A link audit is often treated as a broken link check.
That is understandable. Broken links are easy to explain. A page points to a URL, the URL returns a 404, and someone needs to fix it.
But a useful link audit goes further than that.
Before publishing a page, you should also ask whether the links make sense. Are internal links helping readers move to the right next page? Is the anchor text descriptive, or does the page rely on vague labels like “click here” and “read more”? Are important internal links accidentally marked as `nofollow`? Do external links that open in a new tab use safe attributes? Are redirects hiding behind links that look clean on the surface?
A page can have zero broken links and still have weak links.
That is why internal links and anchor text deserve their own pre-publish review.
Why internal links matter before publishing
Internal links are part of how a website explains itself.
They help readers move from one useful page to another. They help search engines discover URLs. They show relationships between topics. They connect commercial pages, documentation, support content, blog posts, comparisons, and product explanations.
A good internal link says:
```txt This page is related to that page, and the words in this link explain why. ```
A weak internal link says:
```txt Something is probably connected here, but the context is unclear. ```
That may sound small, but it matters during publishing.
New pages are often added quickly. A blog post links to an old article. A landing page points to documentation. A pricing page links to terms. A feature page links to a case study. A migration changes URL paths. A CMS component adds the same “Learn more” link everywhere.
By the time the page goes live, the links may technically work, but they may not be helping the reader or the crawler understand the site.
A link audit is not only a status-code check
A basic link check asks:
- Does the URL return 200?
- Does it redirect?
- Does it return 404?
- Does it timeout?
That is useful, but it is only the first layer.
A stronger audit asks:
- Is the link internal or external?
- Is the destination relevant?
- Is the anchor text clear?
- Is the same generic anchor repeated many times?
- Does the link point to the canonical version of the URL?
- Does the link go through an unnecessary redirect?
- Is the internal link accidentally marked `nofollow`?
- Does an external link open in a new tab safely?
- Does the link appear in a useful context?
- Does the page link to the next page a reader would naturally need?
That is the difference between checking whether links function and checking whether links communicate.
Both matter.
Start with internal links
Before reviewing every external reference, start with internal links.
Internal links are the links you control most directly. They shape the paths users and crawlers follow through your site. They also help define which pages are important.
On a typical page, review links in:
- navigation
- footer
- body copy
- cards
- buttons
- related article blocks
- product sections
- documentation references
- breadcrumbs
- sidebar menus
- FAQ answers
- call-to-action areas
A page may have many internal links, but not all of them carry the same weight. Focus first on links that affect the main user journey.
For example, on a SaaS landing page, the most important internal links may be:
- pricing
- documentation
- product demo
- support
- security
- changelog
- comparison pages
- related use cases
On a blog post, the most important internal links may be:
- related guides
- product pages
- glossary pages
- technical documentation
- previous articles in the same topic cluster
A good audit asks whether those links are present, useful, and easy to understand.
Audit anchor text
Anchor text is the visible text inside a link.
For example:
```html <a href="/blog/schema-markup-audit">schema markup audit</a> ```
In this case, `schema markup audit` is the anchor text.
Good anchor text gives readers a clear expectation of what they will get after clicking. It also gives search engines more context about the linked page.
Weak anchor text hides the destination behind vague language.
Bad examples:
```html <a href="/blog/schema-markup-audit">click here</a> ```
```html <a href="/blog/schema-markup-audit">read more</a> ```
```html <a href="/blog/schema-markup-audit">this article</a> ```
Better examples:
```html <a href="/blog/schema-markup-audit">schema markup audit checklist</a> ```
```html <a href="/blog/check-broken-links-before-launch"
broken link launch checklist</a
```
```html <a href="/blog/how-to-audit-images-for-seo-before-publishing-a-page"
image SEO audit guide</a
```
The better anchor text is not longer because longer is always better. It is better because it describes the destination.
A person can understand what the link offers before clicking. A crawler can understand how the source page relates to the destination page.
Do not turn anchor text into keyword stuffing
Descriptive anchor text is good. Over-optimized anchor text is not.
This is too much:
```html <a href="/blog/schema-markup-audit"> best schema markup audit checklist JSON-LD structured data SEO validation tool </a> ```
That does not read like a natural link. It reads like someone tried to put every keyword into one clickable phrase.
A better version:
```html <a href="/blog/schema-markup-audit"> schema markup audit checklist </a> ```
Anchor text should be specific, but it should still sound like something a human would write.
A practical rule:
```txt If the anchor text feels awkward in the sentence, rewrite the sentence. ```
Do not force an exact keyword where a natural phrase would work better.
Watch for generic anchor text
Generic anchor text is one of the most common link quality issues.
