Reciprocal linking
What Are Reciprocal Links, and Are They Bad for SEO?
A reciprocal link is two websites linking to each other. That is the whole definition, and it is also the reason this question is harder than it looks: direction is close to the least informative thing you can know about a link. Two integration partners referencing each other and two hundred unrelated sites trading placements produce the identical shape in a backlink report.
So no, reciprocal links are not automatically bad for SEO. What matters is why each link exists, how many of them you have, and whether anyone benefits from following one. This page is about the reciprocal link itself. For where Google's spam policies draw the line on trading at scale, read are link exchanges against Google's guidelines.
- 16 min read
- Published August 18, 2026
- Updated August 18, 2026
- The LinkSwapy team
- Sourced from Google Search Central
The short answer
- A reciprocal link is two sites linking to each other. Nothing about the direction tells you whether either link is any good.
- Not automatically bad. Google's policy names excessive exchanges, not reciprocity.
- A relevant reciprocal link can send real referral traffic. A traded one usually sends none, which is its own tell.
- There is no safe percentage. Nobody has published one and nobody can.
- Most teams cannot say what their reciprocal footprint is, because no standard report cross-references outbound links against referring domains.
Contents
What this covers
Start with the definition if the vocabulary is the problem. Skip to the footprint section if you already know the theory and want to find out what you actually have.
Definition
What a reciprocal link is
A reciprocal link is a pair of links between two websites, where each one points at the other. Site A links to site B, and site B links back to site A. The term describes a shape in the link graph and nothing more. It carries no information about who asked, whether anyone asked at all, or whether either link deserves to be there.
01
Reciprocal backlink
The same thing, named from the receiving side. People use it when they mean the inbound half of the pair, which is where half the confusion in this topic comes from.
02
Two-way link, mutual backlink
Older phrasing for the same shape, still common in outreach emails. If somebody offers you a two-way link, they are proposing a trade and using softer words for it.
03
Link exchange
An arrangement, not a shape. Every exchange produces reciprocal links. Plenty of reciprocal links come from nobody agreeing to anything, which is why the two terms are not interchangeable.
04
Cross-linking
Usually means the same trade across a group of sites rather than a pair. When it appears in Google's policy it is attached to partner pages that exist for the purpose.
The core problem
Why direction tells you almost nothing
Here is the thing that makes reciprocal linking such a durable argument in SEO circles. Reciprocity is trivially easy to measure and almost useless on its own, so everybody can see it and nobody agrees on what it means.
Reciprocity is the easiest thing to measure and one of the least useful things to know.
Picture two rows in a backlink export. Row one: a monitoring platform links to a security company from its integrations page, and the security company links back from its partners list. Row two: a roofing site links to a crypto blog from a resources page, and the crypto blog links back from an identical page. Both rows are reciprocal. Both would flag in any tool that checks for mutual links. One is a product relationship, the other is a trade between strangers, and the reciprocity flag cannot tell you which is which.
Everything that actually separates those two rows sits somewhere else: topical relevance, what kind of page each link lives on, what the anchor says, how many other links share the same pattern, and whether either party would have linked without the other. Direction is the one variable that is easy to check and the one that carries the least signal.
The vocabulary does not help much here either, and it is worth getting it out of the way once: link swap and link exchange describe the arrangement, reciprocal link describes the shape it leaves behind. Which is why we would not build a review process around finding reciprocal links. Find them, yes, because you should know what you have. But the reciprocity is the starting point of the assessment, never the conclusion. Finding them is a different job from deciding what they mean, and it is the second one that tells you something.
Three cases
Are reciprocal links bad for SEO?
Not automatically, and lumping these three together is how teams end up removing links they should have kept. The link graph shape is identical in all three. Everything else differs.
Nobody arranged anything
Two sites end up pointing at each other through separate decisions made at separate times. Neither party knew the other link would appear. Ask either editor why the link is there and you get an answer about a reader, not about a deal.
You cite a company's research. Months later they reference your analysis of their data, because it is the best thing anyone wrote about it.
Arranged, but the links belong
Somebody did raise it. A partnership got discussed and both sides agreed to reference each other. That is a negotiated link, and it is fine as long as each placement would survive being read by a stranger who knows nothing about the arrangement.
Two companies ship an integration and each documents it. The conversation started the links. The product justifies them.
The trade is the whole point
One dofollow link for one dofollow link, repeated across a list of domains chosen by their metrics. Nothing about the content on either side changed as a result. This is where the pattern starts producing the footprint that gets it discounted.
The email opening with a quick collaboration idea that includes a spreadsheet, specifies the anchor, and never mentions anything either site has published.
Side by side
Natural reciprocal links versus manipulative ones
The columns below are what a spam system can actually measure, and they are the same signals behind the wider link spam category. Read down the right side and notice how little of it concerns reciprocity: almost every signal here would still be visible if the links pointed in one direction only.
| Factor | Natural reciprocal link | Manipulative reciprocal link |
|---|---|---|
| The two sites | Same subject, overlapping audience | Unrelated, connected only by the trade |
| Who it serves | A reader who wants the other page | A ranking, and only a ranking |
| How it was decided | Editorially, on the merits | Required as the price of the inbound link |
| Frequency | Occasional, tied to real relationships | Systematic, tracked against a target |
| Anchor text | Brand, URL, or whatever the sentence needed | Commercial keyword, repeated across partners |
| The relationship | A real business connection you could describe | A backlink arrangement and nothing else |
| The destination | A page worth landing on | A page nobody would visit twice |
| The counterfactual | Would exist with no SEO benefit at all | Would vanish the day links stopped counting |
The last row is the one to keep. Every other line in this table is a proxy for it, and when two of the rows disagree, that is the one that settles the argument.
