Noindex vs Nofollow: What Is the Difference (and Robots.txt)?

What is the difference between noindex vs nofollow?

Noindex vs nofollow comes down to one split: noindex tells Google not to show a page in search results, while nofollow tells Google not to follow the links on it. One controls indexing, the other controls link handling. Robots.txt is a third tool that only controls crawling.

Many people mix these up because all three seem to "talk to Google". However, each one acts at a different stage. To understand noindex vs nofollow, you need to know how Google processes a page: first it crawls, then it indexes, and finally it ranks.

In this guide, we explain the difference using Google's official documentation. We also compare both against robots.txt and show which tool fits which job. In our technical SEO audits, confusing these three is one of the mistakes we see most often.

In what order does Google process a page, and why does it matter?

Google handles a page in three steps. First, Googlebot requests the URL. Next, Google processes the page and decides whether to add it to the index. Finally, it ranks indexed pages for each query.

Each tool works on a different link in this chain, so the order matters. First, robots.txt affects crawling. Second, noindex affects indexing. Finally, nofollow is a hint about how to treat links.

• Crawling: can Googlebot reach the URL? Robots.txt decides.
• Indexing: can the page appear in results? Noindex decides.
• Link handling: does Google follow the links? Nofollow hints at it.

So "which one is stronger?" is the wrong question. Ask which step you want to control instead. The answer tells you which tool to pick.

Once you see this order, debugging gets easier. If a page is missing from results, check first whether it was crawled. Then check whether it carries noindex. Only then should you look at ranking.

What does noindex do and how do you add it?

Noindex is a directive that tells Google not to show a page in search results. Google's robots meta tag documentation describes it as "do not show this page, media, or resource in search results". The page stays crawlable, but it does not enter the index.

You can add it in two ways. The first is a robots meta tag in the HTML head. The second is an X-Robots-Tag value in the HTTP response header.

• HTML pages: place <meta name="robots" content="noindex"> in the head.
• Files such as PDFs and images: send X-Robots-Tag: noindex in the server response.
• Targeting Google only: set the meta name to "googlebot".

Most CMS platforms offer a checkbox for this. Even so, you should confirm that the tag is live in the source code or in Search Console. Our meta tag generator helps you produce the right syntax.

Treat noindex as a managed setting, not a permanent verdict. Also, when the purpose of a page changes, revisit the tag.

What does nofollow do, and what are its two uses?

Nofollow is a hint to not follow links, and you use it in two places. One is the page-level robots meta tag. The other is the rel attribute on a single link. They do not do the same job, and that is where most of the confusion starts.

A page-level nofollow asks Google not to follow any link on that page. A link-level rel="nofollow" affects only that one link.

Google treats these attributes as hints, not commands. Google's guide to qualifying outbound links says so clearly. So adding nofollow does not guarantee that a link is never processed.

For example, you would mark user links in comments or paid placements at the link level. In practice, you would rarely use the page-level version. Ask yourself whether you really need to block every link on the page.

Does robots.txt replace noindex?

No, robots.txt does not replace noindex. It only controls which URLs crawlers may access. Google's robots.txt introduction is clear on this: the file is not a mechanism for keeping a page out of Google.

If other sites link to a blocked URL, Google can still list that address without a description. In other words, a robots.txt block does not remove the page from search. Also, it hides the page content from Google, which helps nobody.

Instead, to truly keep a page out, Google suggests three routes: a noindex tag or header, password protection, or removing the page. Robots.txt is not on that list.

So the real job of robots.txt is to manage server load and steer crawling toward important pages. Our common robots.txt mistakes guide covers frequent slips. If you are starting from scratch, use our robots.txt generator.

How do noindex, nofollow and robots.txt compare side by side?

Seeing all three side by side is the fastest way to settle the noindex vs nofollow question. The table below shows which stage each tool affects and what it is typically used for.

The last row matters most, because combining two tools on one page can reverse the result. The next section explains how.

What happens if you use noindex and robots.txt on the same page?

If you block a page in robots.txt and also add noindex, the noindex does not work. Google cannot crawl the blocked page, so it never sees the tag. Google's robots meta tag documentation puts it this way: when crawling is disallowed, indexing and serving rules are not found and are ignored.

This is the most common conflict we see. Teams usually want "double safety", but the outcome is the opposite, because the page can stay in the index.

The correct order looks like this:

1. Do not block the page in robots.txt, and make sure Googlebot can reach it.
2. Add noindex to the page.
3. Confirm in Search Console that the page has dropped out of the index.
4. Consider a crawl block afterwards, only if you still need one.

In short, Googlebot must be able to read the page for noindex to work. Next to the noindex vs nofollow split, this is the rule to remember most.

When should you use noindex and nofollow together?

You use both when you want a page out of the index and its links ignored. In the robots meta tag you write noindex, nofollow, or the short value none.

This combination is rarer than people assume. On most noindex pages, we still want internal links followed, because links from those pages to key pages are part of your site architecture.