Examples include:
- click here
- read more
- learn more
- more
- here
- this page
- this article
- visit
- link
- details
- check it out
These labels are not always wrong. A button that says “Learn more” can work when the surrounding card clearly names the destination. But repeated generic anchors become a problem when they appear without context.
For example:
```html <p> We also wrote about schema validation. <a href="/blog/schema-markup-audit">Read more</a>. </p> ```
The destination is understandable from the sentence, but the link itself is still weak.
Better:
```html <p> We also wrote a practical <a href="/blog/schema-markup-audit">schema markup audit guide</a>. </p> ```
Now the link carries meaning on its own.
That is especially useful for accessibility, scanning, and search understanding.
Review repeated anchors
A single generic link may not be a serious issue.
A page with twenty repeated “Learn more” links is different.
Repeated anchors are common in:
- blog cards
- product cards
- pricing sections
- comparison tables
- feature grids
- documentation lists
- ecommerce category pages
- related article components
Sometimes the repeated label is acceptable because each card has a heading and the link is visually grouped with it. But the audit should still check whether the link has enough accessible context.
A page full of identical anchors can be hard to interpret outside the visual layout. It can also make exports, accessibility reviews, and crawler analysis less useful.
If the link text repeats, ask:
- Is the destination clear from nearby text?
- Is the accessible name unique enough?
- Does the component include a hidden label if needed?
- Would this link make sense in a list of links?
- Is a more descriptive anchor possible without making the design worse?
Good link UX and good SEO often meet in the same place: clarity.
Check internal nofollow
Internal links should usually be followed.
There are exceptions, but an internal `nofollow` often deserves review because it may be left over from an old SEO habit, plugin setting, copied template, or mistaken CMS configuration.
Example:
```html <a href="/pricing" rel="nofollow">Pricing</a> ```
Before publishing, ask why that attribute exists.
If the destination is a normal public page, `nofollow` may send the wrong signal. If the page should not be crawled or indexed, there are usually better controls than randomly adding `nofollow` to internal links.
Use the right tool for the job:
- use internal links when a page should be discoverable
- use `noindex` when a page should not appear in search results
- use authentication for private content
- use robots rules carefully for crawl control
- avoid using internal `nofollow` as a vague “SEO safety” switch
The goal is not to remove every internal `nofollow` automatically. The goal is to make sure each one is intentional.
Check external links
External links are part of a page’s trust surface.
They may point to documentation, research, sources, tools, partner pages, customer stories, social profiles, or examples. They can help support a claim, but they can also create risk if they are outdated, broken, low-quality, or added without review.
Before publishing, check:
- Does the external link still work?
- Is the destination the intended page?
- Is the source trustworthy enough to cite?
- Does the link redirect somewhere unexpected?
- Is the link commercial, sponsored, user-generated, or editorial?
- Does it open in a new tab?
- Does it need a `rel` attribute?
For normal editorial links, you usually do not need to add special `rel` attributes just because the link is external.
But some links should be qualified.
For example:
```html <a href="https://example.com" rel="sponsored">Sponsored partner</a> ```
```html <a href="https://example.com" rel="ugc">Community submission</a> ```
```html <a href="https://example.com" rel="nofollow">Untrusted reference</a> ```
Use these attributes when they describe the relationship honestly. Do not add them mechanically to every outbound link.
Check noopener on new-tab links
Many websites open external links in a new tab:
```html <a href="https://example.com" target="_blank">External resource</a> ```
When a link opens a new browsing context, `rel="noopener"` is commonly used to prevent the opened page from having access to the original page through `window.opener`.
A safer pattern:
```html <a href="https://example.com" target="_blank" rel="noopener"> External resource </a> ```
Some modern browser behavior has reduced the risk in common cases, but it is still worth auditing new-tab links explicitly. It keeps the code clear, helps older patterns surface during review, and prevents avoidable security warnings in many audit workflows.
For untrusted external links, teams may use:
```html rel="noopener noreferrer" ```
The difference is that `noreferrer` also affects referrer information. Use it when that is the intended behavior, not just as a habit.
Check links inside buttons and cards
Not all links look like text links.
Many important links appear as:
- buttons
- cards
- image links
- icon links
- navigation items
- carousel items
- table rows
- feature tiles
These need the same review as regular links.
A button that says “Get started” may be fine if it goes to signup. But if the same page has several “Get started” buttons pointing to different destinations, the context should be clear.
A card may have a visible title, but the actual clickable element may only say “Read more.” Depending on the markup, the accessible name may not be as clear as the visual layout suggests.
A linked icon may look obvious to a designer but be unclear to a screen reader or in an extracted link report.
During the audit, do not only scan paragraphs. Inspect the actual links the page creates.
Check image links
Image links need special attention.
