Search Console Links Report: How to Check Backlinks

If you want to know which sites link to yours according to Google itself, the Search Console links report is the place to look. It is free, it covers the property you own, and it lists linking sites, linked pages and anchor text. It is also easy to misread. Google describes it as a sample, it can contain links that have since disappeared, and it does not tell you whether a link carries a nofollow attribute. This guide shows how to use the report for what it does well and how to cover the gaps it leaves.

The explanations follow Google's own Search Console Help page on the links report, plus Google Search Central pages on link qualification and spam policies, all read on 2026-10-28 and listed at the end. The example tables are invented for illustration and contain no real site data. Nothing here promises a ranking result; the report shows what Google has recorded, not what a link is worth.

What the Search Console Links Report Actually Is

Two sections, one property

The report has an external links section and an internal links section. External links are pages on other sites that point to your site, and anything outside the property you are viewing counts as external. Internal links are links from your own pages to other pages on your site. Everything in the report is scoped to the property you selected, so the first habit worth building is to note which property you opened, a whole domain or a single URL prefix, before you read a single number.

Why it matters for link buyers and site owners

Third-party backlink tools crawl the web with their own bots, so their coverage and counts differ from Google's. The Search Console report is the only view that comes from Google's own records for your property, which makes it a useful reference point when you check an agency report or a publisher's delivery note. It does not replace those tools for studying competitors, because it only shows your own property. If you want to analyze other sites, see how to do a competitor backlink analysis; if you want to judge whether a given link is worth having, read how to spot a quality backlink. This article stays on one narrower question: how to read Google's own data about links to your site.

What it will not tell you

The report will not show a competitor's links, will not say whether a link helped your rankings, and will not label a link as good or harmful. A site appearing in the list says nothing about its quality, and a site missing from the list does not prove it has no link to you. Keep these three limits in mind and most misreadings disappear.

Opening the Report and Choosing the Right Property

Domain property or URL prefix

A domain property covers all subdomains and protocols for a domain, while a URL-prefix property covers only addresses that begin with the prefix you entered. If you open a prefix property for a subfolder, you will see links to that subfolder and not the rest of the site, which can make a healthy campaign look empty. For link checks, the broader property is usually the safer choice. Whichever you use, record it in your notes so a colleague can reproduce what you saw.

Permissions when you work for a client

Agencies and consultants often see a client's property with a restricted role. Permission levels can change what a user can do in Search Console, so check the current Help documentation before promising a client a specific export. This article does not make claims about role-specific differences in the links report.

Finding the report

The links report is opened from the navigation in Search Console, and the Help page describes drilling down from the report into external or internal links and then selecting the option for more rows. Menu labels move around over time, so treat the current Help page as the authority on where the entry sits.

Reading the External Links Tables

Top linked pages

This table lists the pages on your site that receive the most links from other sites. Select a row and you see the sites linking to that page. It is the quickest way to check whether a campaign aimed at a particular landing page is showing up. If you bought links to a category page, look for that page here. For advice on preparing the destination before links arrive, see landing page readiness before building backlinks.

If the page you expect is missing, there are three plausible reasons: Google has not found the link yet, the link is not part of the sample the report shows, or the link points to a different address such as an old URL that redirects. The verification section below explains how to tell them apart.

Top linking sites

This table groups links by root domain. According to Google, protocol, subdomain and subdirectory are removed, so a link from a subdomain and a link from the main domain land on the same row, while different top-level domains remain different sites. Selecting a site shows which of your pages it links to most.

This is the table people over-read. A high count means many links from that site, and nothing more. A footer or sidebar link repeated across thousands of pages can produce an impressive number that reflects a template, not editorial interest. Use the table to see who links, and use other evidence to decide whether the link matters.

Top linking text

The anchor text table shows the most common link texts pointing to your site. Google notes that duplicates are exact matches ignoring case, and links without text appear as empty, which usually means an image link. Look at the pattern, not single lines: is the mix mostly brand names, URLs and descriptive phrases, or is one commercial phrase repeated suspiciously often? For how to plan that mix, see anchor text strategy for link profiles.

Linking pages and the target URL column

Drilling into a site or a target page gives you a list of individual linking pages. The table includes a column for the target URL when it differs from the page you selected. Google says a differing target usually reflects a redirect or a canonical choice. In practice, that column tells you where a publisher actually pointed the link, which matters if the agreed destination was a specific page.

