What is a broken-link check, and why does it matter?
The check follows links through a site and records what each address returns: a working page, missing content, a redirect or an error. A useful result identifies the source page, the destination and the action needed. A redirect chain is a finding to investigate, rather than necessarily a broken link: the final destination and its response matter.
What does Google do with a failing address?
Google does not index URLs returning 404 or 410. Previously indexed URLs are removed, and repeat crawling gradually becomes less frequent. This concerns the URL returning the error, rather than automatically the page linking to it.
Do not apply that rule to every 4xx code without exception. Google treats 429 as an overload signal, like 5xx server errors, and temporarily slows crawling. Persistent server errors can also lead to URLs being removed from the index. One failed response does not establish that a page has disappeared permanently.
A soft 404 is different: the server returns 200, but the content looks like a missing-page message or is empty. Google may report it separately in Search Console. A 200 response alone does not guarantee indexing; inspect the content too.
The documentation for these codes describes how the failing URL is handled. It does not establish a general direct ranking penalty for every page that links to it. The visitor’s problem is specific: a click does not reach the expected content.
Is every redirect a problem?
No. A redirect can correctly take someone from an old address to a new one. Check that the final destination works and matches what the link led the visitor to expect. For an internal link, you can often point directly to the correct final page instead of keeping unnecessary intermediate steps.
Google recommends avoiding long chains. For general web crawling, Googlebot normally follows up to 10 redirect hops; other tools and products can have different limits. That is not a recommended chain length. Check for loops, where addresses send the request back to each other, too.
How much does crawl budget matter?
This is the set of URLs Google can and wants to crawl, determined by crawling capacity and demand. On large or very frequently changing sites, unnecessary URLs, soft 404s and long chains can waste resources.
For a small site with a few hundred pages, start with important URLs working, the Page Indexing report and an up-to-date sitemap. Avoid presenting every broken link as a crawl-budget crisis. Repair still matters when budget is not the bottleneck: it restores the route to the content.
What else can a crawl reveal?
Differently sorted versions of a page and other duplicate or unnecessary URLs can increase crawling work. They are a separate kind of finding, rather than necessarily broken links. A noindex instruction does not save the request itself either: Google must receive the response to read it.
If a page has been removed, return an appropriate status code. If it has moved and there is a matching replacement, a redirect may be appropriate. Do not mechanically send every missing URL to the home page. A tool crawl and Search Console provide different views; your own crawler does not prove exactly how Google handled each URL.
What makes a useful fix list?
For each finding, record the source page, the link URL, the response code, intermediate and final URLs where redirected, and the proposed action. The person fixing it can then see whether to change a link, restore content, configure a redirect or investigate a temporary error.
This protects the routes to deeper pages for people and search engines. A broken link can hinder discovery, but does not necessarily mean there is no other route to that page.
Questions
What is the difference between a hard and a soft 404?
A hard 404 returns the actual 404 code. A soft 404 may return 200 while its content looks missing or empty. Check both the status and the page; Search Console may identify the soft case separately.
Does crawl budget matter for a small site?
It is usually not the first problem for a site with a few hundred pages. Start by checking that important URLs work and can be indexed. More detailed budget work depends on size, update frequency and reported problems.
Should every link pass through only one redirect?
Not necessarily, but remove unnecessary steps. Where possible, point internal links straight to the correct final URL. A crawler limit is not a target chain length.
Does noindex save crawling time?
It does not save the request needed to read the instruction. Google must first receive the response. Excluding a page from the index and stopping its crawling are different tasks.
Sources
Related reading
Next step
Restore the route to the next page
A Website Review crawls the whole site and returns a list of every broken link, redirect chain and soft-404 finding, with the address and proposed fix.
or write to contact@ergostrat.com