Best Broken Link Checker: 11 Tools Tested on the Same 3 Sites

Most “best broken link checker” lists are written by the people who make one of the tools, and almost all of them put their own product first. This one is too — ToolsVale is ours — so we have not ranked it first, and we have written down where it loses. What follows is what each tool actually caps you at, taken from the vendors’ own pages.

Which broken link checker is best for your situation?

The single biggest practical difference between these tools is whether the report tells you which of your pages contains the broken link. A flat list of dead URLs is only half a report.

What each tool actually gives you free

Every figure below came from the vendor’s own pricing or documentation page rather than from another comparison article. That matters more than it should: one widely shared list currently states Screaming Frog’s free version caps at 3,000 pages. It is 500, and Screaming Frog says so on their own site.

Free tiers change, so we re-check this table quarterly and date it.

Last verified: [DATE]

Three things stand out.

“Free” usually means “free until you hit a number.” Of the eleven, only W3C Link Checker states no cap, and it is slow enough that the practical limit is your patience. Everything else stops somewhere: 20 pages, 50 links, 100 pages, 500 URLs, 1,500 links. If your site is larger than 500 pages — and most real sites are — the majority of these will hand you a partial report and present it as finished.

Links and pages are not the same limit. Dr. Link Check’s 1,500 counts links, not pages, and that includes images, stylesheets and scripts alongside anchors. A page carrying 30 outbound references consumes 30 of your allowance. Tools that cap on links run out considerably faster than the number suggests.

Verification is a barrier for agency work. Ahrefs Site Audit needs you to prove you own the domain, via DNS record or a file upload. That is sensible for a monitoring product and awkward if you are auditing a prospect’s site before you have any access to it.

Why your link checker reports working pages as broken

If you have ever run a link check and got back “broken” links that load perfectly when you click them, this is why.

Free checkers crawl from a single IP address. Sites behind Cloudflare, Sucuri, or a strict rate limiter see a burst of automated requests from one address and refuse them — returning 403 Forbidden or 429 Too Many Requests. The page is fine. The crawler simply was not allowed in.

It is worth being precise, because “false positive” is the wrong word. A 403 does not mean the page is healthy; it means the crawler was not authorised to fetch it. The failure is not that the tool read the status wrongly — it is that the tool collapsed a different condition into the broken bucket. What you want from a report is the correct label, because the right response differs for each.

The practical test when choosing a tool: run it against a page you know is behind Cloudflare and see what it calls the result. If it says “broken”, every audit you run with it will contain work you do not need to do.

We should be clear about our own limitation here: ToolsVale crawls from a single IP too, so strict firewalls refuse us as well. What we do is label those as Blocked, separately from genuine breakage, so they do not enter your fix list. That saves time; it does not solve the underlying problem. No free tool solves it — that needs rotating proxies, which is what paid bulk checkers charge for. If a crawler is being refused by your own server rather than a third party’s, it is worth checking your robots.txt before assuming the tool is at fault.

Finding the link is the easy half

A report telling you https://example.com/old-page returns a 404 has done half a job. On a 5,000-page site, the question you actually need answered is: which of my pages links to it?

Most free checkers hand you a flat list of dead destination URLs. You are then left searching your own database for the offending href — which on a large site is slower than the crawl was.

The tools that report the source page turn that list into something a developer can act on directly. Screaming Frog, Ahrefs and ToolsVale all do it; several of the quick web-based checkers do not. If you are handing a fix list to someone else, this single column decides whether the report is useful or just evidence that a problem exists.

What crawlers cannot see

A crawler reads the HTML a server returns. Links that only appear after JavaScript runs are invisible to most of them, and this is rarely mentioned in tool comparisons even though it decides whether your report is complete.

If your navigation, related-posts block or product grid is rendered client-side, a standard crawler will not follow those links. Screaming Frog renders JavaScript on its paid tier; most free web-based checkers, including ours, do not. Neither can any crawler find a true orphan page — if nothing links to it and it is not in your sitemap, there is no path to it. Generating a complete XML sitemap is the only reliable fix for that, because crawlers read it as a starting point.

What “free” costs you in friction

Page limits are the obvious constraint. Friction is the one people notice second, and for an urgent job it decides which tool actually gets used.

Of the eleven tools here, four let you paste a URL and get a result with no account at all: ToolsVale, Broken Link Scan, W3C Link Checker, and Dead Link Checker for a single site. Dr. Link Check, Sitechecker and Ahrefs require registration. Screaming Frog requires a desktop install. The WordPress plugin requires a WordPress installation and an account.

None of that matters when you are planning an audit next week. It matters enormously at two in the morning during a migration, when you need to check a website for broken links immediately and an email verification step is the difference between using a tool and closing the tab.

The WordPress question

If you run WordPress, the standard advice is to install a plugin. The best-known option is Broken Link Checker by WPMU DEV, with over 17 million downloads.

There is a caveat worth knowing before you install it: several managed WordPress hosts do not allow it.

WP Engine lists Broken Link Checker among its disallowed plugins, stating that it overwhelms the server with a very large volume of HTTP requests. That is the plugin working as designed — to check every link on your site, something has to make a request for every link, and doing that from inside your own WordPress installation puts that load on the server your visitors are using.

WP Engine’s own guidance is to use an external tool instead, noting that because such tools are not plugins they have no effect on your site’s performance. That is a hosting provider recommending exactly the architecture an external crawler uses.

The plugin also appears on the disallowed or discouraged lists of Bluehost, Flywheel, GoDaddy, HostGator, Pagely and Pressable. Check your own host’s policy before installing — these lists change.

None of this makes it a bad plugin. On hosting that permits it, the monitoring and one-click fixes are genuinely convenient, and it is the only tool here that lets you fix a link without leaving the WordPress dashboard. But if you are on managed hosting, an external crawler is not merely an alternative — it may be your only option.

Source: WP Engine — Disallowed Plugins

Tool by tool

When ToolsVale is not the right choice

Since we build it, here is the version we would give a friend.

• You need weekly automated monitoring → Dr. Link Check or Dead Link Checker. We do not offer scheduling.
• You need a highly configurable desktop crawler → Screaming Frog.
• Your links are injected by JavaScript → a crawler that renders it. We do not.
• You already pay for Ahrefs → use Site Audit. You have already bought this.
• You want to see who links to your broken pages → Ahrefs. Its backlink database is unmatched here.
• You are checking thousands of external URLs → a paid tool with proxy rotation.

Where we do fit: a large one-off crawl, right now, without installing anything or creating an account — with the source page and anchor text for every broken link.

What no free tool in this comparison can do

Including ours:

• Render JavaScript-injected links, except Screaming Frog on its paid tier
• Discover true orphan pages — nothing links to them, so no crawler can reach them
• Avoid all 403 and 429 misclassification without proxy rotation
• Offer genuinely unlimited free crawling at useful speed

Do any of them have no limit at all?

Effectively no, and that includes ours.

W3C Link Checker states no cap but is slow enough that large sites are impractical. Everything else stops at a stated number. ToolsVale’s 10,000 pages per run is the highest free ceiling here, and “Crawl more” continues the same crawl rather than restarting, but it is still a ceiling.

Limits exist because crawling means one HTTP request per link, and someone pays for that compute. An uncapped crawl still has a practical constraint — crawl speed, server capacity, request rate, or how long you are willing to wait.

Frequently asked questions