Limits You Need to Know Before You Trust the Numbers

It is a sample, not an inventory

Google states that the report is not comprehensive. It shows a sample of links, and some URLs, such as pages that are not indexed or that were deduplicated, may be left out. This single fact should shape how you write up findings. "Not in the report" means "not in the sample I was shown."

Row limits and exports

Tables are limited to 1,000 rows, and a single-table page exports up to 1,000 rows. The main links export offers up to 100,000 rows in two slices: latest links, sorted by discovery date, and more sample links, a sample from the full set known to Google. Values shown as a tilde or a dash become zeros in exported data, so do not read those zeros as measured values. These limits are Google's current documentation and may change, so confirm them on the Help page before building a process around them.

Old links can linger

The report includes links Google has found over time, and the Help page says those links may since have been removed or the page may no longer exist. For anyone who bought a time-limited placement, this matters: a row can remain after the placement has expired. Treat a listed link as evidence that a link existed at some point, and confirm that it is live by opening the page.

No nofollow information

Google states that the report does not specify whether a link is marked nofollow. A paid link correctly marked as sponsored will sit in the list next to an editorial link with no attribute. To see which attribute a link carries, you have to inspect the source of the linking page. If you are unsure what each attribute means, dofollow vs nofollow vs sponsored vs UGC links explains the differences.

Grouping and normalization

Pages are grouped by canonical URL, some anchors and parameters are removed, and duplicate links from the same source to the same target are combined after normalization. So two links from one article to one page can show up as a single entry. When your own count of links does not match the report, this is one of the first things to rule out.

Three Common Jobs and How to Handle Each

Job one: confirm that a purchased link was delivered

Start with what you were promised: the linking URL, the destination page, the anchor text and the attribute. Then work in this order.

1. Open the linking page yourself and find the link. This is your primary evidence.
2. View the page source and note the target address, the visible text and the rel attribute.
3. Check that the page is public and not blocked from crawling, so Google can see the link at all.
4. Look for the linking site in the top linking sites table and the linking page in the drill-down list.
5. If it is not there yet, record the date and check again later rather than concluding it was not delivered.

The Search Console check is a second layer, not the first. Delivery tracking more broadly is covered in backlink package delivery tracking, and footer placements have their own steps in how to check footer backlinks.

Job two: audit your own profile for surprises

Export the data, then sort sites by count and scan the list for names you do not recognize. Unknown does not mean harmful. Directories, scrapers and aggregators link to many sites without anyone's involvement. Google's disavow Help page says that in most cases Google can assess which links to trust without additional guidance. Look at each unfamiliar site briefly: what is it, why would it link to you, and is the context plausible? Only a small subset should need action.

Job three: cross-check an agency report

Agencies often report links using third-party tools that count differently from Google. Compare the agency's list with the Search Console export. Links in both are the easiest to confirm. Links only in the agency's report may simply be outside the sample or not yet discovered. Links only in Search Console may be ones the agency did not build. Treat the differences as questions to ask, not as accusations.

A sample verification log

The table below is fictional and shows a recording format only. It does not describe any real placement.

The last row shows why the report cannot be your only evidence: a listed link was gone when the page was opened.

Paid Links, Attributes and Search Policy

What Google says about qualifying links

Google's guidance on qualifying outbound links says links that are advertisements or paid placements should be marked with the sponsored value, that nofollow is still acceptable, and that sponsored is preferred. User-generated links such as those in comments and forum posts should use the ugc value. Values can be combined in a list.

What Google says about buying and selling links

Google's spam policies treat buying or selling links for ranking purposes as a violation, and they describe exchanging money or goods for links among the examples. The same page says buying and selling is not a violation as long as the links are qualified with a nofollow or sponsored value. The balanced reading is that a properly qualified paid placement is not a policy problem, but it should not be bought in the expectation of passing ranking credit. Reasons to buy a qualified placement are visibility, referral visits and brand exposure. For a longer treatment of the policy, see Google link spam policies and safe link building.

Where this meets the report

Because the report does not display attributes, you cannot use it to confirm that a paid link is qualified correctly. Source inspection is the only check. If a placement you paid for turns out to carry no attribute, raise it with the seller, and for a placement you manage as a publisher, fix it on your side.

Handling Links You Did Not Expect

Triage before action

Ask three questions about each unfamiliar linking site. Is it an actual site with content, or a scraper? Does the link sit in a context that makes sense? Could it be a link you or a partner arranged and forgot? Keep a short note for each decision. Most unfamiliar links need no action.