Reasonable cases for using both include, for example:

• Staging pages, although password protection is safer there.
• Auto-generated pages with no value that also link out.
• Temporary campaign pages until the campaign ends.

As a rule, do not add nofollow when noindex is enough. Extra restriction can damage your internal link structure. For the logic behind links, read our internal linking guide.

Which pages should you noindex?

So noindex pages that help users but make no sense in search. First, the aim is to let only pages that answer a search intent into the index. Also, this improves the overall quality picture of your site.

• Thank-you and order confirmation pages.
• Internal site search result pages.
• Thin pages generated by filter and sort combinations.
• Account, cart and checkout steps.
• Duplicate or near-duplicate tag and archive pages.
• Short-lived landing pages built for ads with no organic value.

Parameter URLs are trickier, because noindex is not always the best fix. We cover that topic in our URL parameters guide.

When deciding, ask one question: should this page earn traffic from Google? If the answer is no, noindex is a good candidate.

Which pages should you never noindex?

Never noindex pages that carry traffic, conversions or internal link value. That sounds simple, but it still happens during launches. Moving a noindex setting from a staging site to the live site is a classic disaster.

Be especially careful with these pages, specifically:

• The homepage, service pages and category pages.
• Blog posts that bring organic traffic.
• Key pages in a pagination chain, which is a debated topic and needs its own review.
• Alternate language versions on multilingual sites.

Also, site migrations and redesigns multiply this risk. Our website migration checklist helps you avoid it.

If a page needs to stay indexed, consider a different fix. For duplicate content, a canonical tag is often the better tool.

How does noindex vs nofollow affect link value?

Whether links on a noindex page pass value is not a settled question. Google's robots meta tag documentation makes no detailed promise about it. Relying on a noindex page for link value is therefore not a safe strategy.

On the nofollow side, however, Google uses the link-level attribute as a hint. You say "this link should not carry a signal", but there is no certainty.

In practice, we think of it this way:

• Link to important pages from pages that are indexed.
• Do not use a noindex page as a link warehouse.
• Choose the right rel value for paid or user-generated links.

For link quality and earning outside links, see our backlinks guide. It covers the external side of noindex vs nofollow in more depth.

What is the difference between nofollow, sponsored and ugc?

Also, Google defines three link attribute values: nofollow, sponsored and ugc. Each has a different purpose, and Google treats them as hints.

Google's guide to qualifying outbound links also notes that you can combine values, for example rel="ugc nofollow".

For paid links, sponsored is the safest choice, because it is also the transparent thing to do, not just an SEO detail.

How does X-Robots-Tag work for PDFs and images?

X-Robots-Tag is the HTTP header version of the robots meta tag. You cannot place a meta tag inside a PDF or an image, so this method steps in. According to Google's documentation, any rule you can use in a robots meta tag also works in this header.

For example, you can return X-Robots-Tag: noindex from your server for a price-list PDF that should not appear in search. The same logic applies to image files.

• On Apache, you can add headers by file type.
• On Nginx, you use an add_header line in the location block.
• CMS plugins usually manage only HTML pages, so files need a server setting.

Check that the header is really returned, using browser developer tools or curl. Writing the setting is not enough, so verify it.

How long does noindex take to work?

The effect of noindex depends on when Google recrawls the page, so we cannot give a fixed time. Frequently crawled pages change sooner, but rarely crawled pages take longer.

Instead of waiting, you can speed the process up with these steps:

1. Inspect the page with the URL Inspection tool in Search Console.
2. Test the live URL and confirm that the noindex tag is seen.
3. Request indexing again if needed.
4. Update the lastmod value of the URL in your sitemap.

For the date logic, see our sitemap lastmod guide.

We also suggest removing noindex pages from the sitemap, because a sitemap says "index this" while noindex says "do not". Avoid sending conflicting signals.

How is a canonical tag different from noindex and nofollow?

A canonical tag names the preferred URL for a page, while noindex removes the page from the index. For duplicates, canonical says "this is the main page". Noindex says "do not show this page at all".

For example, if color variants of one product live on separate URLs, canonical makes more sense, because it gathers signals on the main page. Noindex, however, gathers nothing; it only excludes the page.

• Duplicate or very similar page: think canonical first.
• Completely useless page: think noindex.
• Page that should never be opened: think password or removal.

Also, using canonical and noindex on the same page sends conflicting signals. Choose one clear decision per page.

How can a wrong noindex tank traffic, and how do you spot it?

A misplaced noindex pulls pages out of the index and can drop organic traffic overnight. It is hard to notice because the page looks normal in the browser. Instead, the problem hides in the source code or the header.

When you suspect it, then follow these steps:

1. In Search Console, open the Pages report and find the group excluded by a noindex tag.
2. Check whether pages appear there that should not.
3. Inspect their source code and HTTP headers.
4. Find where the tag comes from: a template, a plugin, the server or a CDN rule.
5. Fix it and request indexing again.

To investigate drops in a structured way, start with our Search Console traffic drop guide.

