open Library directly without a metadata license fee. The Actor's value is the ready-to-run workflow, source links, normalized ISBNs, exports and spending-limit handling.

Quick start

1. Paste ISBN-10 or ISBN-13 values into ISBNs to look up, separated by lines, commas or semicolons.
2. Set Maximum unique ISBNs to your batch size. The default example looks up one book.
3. Start the Actor and open its dataset. Use Catalog import for catalog fields or Book covers for cover links.
4. Download Catalog import CSV or Book cover CSV from the Output tab. Check the summary and uncharged diagnostics for unfinished or missing lookups.

To find a book without an ISBN, enter its title, author, or both in Find by title or author. A nonempty search query replaces the ISBN inputs, including the prefilled example.

Single ISBN

{
  "isbnText": "9780441172719",
  "maxItems": 1
}

ISBN batch

{
  "isbns": ["9780441172719", "9780452284234", "9780143127741"],
  "maxItems": 3
}

Equivalent ISBN-10 and ISBN-13 values are looked up only once. When using the array or Single ISBN field in Console, clear the main paste box to avoid adding the example book.

Search by title and author

{
  "searchQuery": "Pride and Prejudice Jane Austen",
  "maxItems": 3
}

Search returns relevant source-selected editions with valid ISBNs, rather than guessing an ISBN from a work-wide list. Results include the search query, source rank and edition ID. Search is relevance-based; review titles, authors and edition details before importing. Editions without ISBNs are skipped and are not billed as result rows. Search supports up to 100 results per run; maxItems is a maximum, not a guarantee that enough matches exist.

Optional work-level cover fallback

{
  "isbns": ["9780441172719", "9780452284234"],
  "maxItems": 2,
  "useWorkCovers": true
}

This option is off by default. If the source edition has no cover, it permits a real cover ID from the same work. Such images may show another edition and are labelled coverSource: "work". Leave it off when edition-specific artwork matters. Missing covers remain missing; no placeholder images or guessed ISBN image URLs are returned.

Input

Field	Use
isbnText	ISBNs separated by lines, commas or semicolons. Spaces, hyphens and ISBN labels are accepted.
maxItems	Maximum unique books to process. Default/prefill 1. Set explicitly for batches. ISBN lookup accepts up to 5,000; search is limited to 100 matches.
searchQuery	Optional title/author search. When nonempty, replaces all ISBN input fields.
isbns	Alternative array of ISBN strings.
isbn	Alternative single ISBN string.
useWorkCovers	Optional same-work cover fallback; default false.

The ISBN input fields are combined and deduplicated in ISBN mode. This is a source lookup tool; it does not import or analyse inventory files. Join the downloaded rows to your catalog using ISBNs.

Existing API integrations keep their original fields. includeCovers and includeDescription default to true; maxConcurrency defaults to 10 and is capped at 10. The legacy includeGoodreads option stays off by default and best effort only. Goodreads ratings are not a standard or guaranteed output. Most customers need no technical settings, source account, proxy or cookies.

Output

The default dataset contains useful book results, with success or partial status. Example abbreviated row:

{
  "input": "9780441172719",
  "normalizedInput": "9780441172719",
  "isValidIsbn": true,
  "isbn13": "9780441172719",
  "isbn10": "0441172717",
  "title": "Dune",
  "authors": ["Frank Herbert"],
  "status": "success",
  "sourcesUsed": ["openLibrary"]
}

Fields	What they mean
isbn13, isbn10	Validated, normalized identifiers. Keep ISBN columns as text in spreadsheets to preserve leading zeroes.
title, subtitle, authors	Source book title and author names.
publisher, publishedDate, publicationYear	Edition publication information. The year comes from the edition date, not the original work's first publication.
editionId, workId, editionName, physicalFormat	Source edition/work IDs and available format details. Useful for checking which version was matched.
description, pageCount, language, categories	Available catalog fields; absent source values are not invented.
coverUrl, coverImages	Largest available image link plus small/medium/large variants where available. Images are not downloaded.
coverSource	book_api: supplied by the book API; edition: edition cover fallback; work: explicitly enabled same-work fallback.
sourceUrl, sourceLinks	Public book record used for the metadata. Existing nested source links remain available.
searchQuery, searchRank, searchSourceUrl	Search provenance, present only for title/author search. Source positions may have gaps after missing-ISBN editions and duplicates are removed.
warnings, errors, status, scrapedAt	Collection outcome and observation time.

