Link Reclamation: The Only Link Building Where the Hard Part Is Done

Link reclamation is the practice of recovering links you have already earned but are not currently receiving: brand mentions published without a link, inbound links pointing at pages that now return 404, and links to URLs you moved without redirecting.

Every other tactic in link building has the same expensive first step. You have to convince someone who has never heard of you that your page is worth referencing. That is the part that costs money, takes months, and mostly fails.

Reclamation does not have that step. Somebody already made the decision. A writer typed your name into a paragraph. An editor put your URL in a citation. The judgement call that every outreach campaign is trying to manufacture has already been made, in your favour, and then something mechanical went wrong afterwards.

Which means reclamation is not really a link building tactic. It is maintenance of an asset you already own. That reframe is the whole article, because it changes when you do the work. Tactics get scheduled. Maintenance gets monitored.

Link rot is not an edge case#

The reason this matters more every year is that the web deletes itself at a rate most SEO plans do not account for.

Pew Research Center sampled roughly 90,000 pages per year from the Common Crawl repository for each year from 2013 to 2023, then tried to load them again. They found that 25% of all the pages collected were no longer accessible as of October 2023. For older content the number is much worse: "38% of webpages that existed in 2013 are not available today, compared with 8% of pages that existed in 2023."

Now look at how that 25% decomposes, because this is the part that decides whether reclamation is worth your time.

Stacked bar breakdown of what happened to a representative sample of web pages collected each year from 2013 to 2023 by Pew Research Center, about 90,000 pages per year, rechecked in October 2023. 75 percent were still accessible. 16 percent were individually inaccessible but sat on a root domain that still functions. 9 percent were gone along with the entire root domain. The conclusion: of the pages that died, roughly two in three died on a site that is still running, which means the inbound links pointing at them are recoverable by the site owner without contacting anybody.

Still accessible 75%

Page gone, site alive 16%

Whole domain gone 9%

0% 100% Of the pages that died, roughly two in three died on a site that still works. Those are links a site owner can restore alone, with a redirect and no outreach. Checked October 2023. Sample drawn from the Common Crawl web repository.

Still accessible 75%

Page gone, site alive 16%

Whole domain gone 9%

Two in three dead pages died on a site that still works. Those links come back with a redirect, not an email. Pew Research Center, 2013-2023.

Sixteen points of that twenty-five are pages that vanished from a root domain that still functions. Nine are whole domains that went dark. Roughly two thirds of the rot is the kind where somebody still has server access and could fix it with a config change.

Ahrefs put a much larger number on the same phenomenon from the link side. In their link rot study, "Since January 2013, 66.5% of the links pointing to the 2,062,173 websites we sampled have rotted," rising to "a total of 74.5% of the links in our study are considered lost" once temporary errors and other issues are counted.

Both studies are pointing at the same operational fact, which is that decay is a function of age. The links you earned longest ago, the ones that have had the most time to accrue authority and get cited onward, are also the ones most likely to be pointing at nothing.

Bar chart of how likely a webpage is to be inaccessible depending on the year it was first observed, from Pew Research Center's sample of pages drawn from the Common Crawl repository for each year from 2013 to 2023 and rechecked in October 2023. Pages first seen in 2013 were 38 percent inaccessible. Pooled across all years the figure was 25 percent. Pages first seen in 2023 were 8 percent inaccessible. The conclusion: a page from 2013 is roughly five times likelier to be gone than one from 2023, so the oldest and most valuable part of a backlink profile is also the most decayed part, and decay is a function of age rather than a one-off event.

Pages first seen in 2013 38%

All years pooled 25%

Pages first seen in 2023 8%

0% inaccessible 40% A page from 2013 is about five times likelier to be gone than one from 2023. So your oldest backlinks are your most valuable and your most decayed at once. Measured from the link side, Ahrefs found 66.5% of links to 2,062,173 sampled sites have rotted since Jan 2013.

First seen 2013 38%

All years pooled 25%

First seen 2023 8%

Share inaccessible, 0% to 40% scale. A 2013 page is about five times likelier to be gone than a 2023 one. Your oldest backlinks are your most valuable and your most decayed. Pew Research Center, Oct 2023.

Whatever your own figure is, it is not zero, and nobody on your team is looking at it.

The four kinds of link reclamation are four different jobs#

Most guides present reclamation as one checklist. It is not one job. It is four, and they differ on the only axis that matters operationally: who has to agree before the link works again.

Four types of link reclamation sorted by who must agree before the link works again. Type one, inbound links to your own 404 pages: you are the only person involved, and a server-side permanent redirect fixes it. Type two, links to URLs you moved: also entirely yours, fixed by repairing or shortening the redirect. Type three, unlinked brand mentions: the publisher must agree, so it requires outreach to a named person. Type four, images and data used without attribution: the publisher must agree, and you are asking for credit rather than a favour. The conclusion: two of the four require no outreach at all and should never be batched into a quarterly campaign.

