Archive Website Before Redesign: Save Pages and Old URLs

To archive a website before a redesign properly, make two copies while the old site is still online. The first is a restorable backup (files, database or platform exports) that lets you roll back. The second is a browsable offline copy of every public page that opens in any browser. The offline copy also gives you the list of old URLs you need for 301 redirects. Archive website before redesign launch day, not after it: once the new site replaces the old one, there is nothing left to copy.

Why archive your website before a redesign, not after launch

On launch day the old site usually disappears in one of three ways. The new theme overwrites the old one on the same installation, the domain is pointed at a new host, or a site builder publishes the new design over the old one. From that moment the old pages no longer exist anywhere you control.

The problems that need the old site rarely show up on launch day. They show up weeks later:

• A page with good backlinks now returns a 404 because nobody knew it existed.
• The client asks for the wording of an old service description or pricing table.
• Rankings drop for a URL that was never mapped to a new page.
• Someone wants to know whether a legal notice, disclaimer or price was on the old site, and in what form.

The usual fallback is the Internet Archive. Its captures are irregular, though: some sites are captured often, others rarely, and subpages, images and PDFs are often missing. Restoring from the Wayback Machine is a rescue for when nothing else exists. It is not a plan.

A practical schedule is two copies. Make the first when the project starts, as the reference for your content audit. Make the second right before the switch, after the last edits on the old site, so it shows exactly what visitors saw on the final day.

Restorable backup vs. browsable archive: you need both

Most redesign checklists say "take a backup" and mean a files-plus-database backup made with a WordPress plugin. That backup matters, but it answers a different question than an archive.

A database dump does not show what the homepage looked like. A folder of HTML files cannot bring the old CMS back. A copywriter checking old wording, an SEO checking titles and headings, and a developer rolling back a broken launch need different things, so keep both.

What should you back up before a website redesign?

Use this as the backup part of your website redesign checklist:

• Every public page, including old blog posts, landing pages from past campaigns and pages that are no longer in the menu but still have traffic or links.
• Media and downloads: images, PDFs, brochures, price lists. These often have their own URLs with backlinks.
• The old URL list: every address that currently returns a page, so each one can get a redirect.
• On-page SEO data: page titles, meta descriptions, headings and structured data. In an offline copy these stay inside each HTML file, so you can check them later.
• Existing redirects: rules in .htaccess, the server config, a redirect plugin or the builder's redirect settings. Old redirects have to be carried over too, or you break links that already work.
• robots.txt, sitemap.xml and tracking codes, so you can compare settings after launch.
• Data behind the front end: form submissions, customer and order data, member accounts, newsletter subscribers. None of this is visible on public pages, so it has to be exported from the CMS, shop or builder directly.

How to make an offline copy of the old site

A browser-based downloader is the quickest way to get a complete, clickable copy without setting up command-line tools. With Website Downloader:

1. Enter the old site's address in the download field on the home page.
2. Start the download and watch the live progress as it counts pages. You can close the tab or switch off your computer; the crawl runs on the server.
3. Download the finished ZIP, unzip it and open index.html in a browser.

Before you file the archive away, check it:

• Click through the main navigation and a few deep pages, such as an old blog post or a product detail page.
• Check that images, fonts and layout load. If they don't, see downloaded website missing images and CSS.
• Compare the number of pages in the archive with your sitemap. A large gap usually means pages that nothing links to, or content that only loads through JavaScript.

On size and plans: the free plan covers up to 100 subpages with no account, which is enough for many small business sites. For a larger site, the Single Site plan is a one-time 7.99 EUR for one website with up to 5,000 subpages; larger sites need a monthly plan (see pricing). Download the ZIP to your own storage promptly, because the service keeps it only for a limited time: 1 day on the free plan, 30 days on paid plans.

How to keep a list of old URLs for 301 redirects

Redirect mapping goes wrong when nobody has a complete list of the old URLs. The archive gives you one, because the subpages are stored as folders and HTML files that follow the site's structure.

