Restore Website from Wayback Machine: Step-by-Step Guide

You can restore a website from Wayback Machine captures even without a backup, provided the Internet Archive crawled your pages before they went offline. The job has three parts. First, list every URL the Archive saved. Next, choose a snapshot date when the site looked complete. Then download those captured pages, with their images and stylesheets, into a folder or ZIP that opens offline. What you get back is a static copy of what visitors saw. It does not include your database, admin panel or server code. This guide explains how to restore a website from Wayback Machine captures and what such a copy can and cannot give you.

What can the Wayback Machine give back, and what can't it?

The Wayback Machine is a public library of web captures. It is not a backup service. Its crawlers visit sites on an irregular schedule. A small or new site may have been captured a handful of times or not at all. Some pages were never reached because nothing linked to them, and others were later excluded at the owner's request. There is no guarantee your site is in there, so check before you plan anything around it.

The Internet Archive also won't recover a deleted website from archive.org for you. It doesn't package lost sites into downloads or send files to former owners. The captures are public, and getting them back out is up to you.

When a site was captured well, you can usually recover:

• The HTML of each captured page, as a visitor's browser received it
• Images, CSS, JavaScript files and fonts that the crawler fetched
• PDFs and other linked documents, if they were captured

You cannot recover anything that only existed on the server:

• The database (WordPress posts as database rows, users, orders, comments as data)
• PHP or other server-side code, plugin settings, the admin area
• Pages behind a login, form submissions, email accounts

So you can recover an old website as something people can read and browse. You cannot get back a working copy of the software that ran it.

Step 1: How do I see every page and file the Wayback Machine saved?

Before you look at any single snapshot, get the full inventory. Type this into your browser's address bar, replacing the domain with your own:

web.archive.org/*/www.yoursite.com/*

The asterisk after the slash is the important part. It switches the Wayback Machine from the calendar view to a URL listing. This shows every address it holds under that domain, with the dates of the first and last capture and the number of captures. There is a filter box for narrowing the list, for example to /blog/ or to .jpg files.

Some tips for reading the list:

• Try variants. Run the search with and without www. If the site ever moved between http and https, captures can sit under both.
• Check your most important pages by name: home, about, contact, best-performing posts, product or service pages. Note which are missing.
• Look at the files, not just the pages. Images, stylesheets and scripts appear as their own URLs. If the CSS file was never captured, the restored pages will look unstyled.

If you are comfortable with raw data, the Archive's CDX interface returns the same inventory as plain text or JSON, one capture per line. Use web.archive.org/cdx/search/cdx?url=yoursite.com/*. It is handy for counting pages or feeding the list to a script.

Step 2: Choose the right snapshot

Open the calendar view at web.archive.org/web/*/yoursite.com. Each dot marks a capture, and the colour shows what the crawler got back. Blue is a normal page. Green is a redirect. Orange is a client error such as 404. Red is a server error. You want a date with a blue dot on the home page and on the pages that matter.

Which date is the right one depends on how the site was lost:

• Hosting expired or the account was closed: pick the last capture before the site turned into an error page or a hosting company's placeholder.
• Site was hacked: go further back. Look for the last capture before spam links, odd redirects, foreign-language pages or injected scripts appeared. A recent but compromised snapshot can bring the hack back with it.
• Domain was sold or parked: pick a date before the parking page took over. That page often gets captured many times and can drown out the real site.

Then click through the snapshot as a visitor would and watch the dates in the toolbar. When you follow a link, the Wayback Machine doesn't promise to show that page from the same day. It shows the closest available capture of that URL, which can be months or even years away. A snapshot from 2019 can lead you to an "About" page from 2016 and a product page from 2021. A restored copy built from it mixes versions in the same way. That's usually acceptable, but check the pages where old prices, staff names or contact details would cause trouble.

Step 3: Restore the website from Wayback Machine as a ZIP

Once you know what exists and which period you want, you have three practical ways to download the entire website from the Wayback Machine.

Option 1: Copy pages by hand

For a site of five or ten pages, you can open each snapshot and use your browser's "Save page as". The catch is that a normal snapshot includes the Wayback toolbar and rewrites every link to point back to web.archive.org. A useful trick avoids this. Add id_ directly after the timestamp in the address, for example web.archive.org/web/20190315000000id_/http://www.yoursite.com/. This returns the file as it was originally captured, without the toolbar or rewritten links. You still have to save every image and stylesheet yourself and fix the paths. That is fine for a few pages and tedious for fifty.

Option 2: Command-line tools

Several open-source downloaders are built for the Wayback Machine, and some of them need Ruby installed first. They query the capture list and fetch the original files in bulk. They are flexible and free, but you have to install them and work out their options. Pointing a general mirroring tool such as wget --mirror straight at web.archive.org tends to work badly. It follows the rewritten links and the toolbar and pulls in far more than your site. If you are deciding between command-line and browser-based tools in general, see wget vs HTTrack vs an online website downloader.