Two of the four are yours alone. They should be monitored, not batched into a campaign.

1. Your own 404 pages They linked. You deleted it. 301 redirect. No outreach. 2. URLs you moved Chains, loops, stray 302s. Repair it. No outreach. 3. Unlinked mentions Your name, no link on it. Ask a named person. 4. Assets, no credit Your chart or data, unattributed. Ask for credit, not a favour.

Two of the four are yours alone.

The first two rows are the ones that get skipped, and they are the ones with no counterparty. That is the strange economics of this tactic: the highest-yield link building action available to most sites is a server configuration change.

Reclaiming links from 404 pages needs no outreach at all#

If a page on your site returns 404 and external sites still link to it, those links exist, they were earned, and they are currently pointing into a hole.

Google's redirect documentation is precise about what fixes this. For a permanent redirect, "Googlebot follows the redirect, and the indexing pipeline uses the redirect as a signal that the redirect target should be canonical." For a temporary one, "Googlebot follows the redirect, but the indexing pipeline doesn't use the redirect as a signal that the redirect target should be canonical."

So the mechanism you want is the permanent one, and Google ranks the implementations by reliability, noting that "a server side redirect has the highest chance of being interpreted correctly by Google." An HTTP 301 or 308 from your server beats a meta refresh, which beats a JavaScript redirect.

Three rules that decide whether this actually works:

• Redirect to the closest equivalent page, not the homepage. A mass redirect of every dead URL to / is the standard mistake. Google treats an irrelevant redirect target as a soft 404, and you get nothing.
• If no equivalent page exists, consider rebuilding the page rather than redirecting it sideways. If people linked to it, it had value.
• Do not leave it on a 302. You will have transferred nothing and told Google the old URL is still the canonical one.

This work is unglamorous, invisible in a monthly deck, and routinely worth more than the campaign running alongside it.

Link reclamation vs broken link building#

These two get confused constantly, including by people selling both. They are mirror images, and the difference is the direction of the arrow.

Comparison of link reclamation and broken link building by the direction of the broken link. In link reclamation, the dead page is yours and the link already points at you, so you are recovering an existing link and you usually need no permission. In broken link building, the dead page belongs to someone else and the link points at a third party, so you are competing for a new link and the publisher must choose you over the alternatives. The conclusion: reclamation is a recovery job with a near-certain yield, while broken link building is an acquisition job with a normal outreach hit rate, and they should not share a workflow.

One is a recovery job with a near-certain yield on the half you control. The other is an acquisition job with a normal outreach hit rate. They should not share a workflow, a target, or a report.

Link reclamation The dead page is yours. The link already points at you. You are recovering, not competing. Often needs no permission. Broken link building The dead page is someone else's. The link points at a third party. You are pitching to replace it. Needs the publisher to say yes.

One is recovery. One is acquisition. They should not share a workflow, a target, or a report.

If you want the other side of that coin, we wrote about broken link building separately. Do not run them as one project.

How to find link reclamation opportunities#

This is a recurring audit, not a research project. The whole sweep takes an afternoon the first time and minutes thereafter.

1. Pull your inbound links to 404 URLs. Any backlink tool will report referring pages pointing at URLs on your domain that do not resolve. Sort by number of referring domains, not number of links. Ten links from one site is one relationship; one link each from ten sites is ten.
2. Crawl your own redirects. Look for chains longer than one hop, loops, and any 302 that was supposed to be temporary three years ago. These are silent and no tool will flag them unprompted.
3. Search for your brand name minus your domain. The unlinked mention query is your brand name as a phrase, excluding your own site. Run it, and run the variants people actually type, including misspellings and your product names.
4. Set up an alert, not a reminder. Mentions are perishable. A mention found the week it publishes gets linked far more often than one found in next quarter's audit, because the writer still remembers writing it.
5. Reverse image search your original charts. If you publish original diagrams or data, they get reused. Most reuse is not malicious and most people will add a credit if asked.

Reclamation type	How you find it	Who has to say yes	Typical effort
Inbound links to 404s	Backlink tool, best pages by links report	Nobody	Minutes per URL
Broken or chained redirects	Site crawl, redirect report	Nobody	Minutes per URL
Unlinked brand mentions	Brand query excluding your domain, plus alerts	The publisher	One email, plus follow-up
Uncredited assets	Reverse image search	The publisher	One email, plus follow-up
Links lost to a competitor rebuild	Manual review of old referring pages	The publisher	Full outreach

The bottom half of that table is where reclamation stops being free. Finding a mention takes seconds. Finding the right person at the publication, working out what they cover, and writing something they will actually reply to is the same research problem as any other outreach, and it is the part that gets dropped when the quarter gets busy. It is the part GenLink is built to keep running.

