Resource Page Link Building: Finding Them Is Free, Qualifying Them Is the Job
A resource page is the one link type where the publisher writes their acceptance criteria down in public. Almost nobody reads them before pitching.
Every guide to resource page link building opens the same way. Here are the search operators. Here are four hundred pages. Here is a template. It is the beginner tactic, the one you run first because the pages are easy to find and the ask is easy to make.
Two of those claims are true. Finding resource pages is nearly free, and the ask genuinely is simple. The conclusion drawn from them is backwards.
A resource page is a curated list of links that somebody maintains for an audience they have in mind. What separates it from every other link target is not that it is easy. It is that the publisher has written their acceptance criteria down, in public, on the page you are about to email them about.
Nobody reads them. The industry has decided this is a discovery problem, so it spends its budget on discovery, which is the one part of the job that costs nothing and produces no advantage. Everything that decides whether you get the link happens in the step that gets twenty seconds.
Resource page link building is a qualification problem, not a discovery problem#
The discovery half takes about twenty minutes and you already know how to do it.
Run "your topic" inurl:resources, "your topic" intitle:"resources", "your topic" inurl:links, "useful links" "your topic", "recommended tools" "your topic", and the same set again restricted with site:.edu or site:.org. Pull the results into a sheet. You will have somewhere between two hundred and a thousand URLs before your coffee goes cold.
Here is the problem with treating that as an asset: everyone in your category runs the same operators against the same index and gets substantially the same list. The output of discovery is not proprietary. It is a shared public artifact that your two closest competitors also have open in a tab.
Your advantage cannot live in a step that a search operator performs identically for everyone. It lives in what you do between the export and the first email, and that is the step most teams delete in order to hit a send target.
The usual ratio is inverted. Most campaigns put the overwhelming share of their effort into list size and the remainder into fit, then wonder why the reply rate is a rounding error. Resource page link building rewards the reverse allocation more than any other tactic, because it is the only one where the qualifying information is sitting in public on the target page, waiting to be read, in the publisher's own words.
A resource page publishes its acceptance criteria, and they are narrower than "useful"#
This is not a metaphor. On a large class of the highest-value resource pages, the criteria are a formal document with a URL.
The National Institute of Neurological Disorders and Stroke publishes the rules it applies before linking out. Eight of them. Among them: the site must have "accurate, medically reviewed, science-based information for people with lived experience and members of the public available for free", it must be "managed by a non-profit or not-for-profit entity that is not advancing a commercial product or service", and it must be "designed to reach a national audience and is not providing only information specific to a city or state."
Read that as a link builder. Most commercial software is eliminated by the non-profit criterion before anybody writes an email. No subject line recovers from that.
The Department of Education is blunter still. Its published scope for the links it lists says links to outside organisations "will be provided only if the information is judged to be clearly central, essential, and indispensable to the implementation of the mission of ED and the responsible ED office", and reserves the right to remove a link "if it is determined that a group's website is primarily commercial in nature (i.e., advertising a product, charging a usage fee, soliciting user information, etc.)".
Clearly central, essential, and indispensable. That is the bar, in writing, on a page whose links every SEO in education wants.
Universities run the same convention, often with the word "criteria" in the heading. Tarleton State's web site link policy states that "Every external website link request must undergo review to determine its relevance and appropriateness to the purpose of Tarleton", and then tells you exactly how to submit: "Each link request should address how the site meets the two acceptance criteria outlined above." UT Southwestern's linking policy uses almost identical language, down to the same sentence asking submitters to address the criteria.
They are telling you what the email has to argue. Most outreach does not argue it.
Not every resource page is a .gov or a .edu with a legal team. On the informal ones the criteria are unwritten, but they are just as real, and the next section is how you read them off the page in four minutes.
And when a page has a "suggest a resource" form, the form fields are the criteria, in order. If the form asks for an audience, a licence and an accessibility statement, those three things are the review. A pitch that ignores them is not a weak pitch, it is an incomplete submission.
This is the structural advantage of the tactic, and it is exactly what volume destroys. You are not guessing what a stranger wants. You are reading a brief. The brief takes four minutes per page, and it is the entire difference between this and cold outreach.
Link rot is the opening you are looking for, and it is measurable#
There is a second thing sitting in public on these pages, and unlike the criteria it is quantifiable.
Resource pages are old by construction and long by construction. Both of those are the variables that drive link decay, which means the page you are looking at very probably contains dead links right now.