Option 3: An online restore with Website Downloader

The Wayback Machine downloader on this site does the job in the browser, with nothing to install. You enter the domain, the service rebuilds the site from its Internet Archive capture, and you receive one ZIP file that opens offline in any browser. The Wayback playback toolbar and the links back to archive.org are removed, so the pages link to each other as they did on the live site.

• Free: a restore of up to 100 pages, with no account needed.
• Wayback Restore: a one-time payment of 14.99 EUR for one website, up to 5000 subpages. The archive is kept for 30 days so you can download it.

Monthly plans do not include restores; they are for downloading live websites. Details are on the pricing page. For more on how the ZIP format works in general, see Download a Website as ZIP Online.

Why are images, styles or pages missing from the restored copy?

Gaps are normal, even with a good tool, because a download can only contain what the Archive captured. Gray placeholder images and the occasional missing page don't necessarily mean anything went wrong. The usual causes are:

• Never captured. The crawler didn't reach that page or file, often because it was deep in the site or only linked from a form or script.
• Captured as an error. The only capture is a 404, a 500 or a redirect to a login page.
• Hosted elsewhere. Images on a CDN, web fonts from a font service and embedded videos sat on other domains and may not have been captured alongside your pages.
• Built by JavaScript. Content loaded by scripts after the page opened often wasn't in the captured HTML at all.
• Excluded. The site's robots.txt or a removal request kept parts of it out of the Archive.

To fill the gaps, first go back to the URL listing from Step 1. A missing image may exist under a different date, with or without www, or over http instead of https. After that, look outside the Archive. Check old files on your own computer, email attachments, the original photos from whoever took them, drafts in a document editor, exports you once made from your CMS, and posts you shared on social media. The restored copy gives you the structure and most of the text, and these sources can fill in the rest.

Can I put the restored site back online or into WordPress?

Yes, and how much work it takes depends on what you want the site to do.

As a static site on a new host

This is the quickest route. Unpack the ZIP and upload the contents to your web space so that index.html sits in the root folder, usually public_html or www. Then point your domain at the new host. Any ordinary hosting package can serve plain HTML, and so can static hosting services. Keep the original page paths where you can, so that old links and search results still lead to the right pages.

Rebuilt in WordPress or another CMS

A restored copy is HTML files, not a WordPress database, so there is no one-click import back into the old installation. In practice you install a fresh WordPress and rebuild page by page. Copy the text from the restored HTML, upload the images, recreate menus, and set the permalinks to match the old URLs. It is slow, but you end up with an editable site, and the restored copy works as a complete reference while you rebuild.

What stops working

Anything that needed the server to respond will not work in a static copy. Contact forms don't send, site search returns nothing, shopping carts can't take orders, comment sections can't accept new comments, and member logins have nothing to log into. You will need to replace these, for example with a form service, a CMS, or a shop platform.

Ownership, copyright and expired domains: what to settle before republishing

This section explains general principles, not legal advice. The rules differ by country and depend on the facts of your case.

• Your own site: if you or your organisation created the content, restoring and republishing it is generally straightforward. Watch for third-party material you had licensed, such as stock photos, fonts and themes, if the licence was tied to the original site or has expired.
• An expired domain you bought: owning the domain name does not give you the copyright in the texts and images that used to be published under it. Those normally still belong to their authors. Republishing them as your own can infringe copyright and mislead visitors who knew the old site.
• Personal data and trademarks: old pages can contain names, photos and contact details of real people, plus brand names that someone else still uses. Republishing these raises separate questions.
• A client or employer's site: confirm in writing that you are authorised to restore and publish it.

If money, a business or a dispute is involved, or you plan to republish content you didn't write, speak to a lawyer in your jurisdiction first.

FAQ

Can I get my website back if my hosting expired or the site was hacked and I have no backup?

Often yes, as long as the Wayback Machine captured the site before it went down. Use the URL listing to check, then choose a snapshot from before the hosting lapsed or before the hack. You will get the public pages and files back, but not the database, the admin area or any server code.

Will the Internet Archive restore my lost website for me?

No. The Internet Archive preserves captures for public access, but it does not package sites for their former owners or guarantee that a site was ever captured. You restore the site yourself, by hand, with a command-line tool, or with an online service such as the Wayback Machine downloader.

Does Save Page Now save my whole site?

No. Save Page Now captures the page you submit, not a full crawl of your domain. To make sure several pages exist in the Archive, you would have to submit each one. For your own backup, it is simpler to download a ZIP of your live site regularly.

How do I remove the Wayback toolbar and the links to archive.org?

For single pages, add id_ after the timestamp in the snapshot address to get the file without the toolbar or rewritten links. For a whole site, a dedicated restore tool is far quicker. Website Downloader's restore removes both the playback toolbar and the archive.org links automatically.

How much does it cost to restore a site this way?

Doing it by hand or with open-source tools costs nothing but time. With Website Downloader, a restore of up to 100 pages is free. A larger site of up to 5000 subpages is a one-time 14.99 EUR Wayback Restore, kept for 30 days. More answers are in the FAQ.