failure
Why does copying a table out of a PDF viewer mangle it?
Because a PDF stores text as positioned lines on a page, not as columns and rows — the table you see is drawn, not structural — so the clipboard hands Excel a stack of lines that merely resembles a table, and none of the damage raises an error.
How the paste actually fails
Three failure modes account for almost every mangled copy. Column merging: the whitespace between two cells collapses to a single space, and Excel reads both as one cell — silently, so every total downstream shifts. Line-based selection: viewers select visual lines, so a cell whose text wraps onto two lines arrives as two rows. And because the result looks plausible, the mangling tends to be discovered by whoever reconciles the sheet later rather than by whoever pasted it.
One status note before the repair steps: you can open these pages and sign in to the workspace today, but that does not mean external platform accounts are connected, and it does not mean the product has formally launched. Everything below runs on files you supply.
What you need before you rescue a paste
A plain-text editor open beside Excel, and the source PDF. The text editor shows what the clipboard actually delivered; Excel immediately hides the evidence by guessing delimiters and reflowing cells.
Roughly how many rows and columns the source table should have. Without that count you cannot tell a clean paste from a quiet disaster.
Step 1: Paste into a plain-text editor first
Look at the raw lines before Excel touches them. Consistent column alignment in monospace means the gaps survived; jagged lines mean cells already merged on the way out. Thirty seconds here tells you whether a rescue is worth attempting at all.
Step 2: Count fields per line
Every line should carry the same number of fields. Ragged counts expose exactly which rows lost a column boundary or gained a wrapped fragment, and they mark the rows to repair instead of leaving you to eyeball the whole table.
Step 3: Rejoin wrapped rows by rule
A continuation line typically lacks the leading date, code, or identifier that real rows start with, so the join rule writes itself: a line with no leading identifier belongs to the row above. Apply it mechanically and spot-check three joins against the source rather than trusting the pattern blindly.
Step 4: Split merged columns deliberately
Where two cells arrived as one, use fixed-width import or text-to-columns with an explicit delimiter, verified against a sample of rows first. Splitting on whatever looks convenient reproduces the original mangling with extra steps.
Step 5: Know when to stop repairing
If fixes run past a few minutes per table, the clipboard route has cost more than structured extraction would have. Repairs also do not scale: next month's file mangles the same way, and hand-fixing it forever is not a plan.
Verification
The rescued table holds when its row and column counts match the source, a handful of cells read identically to the page, and the totals reconcile with figures printed in the PDF. Anything less and the paste is still winning.
Limits worth stating plainly
Tables are not cell-perfect. Check every number against the source page before you rely on it. A whole multi-page report converted in one pass is not reliable today. Rows from different periods can land in one cell and a column can be dropped. Work one table at a time. Audit-style statements with dollar signs, em-dash blanks, and parenthesised negatives are not handled. Data rows can be lost. Scanned PDFs with no text layer are refused rather than guessed at. There is no OCR path you can rely on here today. A document with no printed table is not something to convert. If you are handed a sheet built out of running prose, discard it and tell us.
Tablefold skips the clipboard entirely: it reads the detected grid structure and flags the cells it was unsure about, but uncertain cells stay flagged for your judgment rather than being resolved automatically, and the output is a starting point for checking, not a certified copy of the page.
What Tablefold does instead of the clipboard
Tablefold extracts the table from the document's detected grid rather than from visual lines, so wrapped cells stay whole and adjacent columns keep their boundary. Where the grid is ambiguous, the cell is flagged instead of guessed.
The paste route remains fine for genuinely small one-off tables. When the same report needs converting every month, replacing the clipboard with a checked conversion removes the failure mode at the source instead of repairing it after every copy.
FAQ
Questions this guide is for
Why did two columns become one when I pasted?
The gap between them collapsed to a single space during selection, so Excel saw one field. Paste into a plain-text editor to see the raw spacing, then split with a fixed-width import after verifying the column positions on a few rows.
Why does one record span two rows in my sheet?
Its cell text wrapped onto two lines in the PDF, and the viewer selected each visual line separately. Continuation lines lack the leading identifier real rows carry, which makes them safe to rejoin by rule.
Can I just fix the pasted table by hand every month?
You can, and for one small table it may be honest work. For recurring reports the same mangling returns with every file, so a structured conversion that flags uncertain cells usually costs less than a lifetime of repairs.
Start in the workspace
Skip the clipboard, keep the check
Sign in or create an account and you return to the Tablefold conversation. Upload the report and read the extracted grid, flags included, beside the source page instead of rescuing another broken paste.