The best measurement of this is not from an SEO vendor. Pew Research Center analysed just under a million webpages sampled from Common Crawl snapshots and found that "A quarter of all webpages that existed at one point between 2013 and 2023 are no longer accessible", with the rate scaling hard by vintage: "38% of webpages that existed in 2013 are not available today, compared with 8% of pages that existed in 2023."
They then measured this at the level of individual links. Across more than 14 million links on 500,000 pages from 2,063 news sites, "5% of all links on news site pages are no longer accessible". Across 42 million links on around 500,000 government pages, "6% point to pages that are no longer accessible". On Wikipedia, across just over a million reference links, "11% of all references linked on Wikipedia are no longer accessible".
A per-link failure rate is not the number you need, though. What you need to know before you open a resource page is the chance that this page, with its forty entries, is already broken somewhere.
The arithmetic is the boring part and the conclusion is not. At the 5% per-link rate Pew measured on news pages, a forty-entry resource page has an 87% chance of already carrying at least one dead link. At the 11% rate they measured on Wikipedia references, it is over 99%.
Two honest caveats belong with that curve. It assumes links fail independently, which they do not quite: one dead host can take out several entries at once, and a page that is actively maintained will run below the population rate. And Pew's rates were measured on news, government and Wikipedia pages, not on a sample of SEO resource pages, so I am transporting a rate from one population to another.
Both caveats push in the same direction, and neither changes the shape. The curve is steep early and flat by twenty entries. Which means the question is never whether a resource page has a dead link. It is which one.
That is your opening. Not "I noticed your excellent resource page". A dead link on the page, named, with a working replacement offered. This is broken link building run inside resource page link building, and it works here better than anywhere else because the maintainer's incentive is unusually clean: their credibility is the page, and you are offering to repair it.
It also fixes the thing that makes resource page outreach fail. The default pitch asks the maintainer to do unpaid work for a stranger's benefit. The link-rot pitch offers to do unpaid work for theirs.
How to qualify a resource page: five checks, in the order that kills fastest#
Qualification is cheap if you run the checks in the right sequence. The right sequence is whatever eliminates the most pages per second of your attention. Most people run these backwards, starting with the authority score, which is the slowest and least decisive of the five.
1. Is anyone still home? Look for a last-updated date, a copyright year, or the vintage of the newest entry. Then spot-check three links. A page whose most recent addition is a product that shut down in 2019 has no maintainer, and a page with no maintainer cannot add you no matter how good the pitch is. This check takes thirty seconds and removes a large share of any operator-built list.
2. Who is the named audience? Read the first paragraph. If the page says it serves undergraduates, or members of a professional body, or residents of a county, that is the filter, and no amount of rewriting the email moves you inside it.
3. What is already listed, and what does that imply? Count the entries by type and by publisher. Thirty free archives and zero commercial products is not an oversight, it is the standard. Every entry sharing an institutional affiliation is a provenance rule nobody wrote down. A page that mixes tools, articles and datasets has room for more kinds of thing.
4. Can the maintainer actually edit the page? A departmental page maintained by a named librarian is editable. A page inside a CMS nobody has credentials for any more, on a site that has since been redesigned around it, is not. Look for a contact route that belongs to a role rather than a person who has moved on.
5. Has it linked to anyone comparable to you? This is the highest-value check and it comes last because it costs the most. A page that has already listed two or three companies in your category is applying a standard rather than doing a favour, and it is the single strongest signal that the criterion is one you can satisfy. Multi-page overlap is also the quickest way to find these in the first place, which is where competitor backlink analysis earns its keep: the resource pages that appear in several rivals' profiles at once are the ones running a published standard.
The five checks take about four minutes a page. Run them on four hundred pages and you are looking at a month of work, which is exactly why nobody does it and why the results are so poor.
Run them on the first eighty pages, in kill order, and most die on check one or check two in under a minute. What comes out the other end is usually somewhere between ten and thirty pages, and that list is worth more than the four hundred you started with, because every page on it has a criterion you have read and can meet.
The list you should be proud of is the one that got smaller.
That month of work is the honest cost of doing this properly by hand, and it is the reason most teams do not. GenLink exists to absorb the part of it that is stamina rather than judgement.
Where "resource page link building is dead" came from#
The tactic has a bad reputation in some quarters, and the reputation has a specific history worth knowing, because it explains what actually broke.
