Microsoft retires InfoPath on 14 July 2026. Archiving is easiest while a working InfoPath install is still around — which is the whole reason to do it now.

"Archiving" InfoPath data means one thing: turning files that only InfoPath could read into copies that anything can read, and keeping them in a form an auditor will accept years from now. Here's how to do that carefully, in an order that never risks your originals.

Step 1 — Inventory everything first

You can't preserve what you can't find. Search your PCs, shares and SharePoint sites for:

  • .xsn — InfoPath form templates (the form design + field catalogue).
  • Form-data .xml — the filled-in records people actually submitted.

Record where each lives, which business process it belongs to, and roughly how many records there are. A simple spreadsheet is fine. This inventory is also the backbone of your audit trail later.

Step 2 — Protect the originals (copy, never move)

Make a read-only copy of everything into a clearly named "originals — do not touch" folder, and work only from copies. The golden rule: never let any conversion or export write back to the source file. If a tool can't promise that, don't point it at your originals.

This is a principle we build into our own tool, not just advice: XSN Rescue opens your file read-only and checks after every run that the original hasn't changed — every report carries a source_modified: false attestation. But the principle stands whatever route you choose: copy first, work from copies.

Step 3 — Choose your export route

There's no single right answer; it depends on how many forms you have and who's doing the work.

If you have a handful of forms (free)

Open them in a still-working InfoPath install and export or copy the data out by hand. Perfectly reasonable for a few dozen records. Its limits are scale and the fact that it relies on InfoPath still running.

If you're technical and patient (free)

Because an .xsn is a CAB archive, you can unpack it (with expand, 7-Zip or similar) to read the schema and views, and open each form-data .xml in a text editor or with a script. Feasible and free; slow and error-prone across hundreds of files, and you still have to assemble the readable archive yourself.

If you have volume, or need it auditable (our tool)

XSN Rescue automates the careful version of the above: it parses the .xsn and .xml locally and exports every filled-in value to CSV, JSON and a readable HTML rendering, with an honest recovery report. Nothing is uploaded; there's no account; your originals stay untouched.

Step 4 — Produce an archive an auditor will accept

A raw CSV is data, but it isn't yet an archive. For records-retention and audit you generally want:

  • A durable, human-readable copy of each record (PDF or HTML) that opens with no special software.
  • An index so any given record can be found quickly.
  • A summary/inventory of the whole set for the person signing off.

You can assemble this by hand. Or, if it helps, XSN Rescue's archive output produces exactly this from the same single pass: an archival PDF per record ({name}-record-NNN.pdf), an archive-manifest.csv plus a browsable archive-index.html linking to every PDF, and an estate-inventory executive report as HTML, JSON and a one-page PDF. All generated locally by the tool's own PDF writer — no cloud service, no per-page fee.

Step 5 — Store it well and record what you did

  • Keep the readable archive alongside the untouched originals — belt and braces.
  • Store a copy off the machine (backup, or a second location) per your retention policy.
  • Note who ran the archival, when, and what the recovery reports said — including anything that couldn't be recovered. Honest gaps recorded now save arguments in an audit later.

Set expectations honestly. Any archival route captures the data and static layout. Form logic (code-behind, external data connections, role rules) is a live-application behaviour, not archivable data — a migration project is the answer if those workflows are still active. Keep archival and migration as two separate, clearly scoped jobs.

The short version

  1. Inventory every .xsn and form-data .xml.
  2. Copy the originals to a read-only "do not touch" set.
  3. Export the data to open formats — by hand, by script, or with a tool.
  4. Turn that into a readable, indexed, summarised archive.
  5. Store it safely and write down what you did.