You've got a 30-page report packed with comparison tables, and someone wants it editable in Word. The text bits are easy. The tables are where the converter either earns its keep or makes your afternoon miserable.
Good news: table detection has come a long way. Bad news: a few edge cases still need a human eye.
How table detection actually works
Modern converters look for the structural fingerprints of a table — rows of aligned text, consistent column widths, ruling lines. If the original PDF was exported from Word or Excel, those signals are strong and the table comes across as a real Word table.
If the PDF was scanned, or built by hand-positioning text in InDesign, the table is technically just floating words. The converter has to reconstruct it from geometry alone, which is harder.
Use a converter that rebuilds real tables
Flint's PDF to Word tool reconstructs tables as native Word table objects — cells you can click into, columns you can resize, headers you can repeat. Not text boxes pretending to be tables.
If your PDF is mostly tabular data — invoices, reports, statements — and you actually want a spreadsheet, convert to Excel instead. Word's table tools are decent but Excel is where numbers live.
Where things still go sideways
Merged cells are the classic gotcha. A header that spans three columns confuses geometry-based detection. Fix: select the cells in Word and merge them manually.
Tables without borders are guessed from spacing alone — less reliable. Multi-page tables sometimes get split. Footnotes inside cells drift out. None of these are dealbreakers, but budget five minutes per heavy-table page for cleanup.
Cleaner source = cleaner output
If you have any control over the upstream PDF, ask for the source file. A Word doc exported to PDF, then converted back, will lose less than a PDF built in InDesign. If you must work from the PDF, make sure it isn't a scan — open it and try to select text. If you can, you're fine. If you can't, run it through OCR before converting.
FAQ
Will merged cells survive?
Sometimes. Detection is improving but it's still the most fragile bit. Plan to fix merges manually for anything important — it's a 30-second job in Word.
What if the table has no borders?
Borderless tables are inferred from column spacing. They mostly work, but if columns are very close together the converter may glue them into one. Adding borders to the source PDF beforehand isn't realistic — just spot-check the output.
Should I convert to Excel instead?
If the document is primarily tables of numbers, yes — PDF to Excel handles them with column-aware logic. Use Word when prose surrounds the tables.
Why are some rows joined together?
Usually means the row spacing in the source PDF was tighter than the converter's threshold. Split rows manually via Layout > Split Cells.
Tables are the hardest part of PDF-to-Word, but a good converter gets you 95% of the way. Convert your PDF to Word and we'll handle the heavy lifting.