Resource page link building was industrialised somewhere around 2012, in the window when Google's first Penguin updates were reshaping what link building could safely look like. It was pitched as the safe, white-hat alternative to paid links and blog networks: the pages exist to link out, so asking to be listed is not manipulation, it is participation.
That was true, and then it was scaled. Operators lists went into mail merges, submission pages filled with identical two-line requests, and a genre of low-quality "resources" pages appeared whose only purpose was to accept those requests. Maintainers of real pages started ignoring the inbox. The tactic's reputation was earned by what volume did to it, not by anything wrong with the page type.
Google's own guidance is both narrower and more pointed than the folklore around it. Its spam policies define the category plainly: "Link spam is the practice of creating links to or from a site primarily for the purpose of manipulating search rankings." The examples are mostly about payment and automation, and two of them land directly on this tactic. One is "Using automated programs or services to create links to your site". The other is "Low-quality directory or bookmark site links".
Read those two together and the line is clear. Asking a librarian to list a genuinely relevant tool on a page whose stated purpose is to list genuinely relevant tools is not on that list. Running an unread export through a mail merge until something sticks, and collecting links from pages that exist only to accept such requests, is a fair description of both of them.
The thing Google names as spam is the volume workflow, not the page type. Qualification is not just what makes the tactic work. It is what keeps it on the right side of the line.
So "resource page link building is dead" is a claim about a workflow, not about a page type. The pages are still there, they still get maintained, and they still link out. What died is the version where the list is the strategy.
Which brings up the objection worth taking seriously, because it comes from a good source.
Ahrefs' own guide to the tactic gives three reasons resource pages are worth your time, in three words each: "It works." "It's scalable." "It's relatively easy." It then hands you a spreadsheet that turns ten keywords into "hundreds of search operators ready to go", and promises that following the instructions "should be enough to land at least a handful of resource page links in most niches."
That is right about the outcome and wrong about the mechanism, and the disagreement matters. A handful of links is exactly the return you would expect from running an unqualified list against pages where a minority happen to fit, and it is the reason the tactic gets filed under beginner. The same effort aimed at ten pages whose criteria you have read is not competing for that handful at all. It is reaching maintainers who stopped opening generic requests years ago.
Scalable discovery is genuinely useful. It is the input, not the strategy, and treating it as the strategy is what caps the tactic at a handful.
Resource page outreach when the page has already told you what it wants#
Once the list is qualified, the email almost writes itself, because the page supplied the argument. Name the dead link. Name the criterion the page publishes. State how the thing you are offering meets it, in their words, not yours. Stop.
That is a better email than almost anything a copywriter produces from a template, and it is a direct product of the qualification work rather than of the writing.
But do not overrate the email either, because there is good evidence about what actually moves outreach, and it is not the prose. Backlinko and Pitchbox analysed 12 million outreach emails and found that "Only 8.5% of outreach emails receive a response". Against that baseline, they measured that "Personalized subject lines boost response rate by 30.5%", that "The response rate of messages sent to several contacts is 93% higher than messages sent to a single person", that "Emailing the same contact multiple times leads to 2x more responses", and that "Email sequences with multiple attempts and multiple contacts boost response rates by 160%."
Put those multipliers on the baseline and the ordering is unambiguous. The lever everybody grinds on, the words in the email, is the smallest one on the chart. The levers that move it are structural: reaching the right number of people at the site, and continuing to exist in their inbox after the first attempt.
Those two things are also the two things a human link builder stops doing at about page thirty. Not because they are hard, but because they are administrative, and administration is what gets dropped when a list is long.
That is the shape of the work GenLink is built for. You do the part that requires judgement: reading the page, reading the criteria, deciding whether you can honestly meet them. The agent does the part that requires stamina: finding who at that institution actually maintains the page, writing the pitch from what the page says, sending it from a warmed mailbox, following up on the schedule the data says to follow up on, and triaging the replies so the only ones that reach you are the ones worth answering.
Give GenLink the page you want listed and it will go find the resource pages that would take it.
Two limits on that evidence, since they matter. The Backlinko study was last updated in 2019 and covers outreach in general rather than resource pages specifically, so the absolute 8.5% should not be treated as a forecast for your campaign. And a response is not a link. What the multipliers support is the ranking of the levers, not a promise about any particular number.
