How to archive a webpage as evidence

Somebody published a claim, a price, a review, a job listing, a takedown-worthy post — and you need to be able to show what it said. A screenshot is better than nothing and weaker than most people assume: it is trivially editable, it shows one viewport, and it carries no record of where it came from.

Here is a method that holds up better, without a court-grade preservation vendor.

Capture more than one form of the same page

Redundancy is what makes a record credible. For anything that might matter, take all four:

1. A full-page screenshot — immediately legible to anyone, no tooling required.
2. A self-contained HTML capture — the page as rendered, assets embedded, openable offline. This is the one that shows the page behaved a certain way: links, layout, responsive rendering, the lot.
3. A PDF print — flat, portable, universally readable, and dated by the operating system.
4. A third-party archive — submit the URL to the Internet Archive's Wayback Machine ("Save Page Now") or archive.today. An independent timestamp you did not control is the single most persuasive artefact you can have, and it costs one click.

Record the metadata immediately

Alongside the files, in a plain text note:

• the full URL, including query string
• the exact date and time, with timezone
• your timezone and, if relevant, the country you fetched it from (many sites geo-vary)
• the browser and version, and whether you were signed in
• the SHA-256 of each captured file
• who took it

sha256sum example-com-pricing-2026-09-03.html

Hashing costs nothing and lets you show later that the file has not changed since capture.

Keep the capture complete and unedited

• Scroll the whole page before capturing so lazy-loaded content is present.
• Do not crop, annotate or "clean up" the original file. Make annotations in a separate copy.
• Capture the surrounding context — the page's own navigation and footer, the byline, the timestamp on the post.
• Capture what it links to, if the claim depends on it.
• Do not use a signed-in view unless you have to; if you do, note it, and be careful about other people's personal data in the file.

Why a rendered capture beats "view source"

The HTML a server sends is frequently not the page anyone saw. Content gets assembled in the browser, A/B tests vary it, personalisation changes it. A capture of the rendered page is a record of what was actually displayed — and if it is self-contained, it still displays that way when the original is gone. That is why the test we apply to our own captures is rendering them with the origin unreachable: an archive that quietly re-fetches from the live site is not an archive, it is a window onto whatever they publish today.

File naming and storage

domain-path-YYYY-MM-DD.ext, one folder per matter, hashes in a MANIFEST.txt, backed up somewhere you do not control. Zip the folder when you send it.

When you need more than this

If the matter is already litigation, use a preservation service that provides notarised timestamps and a chain of custody, or have counsel direct the capture. This method is for the far more common case: a business needing a solid, defensible record quickly.

FAQ

Is a screenshot admissible as evidence?

Screenshots are routinely accepted but easily challenged, because they are trivial to edit and show only one viewport. Pair them with a self-contained page capture, a PDF, a hash and an independent archive submission.

How do I prove when a page was captured?

Combine your own metadata note and file hashes with an independent timestamp — a Wayback Machine or archive.today submission of the same URL at the same time is the strongest cheap option.

What is the best format to archive a webpage?

A single self-contained HTML file preserves the most: layout, links, responsive behaviour and assets, with no server needed. Keep a PDF alongside it for universal readability.

Should I archive pages behind a login?

Only when necessary, and note that you were authenticated. Be aware the capture may contain other people's personal data, which brings its own obligations.

How long will an HTML archive remain readable?

A self-contained HTML file needs only a browser — no network, no server, no plugin — which makes it one of the more durable formats available. Store the hash with it so you can prove it is unchanged.

Keep reading

The best website copiers in 2026, measured

HTTrack, wget, SingleFile, monolith, SiteSucker and CopyAnySite — what each is for, with fidelity scores measured offline.

How to convert any website into an editable WordPress page

Turn a captured page into Gutenberg section blocks with its real CSS, fonts and images — editable in the block editor instead of pasted into a Custom HTML block.

How to copy a landing page you like (properly)

A repeatable process for taking a landing page you admire and turning it into your own, without shipping their fonts, images or words.

Why fonts and logos break when you copy a website (and how to fix it)

The nine places a logo hides, the font race that silently reflows your copy, and how to check both survived before you rely on the file.