For example, take a 500-page site where a bad template setting adds noindex to 120 pages. That removes nearly a quarter of the site from search. Note that this is only an example calculation.

How do you check noindex status in Search Console?

Search Console shows noindex status in two places: the Pages report and the URL Inspection tool. In the Pages report you see separate reasons, such as "Excluded by noindex tag" and "Blocked by robots.txt".

So do not mix the two up. The first can be your deliberate choice, while the second means Googlebot could not read the page, which can also mean it never saw your noindex.

Not every row is a problem. If a thank-you page has noindex, that is expected. Instead, the problem appears when an important page shows up in this list.

For more on the tool, read our Google Search Console guide. For a quick first look at a page, our SEO checker helps you review the basics.

How do you apply noindex and nofollow in WordPress and other systems?

In WordPress and similar systems, you usually set noindex per page through an SEO plugin. Nofollow is a checkbox when you add a link. That convenience also makes bulk mistakes easy, so be careful.

Also, one global setting can publish noindex across the whole site. WordPress's option to discourage search engines from indexing the site is the best-known example. Always check that this box is off when you launch.

• Add a noindex check to your pre-launch checklist.
• Audit robots tags in templates on a regular schedule.
• Recheck key pages after plugin updates.

In practice, on larger sites it is safer to tie this check to an automated test. For sites managed in code, a small robots test in the deployment pipeline solves the problem cheaply.

Does noindex save crawl budget?

Noindex pages keep getting crawled, so they do not save crawl budget. Googlebot must visit the page to see the tag. Many teams overlook this detail, so plan for it.

So if your goal is to reduce server load or crawl effort, noindex alone is not the answer. Robots.txt may fit better in that case. Keep in mind, however, that robots.txt gives no indexing guarantee.

• To keep a page out of search, use noindex.
• To cut pointless crawling, fix robots.txt or the URL structure.
• For both goals, noindex first, then weigh a crawl block.

Also, if crawling drops on its own, look for the causes separately. Our guide on why Googlebot crawls less gives a detailed framework.

How do you apply noindex vs nofollow on e-commerce sites?

The main risk in e-commerce is that filter and sort combinations create endless URLs. Most of them are thin, so they add nothing to search. Still, one rule for all of them would be wrong.

A popular filter page, such as a color or brand combination, can answer a real search intent. Adding noindex there wastes an opportunity. Cart, checkout and account pages, on the other hand, are natural noindex candidates.

• Filter pages with real demand: keep them indexed and give them unique content.
• Thin combinations: consider canonical or noindex.
• Cart, checkout, account: use noindex.
• Links in user reviews: mark them ugc or nofollow.

Our e-commerce consulting service and the technical SEO tips guide offer a wider frame.

How do you test your noindex and nofollow settings?

A test proves that Google sees the setting, not just that you wrote it. The most reliable method is to check several layers, because a setting can look right in the browser and still differ in the server header.

1. Search the page source for the robots meta tag.
2. Check the X-Robots-Tag value in the HTTP response headers.
3. Look at robots.txt to see whether the same URL is blocked.
4. Use URL Inspection in Search Console to confirm what Google sees.
5. Look for conflicts between the sitemap and noindex.

For bulk checks, use a crawler. For a single page, our SEO checker gives a quick first look.

Then, if you make these tests part of your release process, you sharply reduce the risk of traffic loss from a wrong noindex.

Which tool should you pick for which situation?

The decision flow is simple: define your goal first, then the stage. To keep a page out of search, pick noindex. Reducing crawl load calls for robots.txt. Qualifying one link calls for the rel attribute.

Do not apply one tool to every problem, because each manages a different link in the chain. Knowing the noindex vs nofollow split gives you a clear edge here.

What are the most common noindex vs nofollow mistakes?

These are the six mistakes we meet most often. Most of them come from mixing up the stage each tool works on.

1. Putting noindex on a page that robots.txt blocks.
2. Treating robots.txt as a removal tool.
3. Forgetting the staging noindex at launch.
4. Leaving a noindex page in the sitemap.
5. Applying page-level nofollow to a whole page without need.
6. Using no mark at all on paid links instead of sponsored.

None of these mistakes is complex, but they are easy to miss. Regular technical audits are therefore essential. Our technical SEO tips give a wider view.

Also, AI crawlers follow a similar split between access and use. We explained it in our AI crawlers guide.

Which checklist does our team use?

In technical audits, we review all three tools with one checklist. It is a working method based on field experience, not a guarantee. Because every site is different, you adapt the list.

• Open robots.txt and confirm that key sections are not blocked.
• Check that no URL in the sitemap carries noindex.
• Confirm that noindex pages are not blocked in robots.txt.
• Look for bulk noindex caused by templates or plugins.
• Verify the right rel value on paid and user-generated links.
• Review the exclusion reasons in the Search Console Pages report.

Then these six steps turn the noindex vs nofollow split into practice. If you want our team to run the audit, see our SEO consulting service.

Finally, remember this: the right tool works at the right stage. Specifically, use noindex for indexing, robots.txt for crawling and rel for link quality.