Read against this post's argument, the study also cuts slightly the other way: if personalisation is only worth about 30%, why spend four minutes a page reading criteria? Because the study measured personalisation of the message, which is a copy variable, and qualification is a targeting variable. It does not change how persuasive you are to a given maintainer. It changes which maintainers are in the list at all, and the policies quoted above are the reason that distinction is not academic. There is no percentage improvement in the pitch that gets a commercial product onto a page that excludes commercial products.
What a qualified resource page programme actually looks like#
Put the pieces in order and the whole thing is unrecognisable next to the standard version.
Spend twenty minutes on discovery, because that is all it is worth. Spend the next several hours on qualification, running the five checks in kill order until you have between ten and thirty pages whose published or implied criteria you have actually read. Check every one of those pages for dead links first, because the arithmetic says most of them have one, and that is your reason to be in the inbox at all. Then write pitches that argue against the stated criteria, send them to more than one contact, and follow up.
That is a week of work producing a list of thirty. The standard version produces a list of four hundred in twenty minutes. The list of thirty wins, and it is not close, because every page on it has already told you what it takes and you have already checked that you have it.
The rest of the cluster fits around this. Link building prospecting is the general case of the qualification argument made here. Link building outreach is what happens after the list is right. Anchor text optimization matters less than people think on these pages, since the maintainer usually writes the anchor themselves. And the whole tactic sits inside the wider model in the link building pillar, where resource pages are the clearest example of a structural link: a page that exists to link out, applying a standard, repeatedly, to anyone who meets it.
Frequently asked questions#
What is resource page link building?
It is the practice of getting your page listed on pages that exist to link out: library research guides, association member resources, "useful links" pages, curated directories maintained by an institution or a community. You ask to be added. What separates it from other outreach is that the publisher has usually stated, on the page or in a linked policy, what they will and will not list. The work is reading that and deciding honestly whether you qualify.
How do I find resource pages?
Search operators, and it takes about twenty minutes. Combine your topic with inurl:resources, intitle:resources, inurl:links, "useful links", "recommended tools" and "helpful resources", then repeat each one restricted to site:.edu, site:.org and site:.gov. Add your competitors' names as a modifier to find the pages already listing companies like yours. None of this is a competitive advantage, because everyone else can run the same searches against the same index in the same twenty minutes.
Does resource page link building still work in 2026?
Yes, with a caveat about what "it" means. Institutions still maintain link pages, still add to them, and still respond to good submissions. What stopped working is the mail-merged version, because maintainers have been receiving identical requests for over a decade. The tactic is healthier than its reputation and much slower than its reputation suggests.
Is asking for a link on a resource page against Google's guidelines?
Not on its own. Google's spam policies target links created primarily to manipulate rankings, with buying, automation and low-quality directory listings as the named examples. A genuine request to be listed on a page that exists to list things like yours is not that. Where people drift over the line is automating the request at volume, or pursuing pages that exist only to accept such requests, which is closer to the directory example than to anything defensible.
How long does it take to get a link from a resource page?
Longer than any other tactic, and the variance is enormous. Institutional pages often route requests through a committee, a communications office or legal review, and some publish that fact openly. A university may take a term. A hobbyist maintainer may reply in an hour. Plan in quarters, not weeks.
Should I build my own resource page to attract links?
It is a reasonable asset but a different tactic, and people conflate them. A page you publish earns links by being the best list of something, which is a content and promotion problem. Being listed on somebody else's is a qualification problem. The useful overlap is that maintaining one teaches you, quickly, how bad most submissions are.
What should I track on a resource page campaign?
Not send volume, which is the metric that produced the problem. Track the size of your qualified list and the reason each page was rejected, because that record tells you whether your offering is genuinely ineligible for a category of page or whether you are just describing it badly. Track replies rather than links as your leading indicator, since the lag on placement is so long, and track dead links found and reported, which is the part of the work that generates goodwill whether or not you get listed.
A resource page is the one place in link building where the other party has written down what they want. Almost nobody reads it, which is why the tactic has a reputation for low yield and why the yield is low for the people who gave it that reputation. Read the criteria, check the page for the dead link that is probably already there, and pitch only where you honestly qualify. That leaves you with a small list and an unusually high hit rate.
The small list is still a lot of work, and it is the same work every time: find who maintains each page, read what they published, write a pitch that is true about it, send it from a mailbox that arrives, follow up on schedule, and triage what comes back. That loop is what GenLink runs. You decide which pages you belong on. The agent goes and makes the case.
Read the criteria yourself. Let the agent do the hundred emails that follow.
