Source: https://restorebunny.com/guides/ocr-proofreading-names-numbers/

# Proofread OCR by Checking Names, Numbers and Page Coverage

Compare recognized text with the scan, concentrating on information where one character changes the meaning. Searchable text is useful only when its important errors are found and corrected.

Document capture2 min readUpdated 19 October 2026

## Put the checklist to work

Use the related browser utility on a working copy, then inspect the result. No signup is required.

[Mark areas to review](<https://restorebunny.com/tools/annotate-image/>)Free · runs in your browser · no signup

Quick answer

## The short version

Compare recognized text with the scan, concentrating on information where one character changes the meaning. Searchable text is useful only when its important errors are found and corrected.

## A capture or editing plan

Keep the image reference and open the OCR output beside it. Check headings, names, dates, amounts and identifiers before reviewing ordinary prose. Look for dropped lines, merged columns and characters confused by similar shapes. Record uncertain readings instead of silently guessing. Test a search for a known phrase, but do not treat one successful search as proof that every page was recognized.

## When to change the approach

OCR is not a certification of a document's contents or authenticity. Use the original source when an exact record is required and review the receiving workflow's requirements.

## Worked planning example

Planning example: OCR reads an identifier ending in O as a zero. Comparing the exact line with the scan catches the error even though a general search of the document appears to work.

## If the first attempt fails

If a name or amount looks plausible but differs from the source, correct the transcription and preserve the scan as the reference. Review similar character pairs such as a letter O and zero in context. Check columns and line order separately from spelling. For consequential records, have an appropriate second reviewer compare the relevant fields; a fluent OCR result is not independent evidence that the text is accurate.

## Before you finish

1. Check high-impact names and numbers first.
2. Inspect columns and missing lines.
3. Mark uncertainty instead of inventing text.

## FAQs

What should I check first for document scanning ocr?

Check high-impact names and numbers first. Inspect columns and missing lines. Mark uncertainty instead of inventing text.

## Sources

Sources for technical facts and documented capabilities. Worked scenarios are hypothetical planning examples.

- U.S. National Archives: [Digitizing family papers and photographs](<https://www.archives.gov/preservation/family-archives/digitizing>)

## Related

- [Mark areas to review](<https://restorebunny.com/tools/annotate-image/>)
- [Scan a Document with iPhone Notes and Check the Saved Pages](<https://restorebunny.com/guides/scan-documents-iphone-notes-workflow/>)
- [Correct a Document Photo's Perspective Without Cutting Off Content](<https://restorebunny.com/guides/document-photo-perspective-margins/>)
- [Browse all photo guides](<https://restorebunny.com/guides/>)

## Ready to check your working copy?

Use the related browser utility on a working copy, then inspect the result. No signup is required.

[Mark areas to review](<https://restorebunny.com/tools/annotate-image/>)

RestoreBunny

Photo workflow notes and editorial examples

Guides follow our [editorial policy](<https://restorebunny.com/editorial-policy/>) for sourcing, product claims, AI limitations, updates, and corrections.