GUIDE · PRESERVING A WEB PAGE
How to preserve a web page as evidence.
A useful record of a web page answers four questions: what the page said, where it came from, when it was captured, and whether anyone has changed the record since. Most quick methods answer the first and skip the rest.
What a good record shows
- What: the content itself, ideally the exact bytes the site sent, not only a picture of them.
- Where: the address you asked for and the address the site actually served, after redirects.
- When: a capture time that does not depend only on your own computer's clock.
- Unchanged: a fingerprint (such as a SHA-256 hash) so anyone can tell whether the saved copy was altered.
The usual options
Screenshot or print to PDF. Free and instant, and fine as a working note. On its own it is weak: it can be edited, and nothing independent fixes the time.
Internet Archive (Save Page Now). Free, and the dated copy sits with a third party. The copy is public, and some sites block the archive.
Perma.cc. Run by the Harvard Law School Library and widely used for preserving pages cited in legal writing.
Capture services and expert collection. For high-stakes matters, firms use dedicated capture services or a forensic examiner who can testify about the method. Ask your lawyer what your matter needs.
Repeated captures. If you need to show how long something stayed online, or when it changed, one capture is not enough. You need a series, ideally linked so that a gap or a swapped copy would show.
Checking a capture yourself
If a capture includes a hash and an RFC 3161 timestamp, you do not have to trust whoever made it. With OpenSSL installed, you can confirm the timestamp covers the saved record:
openssl ts -verify -data manifest.json -in timestamp.tsr -CAfile <root certificate>
and recompute each file's SHA-256 to confirm nothing was changed. Those file names are FluxProof's; other tools name theirs differently.
Limits that apply to any automated checker
A checker sees what the site sends to it. Sites can show different prices by location, device, login or cart contents, so a record shows what that request received, not what every shopper saw. Prices drawn in by JavaScript after the page loads are missed by tools that read only the HTML (FluxProof is one of them; it reads the price a page declares in its HTML or Shopify's public product data). Some large marketplaces block automated requests altogether. This page is general information, not legal advice.
IF YOU NEED A SERIES OF CAPTURES
How FluxProof does this
FluxProof watches a list of public pages on a schedule and saves a proof packet whenever a page changes: the exact page as served, a manifest with the URL, redirects, time, status, headers and hashes, an RFC 3161 timestamp from DigiCert or FreeTSA, FluxProof's Ed25519 signature, and a before-and-after file when something changed. Each packet links to the previous one. No subscription:
- Evidence Starter: up to 25 pages, each checked every 6 hours, $29 for 30 days, paid once.
- Evidence Pro: up to 150 pages, each checked every hour, plus signed webhooks, $99 for 30 days, paid once.
Download a real sample packet and check it with the commands in its README, or see the plans.