The short answer: import a large comic collection from a CSV, review uncertain identities before saving them, and keep valid unmatched issues instead of deleting them. A catalog match can enrich a record, but a missing match should not decide whether the comic you already own is allowed into your collection.
ComicPilot's current CSV importer requires only Series and Issue columns. It can also use Publisher, Series Year, Volume, Variant, Cover, Barcode or UPC, Printing, Read Status, Rating, Quantity, Purchase Price, Condition, and Notes. More identity detail improves duplicate and run checks, but missing optional fields should not force a collector to rebuild a large collection by hand.
The current mobile-first importer accepts up to 15 MB, 5,000 non-empty data rows, 256 columns in a logical row, and 250,000 parsed cells per CSV. A file can be under 15 MB and still exceed one of the shape limits. Remove unused columns or split a larger export into row batches before importing it. These are responsiveness and memory-safety boundaries, not limits on the total number of comics you may keep across completed imports.
Each later batch is compared with collection records saved from earlier batches. A very large group of rows claiming the same series and issue can still be blocked for correction rather than checked against partial duplicate evidence. Consolidate intentional copies into a Quantity field, correct incomplete identity fields, or review that group manually instead of relying on a truncated comparison.
Keep identity fields separate
A spreadsheet row is a collector record, not proof of a catalog identity. Series and issue number establish the minimum local record. Series year, volume, publisher, cover designation, variant, printing, and a full barcode can help distinguish runs or editions when those values are actually present.
Do not combine every visible detail into one title cell. A row such as “Example Comics Vol. 2 #7 Cover B” is harder to compare safely than separate Series, Volume, Issue, and Cover columns. Separate fields let the importer normalize routine differences while keeping each accepted value in the identity field where it can be compared safely.
The most valuable optional fields for identity are usually:
- Series Year when a title has relaunched;
- Volume when the export uses it consistently;
- Variant or Cover for edition-conscious collecting;
- Printing when it is known from the comic; and
- Barcode or UPC when the complete number was recorded accurately.
ComicPilot records a barcode from an ordinary spreadsheet as unverified evidence unless it came from a versioned ComicPilot export with edition-level evidence. Do not treat the barcode alone as proof of an exact edition.
Preview before anything is saved
ComicPilot parses the selected file, recognizes common header aliases, normalizes series titles and issue numbers, and builds a preview. The preview separates rows that are ready, blocked, or waiting for review. It also shows warnings, duplicate flags, and how many records will be preserved locally without a metadata match.
Review rows deserve attention because several catalog candidates may share a title and issue number. Choosing a candidate can add useful metadata, but it also reruns duplicate protection against the confirmed identity. The import should not quietly choose an edition merely because it was the first search result.
The same principle applies while shopping: the duplicate-or-variant guide explains why issue identity and edition identity are related but not interchangeable.
Preserve valid unmatched comics
“No catalog match” does not mean “not a real comic.” A title may be new, obscure, independently published, entered differently, or temporarily unavailable from a configured provider. Provider coverage can also change over time. ComicPilot's data-source safeguards keep provider metadata, local ownership, and collector confirmation as separate kinds of evidence.
When a row still has a usable Series and Issue after parsing, ComicPilot can preserve it as a minimal local comic record. It remains marked as unmatched instead of being dressed up with guessed creators, story relationships, covers, or publication details.
That minimal record is still useful. It can participate in local title-and-issue duplicate checks and conservative numerical-gap checks. Its incomplete edition details limit what the app may claim: it can warn that you appear to own the issue without pretending it has proved the exact cover or printing.
Preservation is especially important for a large established collection. Dropping every unmatched row would create false gaps and weaken duplicate warnings precisely where the collector needs the app to be cautious.
Let duplicate rules run during import
The preview compares rows with the existing collection and with earlier rows in the same file. Exact duplicates and duplicates under the collector's variant setting can remain blocked. An uncertain same-issue relationship can require review instead of being silently accepted or rejected.
The variant preference changes the interpretation:
- When different variants are not treated as unique, another cover of the same sufficiently identified issue can count as already owned.
- When variants are treated as unique, a clearly different cover can remain eligible for import.
- When cover or printing evidence is incomplete, the importer should expose the uncertainty rather than manufacture an exact-edition conclusion.
This check is why it is safer to import in reviewed batches than to append raw rows directly to a collection database. If a file must be split, each later batch is still compared with records saved from earlier batches.
Use the attention view for a large file
Large imports do not require reading every clean row one by one. Start with the preview totals, then focus on rows marked blocked, uncertain, duplicated, or warned. Correct the source CSV when a required field is missing or a value is unsafe. Confirm a candidate only when its run and visible edition details agree with the comic record you intended to import.
Before saving, check four counts:
- Ready rows: records that will be imported now.
- Needs review: records that still require an explicit choice.
- Blocked rows: records that will not be imported in this attempt.
- Preserved local: valid unmatched records that will remain useful without provider enrichment.
After the save, ComicPilot reports how many rows were imported and how many were not. A completed import should not be described as complete merely because the file opened; the saved and excluded counts are the outcome that matters.
Know what an import cannot prove
CSV import is not a grading service, appraisal, market-price guide, or automatic cover-recognition system. It cannot verify that a condition label is accurate, that a purchase price reflects market value, or that a short barcode proves a specific variant. It also cannot recover identity details that were never present and cannot be matched reliably.
Treat imported ratings, notes, prices, and condition as collector-supplied data. Treat matched catalog fields as provider data. Treat unmatched fields as local evidence. Keeping those sources separate makes later corrections safer and keeps recommendation confidence honest.
Finally, inspect the collection after importing a batch. Search for a few known issues, one expected duplicate, and one nearby run. If an unmatched comic sits between two reliably identified issues, the meaningful-gap guide explains why incomplete run identity should make the app more conservative, not more confident.
The goal is not a perfect metadata match rate. The goal is a collection record trustworthy enough to answer the buying question in front of you: do I already own this comic, does it connect to what I collect, and is the app honest about what it still does not know?