The existing confidence field is a metadata-completeness indicator, not a guarantee of bibliographic accuracy. All original fields remain supported.

Included exports

• Book catalog: the original overview with metadata, source links and cover images.
• Catalog import: flat ISBN, description, publication, edition and cover columns. This is data for merging with an inventory, not a promise of a platform-specific store import format.
• Book covers: ISBNs, titles, image URLs and cover provenance.

CSV, Excel and JSON exports use the same delivered dataset, with no separate export event. Cover links do not grant permission to reuse the images; check applicable rights for your use.

Missing books and partial runs

Uncharged lookup diagnostics contain invalid ISBNs, missing records and failed lookups. They are separate from paid dataset rows. A run with no useful books fails with a clear message.

Lookup summary records processed/delivered counts, source warnings, search pagination and spending/time-limit flags. If a limit stops a batch, the summary contains remainingIsbns. Start a new run with those as isbns, set maxItems accordingly and leave searchQuery empty. Completed invalid/missing lookups are in diagnostics. Exclude books already delivered to avoid duplicate charges. Resurrecting a populated run is refused for that reason.

Python API example

For Python 3.11+ with apify-client 3.x:

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("thescrapelab/isbn-book-metadata-enricher-cover-finder").call(
    run_input={
        "isbns": ["9780441172719", "9780452284234", "9780143127741"],
        "maxItems": 3,
    }
)
if run is None:
    raise RuntimeError("The Actor run did not finish successfully")

for book in client.dataset(run.default_dataset_id).iterate_items():
    print(book["isbn13"], book["title"], book.get("coverUrl"))

For search, replace run_input with {"searchQuery": "Pride and Prejudice Jane Austen", "maxItems": 3}. Use the Actor's API tab for endpoint examples and authentication options. The technical Actor name and URL remain unchanged.

Pricing

USD 0.79 per 1,000 delivered book results, on every Apify plan. The same result fee applies to ISBN lookup and title/author search, including available descriptions, edition fields, cover links and exports.

Delivered books	Result fee
1	USD 0.00079
10	USD 0.0079
100	USD 0.079
1,000	USD 0.79

The existing native Actor Start event is USD 0.00005 per GB of allocated memory, minimum one event. For the default 256 MB run, it is charged once. There is no subscription or additional book-data API fee. See the Pricing tab for the active pricing and platform-usage policy.

No result fee is charged for invalid ISBNs, missing metadata, skipped search editions without ISBNs, or duplicates removed within the same run. The start event still applies to a started run. Set a maximum charge in Apify's run options to control result fees; a low limit can return fewer books than maxItems. Batching is usually more efficient than repeatedly running one-book jobs.

Coverage and limitations

• Data comes from Open Library's free first-party book, work, edition and search endpoints. Individual edition JSON is used if a batch lookup is unavailable. No paid metadata provider is required.
• Open Library coverage varies by edition, publication date and language. New, rare or self-published books may be absent; source records can contain errors or conflicting ISBN assignments.
• Search follows source relevance and selects one matching edition per work, then omits entries without valid ISBNs. It is not an exact bibliographic match service or an exhaustive edition catalog.
• Metadata, cover links and descriptions are optional. Source-provided covers can be missing or unavailable; work-level fallbacks may show another edition.
• Publication dates and language names follow source formats. Output order may differ from input order; use ISBNs to join data.
• Representative ISBN batches up to 100 are verified. Larger ISBN batches are accepted by the legacy limit but have no verified throughput guarantee. No 1,000-result performance claim is made.
• Use bounded, on-demand catalog lookups. Open Library is not offered as a high-traffic bulk database backend; respect its access policies and use its data dumps or direct arrangements for large-scale requirements.
• Source changes, temporary errors and rate limits can reduce coverage. The Actor uses bounded retries and does not bypass access restrictions.

Support

Use the Issues tab with the run URL, affected ISBN/query, expected source page and field. Do not post private catalog data or access tokens. If the source has no record, the Actor cannot manufacture one; another edition's ISBN may have different coverage.