Removal request first, disavow second

Google's Help page says that if you find unknown or spammy sites you can ask for removal first and, if that fails, disavow them. Disavowing is a tool to use carefully rather than a routine cleanup step, so keep a written reason for every entry. The mechanics are covered in how the backlink disavow process works.

Separating your own links from strangers

Match the report against your own order and delivery records. Links that match an order are yours to manage with the seller. Links that match nothing are the group to review. This split keeps you from disavowing something you paid for or ignoring something you did not.

Using the Internal Links Section

Is the linked page supported from within your site?

The internal links table shows which pages receive the most internal links and, for a selected page, which pages link to it. Google's Help page notes you can use it to check whether one page can be reached from another in a small number of clicks. A page that attracts external links yet receives almost no internal ones is a candidate for improvement. This is a sensible housekeeping step, not a documented ranking rule.

Quick fixes

• List pages with external links but few internal links pointing to them.
• Add descriptive links from closely related content to those pages.
• Ask publishers to update links that point at old, redirected addresses.
• Re-export after a few weeks and compare.

Turning One Check into a Routine

A repeatable export schedule

One export is a snapshot. Two exports taken the same way a few weeks apart tell a story: which links appeared, which vanished, and how fast new ones are discovered. Choose an interval that fits your campaign pace, such as after each delivery batch, and keep the files with dates in the names. The timing of when a link might start to matter is a separate question, discussed in how long backlinks take to work.

Questions to ask at each review

• Did the expected new sites appear?
• Does the anchor text mix still look natural?
• Do the target pages match the plan?
• Did any previously listed link disappear?
• Did an unrelated source suddenly appear in large numbers?

Connecting links to outcomes

Confirming that a link exists is the first step in judging a campaign, not the last. For the indicators worth tracking, read backlink campaign KPIs: which metrics matter.

Frequent Mistakes

Reading absence as proof

Because the report is a sample and has a discovery lag, a missing row cannot prove that nothing was delivered. Go to the page.

Reading presence as proof

Because old links stay in the data, a listed row cannot prove a link is still live. Go to the page.

Ranking sites by raw counts

Counts reflect templates and site structure at least as much as editorial interest. Context and relevance tell you more.

Mixing up properties

Comparing a prefix property with a domain property, or an old export with a new one taken differently, produces differences that are not real. Keep your method constant.

Disavowing on reflex

An unfamiliar domain is not a penalty. Review first, document your reasoning, and act only where there is a clear case.

Frequently Asked Questions

Can I see my competitors' backlinks in Search Console?

No. The links report covers only a property you have access to. To study another site you need a third-party tool, and its figures come from its own crawl, so they will not match Google's records exactly.

Does the report show which links are nofollow?

No. Google's Help page says the report does not specify whether a link is marked nofollow. Check the source code of the linking page to see the rel value.

How many rows can I view and export?

Tables are limited to 1,000 rows, and single-table pages export up to 1,000 rows. The main links export gives up to 100,000 rows split between latest links and a sample of other links. These limits come from Google's current documentation and can change.

A link I paid for is not in the report. Is that a problem?

Not necessarily. The report is a sample and links may take time to be discovered. First confirm the link is live, public and crawlable in the page source. If it is, check again after a few weeks, and contact the seller with your evidence if it is still absent or if the page no longer carries the link.

Are all the links in the report still live?

No. The report contains links Google found over time, and some may have been removed. Open the linking page to confirm.

Should I disavow every unfamiliar site I see?

No. Google notes that in most cases it can assess which links to trust without additional guidance. Review the site first, request removal if there is a real problem, and consider disavowing only if that fails and you can justify the decision.

Is it a problem if I see the same link twice?

Usually not. Google combines duplicate links after normalization, and separate rows generally stand for different source pages or target addresses. A difference in the target URL column can come from a redirect or a canonical URL choice.

Conclusion and Next Step

The Search Console links report gives you Google's view of who links to your property, in three tables and an export. Its value depends on reading it with its limits in mind: it is a sample, it may include removed links, it hides nofollow information and it is grouped by root domain. Pair it with a direct look at the linking page and its source code, repeat the check on a schedule, and write down what you find.

If you are planning a link campaign and want a structured way to place links, you can browse the backlink packages on Akua Adforce. Each listing states its scope, placement type and conditions on the product page. Delivery, indexing by Google or any ranking effect is not guaranteed, so use the verification steps in this guide after each delivery.