Example:
```html <a href="/product"> <img src="/images/product-dashboard.webp" alt="Product dashboard" /> </a> ```
If the image is the only content inside the link, the image alt text becomes the accessible label for the link.
That means the alt text should describe the destination well enough, not only the visual object.
Weak:
```html <img src="/images/dashboard.webp" alt="Dashboard" /> ```
Better:
```html <img src="/images/dashboard.webp" alt="Open the product dashboard overview" /> ```
If the linked image is decorative and there is nearby text link context, the implementation may differ. The main point is simple: image links should not become nameless links.
A link without a meaningful accessible name is a usability issue before it is an SEO issue.
Check redirect chains
A link can work and still be inefficient.
For example:
```txt /page-old → /old-page/ → /blog/old-page/ → /blog/current-page/ ```
A user may eventually reach the right page, but the link is still messy. Redirect chains slow down navigation, add crawl friction, and make QA harder.
Before publishing, review links that redirect.
Ask:
- Is the redirect intentional?
- Is it a single clean redirect?
- Can the source link be updated to the final URL?
- Does the redirect point to the right destination?
- Is the redirect temporary or permanent?
- Does the final page match the link promise?
For internal links, the preferred approach is usually to link directly to the final canonical URL.
Redirects are useful safety nets. They should not become the normal internal linking strategy.
Check links to noindex or blocked pages
A link may return `200 OK` and still point to a page that is not useful for discovery.
Examples:
- page has `noindex`
- page is blocked by `robots.txt`
- page canonicalizes somewhere else
- page requires authentication
- page is a staging or preview URL
- page has thin placeholder content
- page redirects to a generic homepage
- page is excluded from sitemap intentionally
These are not always errors. But they should be intentional.
If a high-value article links to a support page that is `noindex`, that may be fine. If a product page links to a feature page that is accidentally `noindex`, that is a launch issue.
A mature link audit looks beyond HTTP status.
Check link context
A good link is not only about the anchor text. It is also about the sentence around it.
This is weak:
```md For more details, read our guide. ```
This is better:
```md For a deeper review of crawl access, read our robots.txt and sitemap audit guide. ```
The second version tells the reader why the link exists. It also connects topics clearly.
Link context matters because readers rarely click every link. They scan. They decide whether a link is worth their time based on the words around it.
Search systems also use surrounding context to understand why pages are connected.
A good internal link should feel like a natural next step, not a random SEO insertion.
Avoid overlinking
Internal links are useful, but more is not always better.
A page that links every other sentence can become hard to read. It can also dilute the editorial purpose of the page.
Overlinking often happens when teams try to force every related keyword into every article. The result is not a stronger page. It is a noisy page.
Before publishing, ask:
- Does this link help the reader?
- Is the destination the best next page?
- Is this link repeated too often?
- Would the page still make sense without it?
- Is the link placed where the topic naturally appears?
Good internal linking is selective. It should create useful paths, not clutter.
Build topic clusters with links
Internal links are especially powerful when they connect related pages into a clear topic cluster.
For example, a technical SEO blog may connect:
- website launch SEO checklist
- on-page SEO audit checklist
- broken link checker guide
- robots.txt and sitemap audit
- image SEO audit
- schema markup audit
- AI readiness SEO checklist
- SEO audit Chrome extension guide
Each article can stand alone, but internal links help readers move through the cluster.
A page about internal links should naturally link to a broken link guide, an on-page SEO checklist, and a broader Chrome audit workflow. A page about robots and sitemaps should link to launch SEO and crawlability topics. A schema article should connect to rich results and entity clarity.
This is not just SEO housekeeping. It is editorial architecture.
Check links after migrations
Link audits become more important after migrations.
Migrations often change:
- URL paths
- trailing slashes
- domains
- protocols
- CMS slugs
- blog structures
- documentation paths
- language routes
- product page URLs
After a migration, many internal links still work only because redirects catch them.
That is not ideal long term.
A post-migration link audit should update internal links to final URLs wherever possible. It should also remove links to staging domains, preview URLs, old documentation paths, and outdated support articles.
A migration is not finished when redirects work. It is finished when the site points cleanly to itself again.
Common internal link mistakes
### Using “click here” everywhere
Generic anchors make links harder to understand. Replace them with descriptive phrases when possible.
### Linking to redirected URLs
If the final URL is known, link to the final URL directly.
### Accidentally using internal nofollow
Internal `nofollow` should be reviewed carefully. It is often not the right control for normal public pages.
### Linking to old content
A page can still load but be outdated, inaccurate, or no longer the best destination.
### Repeating the same anchor too often
Repeated anchor text may be fine in navigation, but repeated generic body links deserve review.
### Linking without context
A link should feel like a natural next step in the sentence or section.
### Ignoring cards and buttons
Many link issues hide in components, not paragraphs.
