Why some pages do not appear in the Wayback Machine
Finding a domain on the calendar does not mean every URL, image and feature was preserved at the same moment.
R42 / SUMMARY
A page can be missing from the Wayback Machine because its URL was never discovered, the crawler could not pass a login or form, access rules prevented collection, or required external resources were not captured. Even when the HTML was saved, images, scripts and later-loaded data may be absent or come from different moments. Save Page Now records one specific page once; it does not schedule future captures or automatically archive an entire site.
KEY POINTS
- The Wayback Machine preserves URLs that are discovered and accessible, not a guaranteed copy of the entire web.
- Save Page Now captures one specific page once and does not automatically include the rest of the site.
- Images, scripts, fonts and API responses are separate resources that must also be collected.
- Sites that depend on logins, forms, clicks or dynamic content are harder to capture and replay.
- An archived capture should not be treated as a replacement for a backup controlled by the site owner.
The Wayback Machine can open older versions of web pages, but its calendar is not a complete backup of every site. A capture begins with a URL the system was able to discover and access. From there, the crawler attempts to retrieve the main document and the resources required to replay it. If one step fails, a record may exist in the archive while still appearing without images, menus, video or content.
This distinction explains why a domain’s homepage can have hundreds of dates while a particular article is absent. Web archiving collects a network of addresses and responses; it does not photograph an entire site as one indivisible object.
The crawler must discover the URL
Crawlers commonly reach new pages by following links. A URL with no public links pointing to it, sometimes called an orphan page, can remain unknown. The same applies when content only appears after someone performs a search, submits a form or signs into an account: the robot does not navigate the site as an authenticated person.
The Internet Archive also says pages may be absent because they were password-protected, inaccessible to automated systems, blocked by crawling rules or removed following a request from the party responsible for the site. A public address therefore does not imply that every version of it was collected.
A page consists of many resources
The HTML delivered to a browser is only the initial structure. Images, style sheets, fonts, scripts, video and data can come from different URLs and domains. Each resource must be requested and preserved. If an image was not collected, it appears broken; if a script depends on the original server, its feature may stop working when replayed from the archive.
Modern sites make the process harder because they often deliver nearly empty HTML and use JavaScript to request data afterward. Interactive maps, click-driven pagination, galleries, media players and 3D tours are among the examples Archive-It identifies as difficult to capture completely. Tools that behave more like a browser can record additional interactions, but the service acknowledges that some dynamic material still cannot be collected or replayed in full.
A replay can combine different moments
When a resource is missing from the selected date, the Wayback Machine may use the nearest available copy. This can help an incomplete page open, but it introduces another limit: the HTML, an image and data requested by a script may have been captured at different hours or on different days.
Researchers studying pages assembled in the browser have shown that old HTML can be replayed with JSON responses from another time. This is described as a temporal violation: the screen resembles a single snapshot even though its components did not necessarily coexist in that form. The displayed date should be understood as a reference for the main URL, not a guarantee that every element is perfectly simultaneous.
Save Page Now does not preserve an entire site
Save Page Now requests a one-time capture of a specific page. According to Internet Archive documentation, it does not add the address to future crawls and does not automatically save related pages, directories or a complete domain. Visible links can still lead to URLs that were never archived.
After requesting a capture, it is worth opening the archived version and checking its text, images and navigation. For material you control, this should complement local copies and exports rather than replace them. Wayback Machine availability addresses part of the link-disappearance problem, while protection against bit rot and silent alteration requires another layer of verification, redundancy and maintenance.
In short, a date on the calendar confirms that an attempt or response was associated with the URL, but it does not prove that the full site was preserved. The archive remains immensely valuable; using it correctly means recognizing that a web capture also has a scope, dependencies and failure modes.
Misael
Responsible for reporting and writing this story at Rota42.
R42 / FAQ
Does the Wayback Machine automatically archive every website?
No. Crawlers must discover public URLs and be able to access them. Isolated, protected, excluded or interaction-dependent pages may not be collected.
Does Save Page Now save an entire website?
No. According to the Internet Archive, it saves one specific page once. It does not automatically archive directories, every subpage or schedule future captures.
Why does an archived capture contain broken images?
Because the HTML and each image are separate resources. If an image file was not captured, came from another domain or required the original server, it may be missing during replay.
Can the Wayback Machine save pages behind a login?
It generally does not preserve content that requires a password or form submission. The archive collects publicly accessible material rather than replaying a user’s private session.
Does a Wayback Machine capture replace a backup?
No. The Internet Archive does not guarantee that a site has been or will be archived. Content owners should keep their own copies and verify their integrity.