Unlinked brand mentions are the half that needs a human#

The pitch for an unlinked mention is the easiest one in link building, and people still get it wrong by making it about themselves.

You are writing to someone who already did you a favour. They cited you without being asked. The correct tone is a thank you with a small request attached, not a request with a thank you attached.

What works:

• Reference the specific sentence they wrote. Not the article, the sentence.
• Say what the link adds for their reader, since that is the only reason an editor touches a published page.
• Give them the exact URL so there is nothing to look up.
• Ask once. If they do not reply, the mention is still worth having and you still have the relationship.

What does not work is a templated note about your "valuable resource," and it is worth being clear that you are asking for a link they already almost gave you, not negotiating one into existence. The same rules as any link building outreach apply, with much better odds.

One thing not to do: do not specify the anchor text you want. Suggesting a natural way to reference your page is fine, dictating a keyword string is the pattern Google's spam policies describe. We went through that distinction in detail in anchor text optimization.

Why link reclamation gets scheduled quarterly and should be monitored continuously#

Here is the structural failure, and it follows directly from calling this a tactic.

Tactics live in campaigns. Campaigns have start dates. So reclamation becomes a thing you do in the quarterly backlink audit, which means a link that broke in week two sits broken for eleven weeks, and a mention published on a Tuesday is discovered eighty days later when the writer has moved on to other work.

Two of the four reclamation types have no counterparty at all. There is no reason on earth for those to wait for a quarterly review. A 404 with inbound links is a bug in your site, and you would not batch up bugs for a quarterly bug-fixing campaign.

Treat it as maintenance and the schedule writes itself: a monitor on your 404s with inbound links, an alert on your brand name, and a redirect crawl whenever anybody ships a URL change. The outreach half stays a real workflow with real people in it, and that is fine, because that half is genuinely outreach.

The dividend is quiet and compounds. You are not adding links. You are stopping the ones you already paid for from leaking away, which is the same thing on your dashboard and roughly a tenth of the work. Everything you should know about what a backlink is applies here with one difference: you are not trying to earn these. You already did.

Frequently asked questions#

What is link reclamation in SEO?

Link reclamation is recovering links you already earned but are not currently receiving. That covers inbound links pointing at pages on your site that now return 404, links to URLs you moved without a working redirect, published mentions of your brand that were never linked, and your images or data used without attribution. The defining feature is that someone already decided to reference you.

What is the difference between link reclamation and broken link building?

The direction of the broken link. In link reclamation the dead page is yours and the link already points at you, so you are recovering something you own and often need nobody's permission. In broken link building the dead page belongs to a third party and you are pitching your page as the replacement, which always needs the publisher to agree. Reclamation is recovery, broken link building is acquisition.

How do I reclaim links from 404 pages?

Pull the report of referring pages that link to URLs on your domain that do not resolve, then set up a server-side permanent redirect from each dead URL to the closest equivalent live page. Google's documentation says that for a permanent redirect "the indexing pipeline uses the redirect as a signal that the redirect target should be canonical," and that a server-side redirect has the highest chance of being interpreted correctly. Do not bulk-redirect everything to your homepage; an irrelevant target is treated as a soft 404.

How do I find unlinked brand mentions?

Search your brand name as an exact phrase while excluding your own domain, and repeat it for product names, common misspellings and any former brand name. More importantly, set a standing alert rather than a calendar reminder, because a mention is far more likely to be linked in the week it publishes than three months later.

Is link reclamation worth the time compared to other link building?

It has the best ratio in link building, because the expensive step is already finished. Two of the four reclamation types need no outreach at all, and Pew Research Center found that of a decade of sampled pages, 16% were inaccessible on a root domain that still functions versus 9% where the whole domain was gone, meaning most page-level decay sits on sites whose owners could fix it themselves.

How often should I run link reclamation?

Continuously for the half you control, and as a normal outreach workflow for the half you do not. Inbound links to 404 pages and broken internal redirects are defects on your own site and should be monitored, not batched into a quarterly audit. Brand mention alerts should run in real time for the same reason.

---

Set the redirects yourself. The half that needs a person is the half that gets dropped. Unlinked mentions and uncredited assets are the two reclamation types with a counterparty, and they come down to one repeated job: find the right human at the publication, work out what they actually cover, and write something worth replying to. That is exactly the loop GenLink runs. It is a backlink agent that finds the relevant sites, finds the person, writes a pitch specific to what they publish, sends it from a warmed mailbox, follows up, and triages the replies so you only see the ones that matter. The config half of reclamation is a Tuesday afternoon. The outreach half is a standing workflow, and a standing workflow is a thing you can hand over.

Start with GenLink free.