Straight answers a compliance officer could check.

Open the Compare page, drop the previously approved version on the left and the revised version on the right, pick an output — Word tracked changes, baked colors, or (for spreadsheet pairs) a colored Excel review copy — and click Compare. The result downloads automatically with a comparison report page and a ✔ Verified badge. The full redline is the filing artifact state regulators expect; excerpts are review aids. See the SERFF use case.

Compare the two .docx (or .docm) editions to get a tracked-changes redline that includes the running headers and footers where the form number and edition date live — the detail generic tools miss and states notice. Assertions check the conventions on every run: the edition date must bump, the form number must not drift, and the header and footer must agree. Failures appear as warnings on the report, never silently.

Drop any file the tool produced on the Verify page. Everything is recomputed from the uploaded bytes — no database, no lookup; the file proves itself. Integrity re-hashes every part against the manifest embedded at generation time; origin checks an HMAC only servers holding the signing secret can produce; and the certificate data shows the embedded change-set, filing metadata, and the file's own SHA-256. Caveat: re-saving the file in Word or Excel rewrites the zip and typically breaks integrity, so consume produced artifacts as produced.

Drop the two .xlsx files. You can get either a paged Word redline — the worksheets' print areas laid out as Word tables mirroring the printed rate pages, with real tracked changes — or a colored Excel review copy of the actual workbook. Rows and columns are re-synchronized by content, so one inserted row shows as exactly one inserted row with nothing cascading below it, and the report opens with the aggregate rate impact.

US insurance is regulated state by state, and state filing checklists routinely require a marked-up/redlined copy showing changes from the previously approved version — the strikeout/underline exhibit this tool produces. Requirements vary by state and filing type, so check the specific checklist; where a redlined copy is expected, this is the artifact it describes.

No. The comparison is deterministic — the same inputs always produce the same redline. That is a selling point for a regulated filing: the result is reproducible and can be verified byte for byte, not a probabilistic guess.

Nowhere, and no. The product is self-hosted and stateless: documents are processed in memory or temp files and deleted after the download, never stored. No document is ever sent to a third party; each request is computed and forgotten. There is no cloud service and no Microsoft Office dependency. See Security & compliance.

Yes. Print areas of tens of thousands of rows compare in seconds. Beyond the readable-Word-table size, the redline automatically shows changed pages only, every omission explicitly marked (— print pages i–j of N unchanged — not shown —); a wholly unchanged sheet contributes no page at all.

Word .docx and .docm, and Excel .xlsx. Both files of a pair must be the same kind — you cannot compare a workbook against a Word document. An Excel pair can produce either a Word redline (from the worksheets' print areas) or the colored Excel review copy.

It trims any output to just what changed, with every omission explicitly marked. For an Excel-to-Word redline that means only the print pages containing changes; for a Word pair, only the changed paragraphs and tables with one context block each side; for the colored Excel copy, unchanged sheets are hidden (never deleted). The report still lists every change with working links, and the full redline remains the filing artifact. Detail in Understanding the outputs.