NurViewer · Table recognition
Tables to editable Excel
A screenshot of a table is easy to send and hard to calculate with. Copying text often loses columns and empty cells, leaving you to rebuild the relationships before you can do any useful analysis.
THE PRACTICAL VALUE
Why this step deserves its own workflow
Recover supported table structure from native PDF content or scanned pages, compare the source with recognised cells and export XLSX. Rows, columns and supported merged regions matter because they give each value its context. Review first, then continue calculations or cleanup in Excel instead of re-entering the table from scratch.
A PRACTICAL SEQUENCE
Start with the material you actually need.
Check representative material before deciding how much more to process.
- Select the source page or table region.
- Review row/column boundaries, merges and cell text.
- Export XLSX and verify critical numbers before analysis.
EDITION & SCOPE
Know the scope before you commit.
Recognition may misread small numbers or unusual layouts; formulas are not reconstructed from an image. Free 2 source pages/day with watermark, Standard 20, Pro 120.
Free has no trial expiry: common-format reading, ordinary PDF annotations and supported ordinary document/image-to-PDF export remain available. Recognition, specialist formats and advanced tools have their own allowances or tier requirements.
Compare Free, Standard and Pro →Images may retain historical layouts. Demonstration books, dictionaries and media are not bundled.
A USEFUL QUESTION
Why is structure recognition more useful than plain OCR here?
A value is useful only when its row, column and header stay understandable. Structured export preserves supported relationships for further editing.
CONTINUE THE WORKFLOW
A useful next step.
Keep reading, understanding and organising around the same material.
Table recognition
Borderless table recognition
Many comparison sheets and reports use whitespace instead of grid lines. A generic extraction can turn aligned columns into prose, even though you need a table for sorting and comparison.
Explore the workflow →Table recognition
Merged headers & empty cells
If a multi-level header is flattened or an empty cell disappears, the exported numbers can move under the wrong label. The result may look tidy while saying something different from the original.
Explore the workflow →Table recognition
Tables across PDF pages
A long report splits one table across several pages and repeats its header. Manually joining each fragment in a spreadsheet can introduce duplicate headings or put rows in the wrong order.
Explore the workflow →