Which broken images, PDFs, and embedded files do standard link checkers usually miss?

A page can pass a link check while its hero image, downloadable guide, or embedded video is dead. Here is what typical crawlers skip and how to close the gaps.

Julien Jimenez

September 7, 2026 · 3 min read

What a basic crawler actually checks

Most link checkers follow anchor tags, request each href, and report the status. Some also fetch the src of image tags. Far fewer look at srcset variants, picture sources, CSS background images, video and audio sources, iframes, object and embed tags, web fonts, favicons, or files referenced only from JavaScript. Almost none verify that the response is actually the type of file the page expected. The result is a report that says a page is clean while several things on it are visibly or invisibly broken. Related: How to Find Broken Links at Scale

Modern front ends make the blind spots worse. Lazy-loaded images keep the real URL in a data attribute and leave a placeholder in src, so the checker tests the placeholder. Responsive images list several files in srcset while only the fallback gets checked, and the larger files that most visitors receive can be missing. Download links served through a script commonly return a 200 with an HTML error page instead of the document, which a status-only check will happily record as success. Related: What a 404 Does to a Visitor, and How to Soften It

Keep reading: Why Broken Links Quietly Hurt Your SEO, Internal vs External Broken Links: Which to Fix First, How to Find Broken Links at Scale. See how LinkRotly helps you continuous broken link monitoring for websites.

Assets that break silently

PDFs and other downloads are the classic case. When a file is moved or renamed in a media library, the old URL often falls into a CMS catch-all that returns a 200 HTML page. The status looks fine and the visitor gets a web page instead of the guide they clicked. Checking the Content-Type header catches this, and for PDFs you can also read the first bytes of the file. Watch for zero-byte responses and for files that are dramatically smaller than the last time you fetched them, which usually means a truncated upload.

Embedded media fails without any HTTP error at all. An iframe to a video host or a map service shows a generic unavailable placeholder when the source is removed. Third-party widgets can be discontinued by the vendor and simply render an empty box. CSS background images referenced from stylesheets are rarely crawled, so hero sections go blank and stay that way. Web fonts that fail fall back to system fonts, which changes the look of every page and gets reported by nobody.

Building a more complete check

Start by extending what gets extracted from each page. Parse img src and srcset, picture source elements, data-src and similar lazy attributes, video and audio source elements, iframe src, stylesheet links and every url() reference inside those stylesheets, and download links identified by extension. Group results per page rather than per file, so a broken hero image appears next to the page that depends on it instead of as an anonymous URL in a list nobody can act on. Related: Why Broken Links Quietly Hurt Your SEO

Then validate the response rather than trusting the status alone. Compare the Content-Type against what the page expects, such as an image type for an img tag or application/pdf for a PDF link. Check that Content-Length is above a sensible minimum for the type. For images, optionally decode the header to confirm dimensions. For iframes, fetch the frame URL and look for the provider's unavailable markers when you can; when you cannot, at least confirm the frame URL itself responds with a success status.

Prioritizing what to fix

Rank by visibility and business impact. Images above the fold on high traffic pages come first, followed by downloadable assets referenced from campaigns and sales material, then embedded media, then decorative background images. A missing product image on a top category page costs more than a missing icon in an old blog post. When many pages share the same asset, restore the file at its original path once and every page recovers at the same time.

Prevention is mostly about discipline in the media library. Keep uploaded files under stable paths, avoid renaming or reorganizing media without adding redirects, and include asset directories in every migration redirect map. Make asset checks part of continuous monitoring rather than a one-time audit, because media libraries get cleaned up by people who have no way of knowing which pages reference which files. Related: Redirects Done Right: Avoiding Chains and Loops

Key takeaways

Most checkers follow href and img src and skip srcset, lazy attributes, CSS backgrounds, iframes, and fonts.
A PDF link that returns an HTML page with a 200 status is broken even though the status looks fine.
Validate Content-Type and Content-Length for every asset, not only the status code.
Rank fixes by visibility: hero images and campaign downloads first, decorative assets last.

Written by

Julien Jimenez

Julien Jimenez is an independent software builder based in Paris. He designs, ships, and operates focused SaaS products for small businesses and independent professionals. Read the full author page.