### Forgetting mobile navigation
A desktop audit can miss mobile-only menus, sticky buttons, collapsed sections, and footer variations.
Common external link mistakes
### Opening links in a new tab without review
If external links use `target="_blank"`, review whether `noopener` or `noreferrer` is appropriate.
### Using nofollow on every outbound link
Normal editorial links do not automatically need `nofollow`. Use link qualifications when they reflect the relationship.
### Forgetting sponsored or user-generated links
Paid, sponsored, affiliate, or user-generated links may need specific rel attributes.
### Linking to unstable sources
External pages can change, redirect, or disappear. Important references should be checked before publishing.
### Linking to low-quality sources
A link is also an editorial signal. Do not cite sources you would not stand behind.
A practical link audit checklist
Use this before publishing a page, refreshing an article, shipping a landing page, or completing a migration.
### Internal links
- Check all internal links on the page.
- Confirm important links return clean responses.
- Update redirected internal links to final URLs.
- Remove links to staging or preview environments.
- Check whether internal links point to canonical URLs.
- Review links to `noindex` or blocked pages.
- Make sure key related pages are linked.
- Avoid unnecessary duplicate links in the same section.
- Check mobile-only navigation and footer links.
### Anchor text
- Replace vague anchors like “click here” where possible.
- Avoid repeating “read more” without context.
- Make anchors descriptive but concise.
- Keep anchor text natural in the sentence.
- Avoid keyword stuffing.
- Make sure links are understandable when scanned.
- Review repeated anchor text across cards and lists.
- Check image links for meaningful accessible labels.
### External links
- Check external link status.
- Review redirects and final destinations.
- Use `sponsored`, `ugc`, or `nofollow` where appropriate.
- Do not add rel attributes mechanically to every external link.
- Review `target="_blank"` links.
- Add `noopener` where appropriate.
- Check whether cited sources are still accurate and useful.
### Link context
- Make sure each link supports the section where it appears.
- Link to the best destination, not just any related page.
- Avoid overlinking.
- Connect related articles into topic clusters.
- Make conversion links clear and easy to find.
- Make support, pricing, documentation, and policy links easy to reach when relevant.
How to audit links with Crowra
You can review links manually, but it becomes slow when a page has navigation links, footer links, cards, buttons, body links, social links, external references, and related content blocks.
Crowra is designed for that pre-publish review.
Open the page in Chrome, run Crowra, and review link signals beside the page you are checking. Crowra helps surface page links, link status checks, redirect chains, internal and external link patterns, generic anchor text, internal `nofollow`, and external links missing `noopener` where relevant.
That matters because link quality is not one issue.
A page can have no 404s but still rely on vague anchor text. It can have useful body links but accidentally nofollow an internal product page. It can use strong internal links but send users through old redirects. It can open external resources in new tabs without the attributes your team expects.
A useful link audit puts those signals in one workflow instead of making you jump between the DOM, source code, spreadsheets, browser previews, and separate link checkers.
What to fix first
If the audit finds many link issues, start with the ones that affect users and discovery most.
A good priority order:
- Fix broken internal links.
- Remove staging or preview URLs.
- Update redirected internal links to final URLs.
- Fix links to important `noindex` or blocked pages.
- Replace vague anchor text on important links.
- Review internal `nofollow` on public pages.
- Fix external links that are broken or misleading.
- Add safe attributes to new-tab external links where appropriate.
- Improve repeated card and button anchors.
- Add missing links to related high-value pages.
Do not start by rewriting every small footer link if the main CTA points through a redirect. Do not polish external references while important internal links are broken. Fix the links that affect the page’s purpose first.
When to repeat the audit
Repeat this audit when:
- publishing a new page
- updating a major article
- launching a landing page
- changing navigation
- migrating URLs
- changing CMS templates
- adding a new blog cluster
- updating product pages
- redesigning cards or buttons
- changing documentation structure
- seeing unexpected crawl or indexing issues
- preparing a page for handoff to another team
Link problems often appear after normal work. Someone edits a CTA. Someone renames a page. Someone changes a component. Someone removes an old article. Someone updates the site navigation.
A link audit is not a one-time cleanup. It is part of publishing discipline.
The bottom line
Internal links and anchor text are easy to underestimate because they do not look technical.
They look like words.
But those words help people navigate. They help crawlers understand relationships. They connect topic clusters. They support discovery. They guide users toward product pages, documentation, support, and next steps.
A good link audit is not only about finding broken URLs.
Before publishing, check whether your links are descriptive, intentional, crawlable, secure, and useful. Check whether they help the page explain itself. Check whether the anchor text sets the right expectation. Check whether the linked destination is still the best next step.
The best links do not feel like SEO work.
They feel like good editing.