To turn the archive into a redirect map:

1. Unzip the archive and list every HTML file. On macOS or Linux run find . -name "*.html" in the unzipped folder. On Windows run dir /s /b *.html in Command Prompt.
2. Paste the list into a spreadsheet and turn the file paths back into URLs. Check a handful against the live site so you know exactly how paths map to addresses.
3. Add a few things a crawl cannot find on its own: URLs from sitemap.xml, top landing pages from your analytics, and pages with backlinks from your SEO tool or Search Console. Orphan pages that nothing links to only show up in these sources.
4. Add columns for the new URL, the redirect status and a "tested" checkbox.

Rules that keep the mapping useful:

• Redirect each old URL to its closest equivalent, not to the homepage. A blanket homepage redirect is often treated like a missing page.
• Avoid chains. If an old redirect already points A to B and B is now moving to C, point A straight at C.
• Don't forget PDFs and image URLs that other sites link to.
• After launch, run the old URL list against the new site and confirm every row returns a 301 to the right target.

For a page that is dropped without a replacement, the archive is still your record of what it said. It is also a quick way to rebuild that content later if it turns out to be needed.

Site builders, WordPress and custom sites: what each one needs

Wix, Squarespace and other site builders

Builders have no database you can download and no files to copy over FTP. Exports are partial at best: some platforms offer an export of blog posts or products, others nothing at all. When you redesign inside the same builder, the old design is often simply overwritten. Here the browsable copy is the backup, plus whatever native exports exist for products, customers or form entries. Platform-specific guides: Wix, Squarespace, Webflow and Shopify.

WordPress

Make a full files-plus-database backup with your host's snapshot tool or a backup plugin, and download it rather than leaving it on the same server. Then make the browsable copy as well. A database backup is only useful if someone can restore it on a compatible server. A static copy opens anywhere, even years later. The approach is explained in converting WordPress to static HTML.

Custom and headless sites

Tag the current state in version control, dump the database and save the server configuration, including redirect rules. If the front end renders content in the browser with JavaScript, test the offline copy carefully. Some single-page apps produce an archive that contains the page shell but not the content (why JavaScript sites can go blank offline).

If the relaunch also involves a new host, the same two-copy logic applies to the old hosting account; see backing up a website before cancelling hosting.

Will forms, search and member areas work in the archived copy?

No. A static copy records what the pages looked like. It does not keep the server that made them interactive.

• Forms appear but don't submit. Export past submissions from the CMS or form tool separately.
• Site search, shopping carts and comment forms need a server and won't work. Existing comments that were part of the page HTML remain visible.
• Pages behind a login, such as member areas, customer portals and intranets, are not downloaded. Export that content from the system itself.
• Sites that block automated downloads may answer with a 403. Website Downloader reports this instead of giving you an empty archive. If it is your own site, adjusting the firewall or bot protection for the crawl usually helps.

For a record and a reference, these limits rarely matter. For a rollback they do, which is why the restorable backup is still needed. Details on what is captured are on the features page and in the FAQ.

FAQ

Is Wayback Machine Save Page Now enough, or do I need my own copy?

Save Page Now captures pages one at a time, and the result is public and stored on someone else's service. It rarely gets every subpage, image and PDF of a site. You also can't search the capture as a folder or turn it into a URL list. Use it as an extra public record if you like, but keep your own complete copy.

How long should I keep the old version of the website?

At minimum until redirects are tested and traffic and rankings have settled, which usually takes a few months. A ZIP archive takes little space, so many teams keep it indefinitely as a record. If the site contained contractual terms, prices or regulated information, retention rules depend on your country and industry. Ask a lawyer or your accountant if that applies to you.

Who should keep the backup during a relaunch, and how do we roll back?

The site owner should hold copies of both the backup and the archive, not only the agency. Store them in two places, for example local storage and a cloud drive. Rolling back uses the restorable backup: restore the files and database, or point the domain back to the old host. In an emergency, the static archive can be uploaded to any web host as a temporary stand-in, without working forms.

The new site is already live. Can I still archive the old one?

Only from what others captured. If the Internet Archive has captures of the old site, the Wayback Machine downloader can rebuild them as an offline ZIP: free for up to 100 pages, or a one-time 14.99 EUR restore for up to 5,000 pages. The result is limited to what the archive captured; see restoring a website from the Wayback Machine.