Excel to PDF Hidden Rows & Formula Fixes

Excel-to-PDF has three signature betrayals — a blank page nobody created, rows you swore you hid, and formulas that printed as something else. They have different causes, and each fix takes under a minute if you know which one you're looking at.

We ran a sheet with a deliberately hidden row through the BytesPDF converter and read the text back out of the finished PDF. Test file: 3 rows × 3 columns, with one row marked hidden="1" and one formula cell (=B3*2, cached value 6). Output: 1,254 bytes, 1 page, 2 rows drawn — and the hidden row's text is gone from the PDF's own content stream. Two findings: - Hidden rows are not printed, and the converter says so. Only the two visible rows were drawn; the run reported hiddenRowsSkipped: 1, and the result summary reads "1 hidden row(s) not printed, matching Excel". That matches how Excel's own print and export treat a hidden row. - Formulas print as their cached value, never as the formula. The output contains 6 and does not contain B3*2. Whatever Excel last stored is what you get, which is why recalculating before export is the real fix for "wrong" numbers. Before-and-after on the same file: honouring the hidden row produced 1,254 bytes; the earlier build, which drew it, produced 1,385. Both runs reported 0 skipped charts, 0 dropped symbols and 0 missing formula results. Columns and sheets follow the same rule. A 4-column sheet with two columns hidden converted to 1,180 bytes and 2 columns, against 1,333 bytes and 4 columns with nothing hidden — the surviving columns close ranks rather than leaving a blank lane. Hidden and veryHidden sheets are skipped entirely, and every run reports what it left out ("2 hidden column(s) not printed, matching Excel"). Hiding is still not redaction. Skipping a hidden row, column or sheet keeps its values out of the PDF, not out of the workbook you converted. If the data itself is sensitive, delete it before you upload.

Trap 1: the ghost blank page

Symptom: a perfectly empty page, usually mid-document or at the end.

Cause: an orphan page break anchored inside or beside a hidden/filtered region — the rows are invisible, the break still exists, and paper doesn't care about filters.

Fix: View → Page Break Preview. Find the stray blue line (often sitting in the hidden zone), right-click → Remove Page Break (or drag it somewhere sane). Re-export.

Trap 2: rows you filtered away reappearing

Symptom: the filter hid 300 rows in Excel; the PDF contains them — or contains their page break consequences.

Where BytesPDF lands, measured: hidden rows are honoured. A row marked with the hidden attribute is left out of the PDF, matching Excel, and the conversion reports how many were skipped (see the evidence box above). Keep the distinction straight: that keeps the row out of the printout, not out of your workbook — it is not redaction.

Causes & fixes:

Export path didn't honor the filter — check Print Preview first; it is the PDF's honest mirror.
Breaks placed while rows were visible survive their hiding — clean breaks as in Trap 1.
"Hidden by filter" vs "manually hidden" behave differently across export routes — for deterministic output, don't rely on filter state: set a Print Area over exactly what you want (select range → Page Layout → Print Area → Set Print Area).

Hidden rows and orphan page breaks producing a blank PDF page

Trap 3: formulas "wrong" in the PDF

The PDF renders cached values — the last numbers Excel stored. Not the live formula bar, not your mental math:

Recalculate before export: Ctrl/Cmd + Alt + F9 (full rebuild) — stale caches from scripted or imported workbooks are common.
Fix visible errors first: a #REF!, #DIV/0!, or #NAME? prints exactly as shown. Hide-with-formatting doesn't help — the text is still there.
Understand values-only rendering: third-party/browser converters read those cached cells without recalculating (the BytesPDF converter is explicit about this — what it does and doesn't do). We confirmed it on a real file: a =B3*2 cell came out as 6, and the text B3*2 was nowhere in the PDF. The belt-and-braces sequence: recalc → fix errors → Print Preview → export from Excel, or convert the verified file and re-check page one.

The export ritual (30 seconds, catches all three)

Full recalc (Ctrl+Alt+F9).
Clear visible errors.
Confirm/refresh the Print Area (or accept filtered view deliberately).
Page Break Preview — no stray breaks.
Print Preview — what you see is what exports.
Export, then open the PDF and check: first page, a formerly-hidden region, one formula-heavy cell.

That's the whole checklist — and it slots into the bigger one: what to check before sending.

Frequently asked questions

Why does a blank page appear in my Excel PDF?

Almost always an orphan page break sitting inside or next to a hidden/filtered region: the break still forces a new sheet of paper even though the rows there are invisible. Open Page Break Preview and remove or drag the stray break; the ghost page disappears.

I filtered rows out — why do they show in the PDF?

Depends on the path: 'Print Active Sheets' honors filter-hidden rows in most builds, but exports via fixed-format with certain settings (or printing 'entire sheet' equivalents) can resurrect them — especially when manual page breaks were placed while the rows were visible. Verify with Print Preview before export, and re-check the filtered region after hiding.

Why do my formulas show as values (or errors) in the PDF?

PDFs render cached results — the last values Excel calculated and stored in the file. Stale cache (file from a macro/script that didn't recalc), errors like #REF! baked in, or a converter reading cached values without recalculating all produce what you see. Press F9 (recalc) in Excel and fix visible errors *before* exporting.

How do I export only the rows I want?

Set a print area over exactly the visible selection: select the cells → Page Layout → Print Area → Set Print Area. For one-offs, Print Selection does the same. This bypasses both hidden-row surprises and stray breaks outside your range.

Does the BytesPDF Excel-to-PDF tool show hidden rows?

No — that one is measured. On a test sheet with a row marked `hidden="1"`, the PDF contained only the two visible rows and the conversion reported one skipped hidden row, so it behaves like Excel's own export. Formulas print as their cached values (a `=B3*2` cell printed as `6`, and the formula text never appeared), and wide sheets scale or split ([fit-to-width guide](/blog/convert-excel-to-pdf-without-cut-off-columns)). One caveat: a hidden row is left out of the PDF, not removed from your workbook — if the data is sensitive, delete it before converting.