Concrete cases
Two worked examples
Principles are easy to nod along to and hard to apply to the email sitting in your inbox, which is the daily reality of full time link building. These are the two ends of the range, described the way they actually turn up.
Reasonable reciprocity
A monitoring platform and a security company have shipped an integration together.
- 01
The security company publishes its tooling page
It lists the monitoring platforms its customers connect to, with a paragraph on each about what the integration does. The link sits in that paragraph.
- 02
The monitoring platform documents the connector
Setup instructions, screenshots, and a link back to the security company because a user configuring it needs both halves of the documentation.
- 03
Nobody negotiated the anchors
Each side used the other's product name, because that is what the sentence needed. Neither page would be improved by removing the link.
The version to decline
An agency emails several hundred sites offering to swap resource page placements.
- 01
The list has no subject in common
Travel, crypto, pets, construction, fashion. The only thing connecting these domains is that they all replied to the same email.
- 02
The pages are built for the purpose
Each resources page is a list of outbound links with a heading above it and no copy. Two dozen entries, none of them explained.
- 03
The anchor comes pre-written
The email specifies the exact phrase, which happens to be a commercial keyword the sender is trying to rank for. At that point you are not discussing content.
Most real decisions land between these two, and that is fine. The point of holding both in your head is that you can tell which end a given offer is closer to within about thirty seconds of opening the page.
The policy
What Google says about reciprocal links
Google's spam policies do not mention reciprocal links as a category. What the link spam section names is excessive link exchanges, the repeated link to me and I will link to you arrangement, and partner pages that exist for the sake of cross-linking.
Read that carefully, because it is a description of behavior rather than of shape. Excessive, not any. Exclusively, not partly. Nothing in it says that two sites pointing at each other is a violation, and if it did, every association directory and integration doc on the web would be caught in it.
We have written the long version of this elsewhere, including the exact policy wording, how devaluation differs from a manual action, and a scoring rubric for a single opportunity. If your question is really about the policy rather than about reciprocity, Google's rules on link exchanges is the page you want.
One honest limit, since a lot of pages on this topic skip it. Nobody outside Google knows where the internal thresholds sit, and we do not either. Everything here is pattern reasoning against published policy plus judgment from watching link profiles, and saying so is more useful than a confident number we cannot source.
Consequences
Do reciprocal links hurt your rankings?
Two different things can happen and only one of them is a penalty, which is why this question produces so many contradictory answers.
The usual cost of bad exchanges is not a penalty. It is a quarter of link building that quietly counted for nothing.
The common outcome is devaluation. Google's spam systems, SpamBrain among them, are used to neutralize the credit that unnatural links pass. Google described exactly this when it announced the December 2022 link spam update: links identified as unnatural stop counting, and sites that relied on them can see rankings fall. There is no notification and nothing appears in Search Console. The links keep existing and stop doing anything at all.
The rarer outcome is a manual action for unnatural links to your site. That one does show up in Search Console with a description, involves a human reviewer, and needs a reconsideration request after cleanup. It is not what happens to a business with a dozen partner links.
So the realistic downside for most sites is not a penalty. It is spending months and a budget on links that were never going to count, then wondering why the rankings did not move. That is a quieter failure and a more expensive one, because nothing tells you it happened.
The other side
Can reciprocal links help SEO?
Yes, and the SEO answer is the less interesting half of it.
A reciprocal link between two genuinely related sites behaves like any other relevant link. It can be crawled, it can pass signals, and it puts your page in a context that makes sense. Nothing about the return link cancels that out.
The part worth more attention is referral traffic, because it is measurable, it doubles as a quality test, and it can be priced. A link from an integration partner's setup docs sends people who are mid-task and already using something adjacent to your product. A link from a partners wall on an unrelated domain sends nobody, ever. Check your analytics after a month: if a reciprocal link has produced zero sessions and the page it sits on gets traffic, you have learned something about where the link is buried.
None of which is an argument for scaling exchanges. It is an argument for the specific case where the link would have been worth having with no backlink attached, which is the same test as everywhere else on this page.
- Referral traffic from people already doing an adjacent task
- Discovery, for readers and crawlers, of pages that belong together
- Topical context around the linked page
- A partnership that produces more than one link over time
Mechanics
Do reciprocal links pass PageRank?
They can. Nothing about reciprocity strips a link of the ability to pass signals, and there is no documented mechanism where two links cancel each other out into zero.
What you cannot assume is the version most people quietly believe: they linked to me, I linked back, therefore we each gained the same amount and we are square. That is not how any of it works. Google weighs a link on the relevance of the linking page, where the link sits, the quality of the page it sits on, the anchor, how many other links share that page, and the pattern the link belongs to. Two links between the same two domains can be worth very different amounts, or nothing.
The practical consequence is that a reciprocal link has no fixed value you can plan around. If somebody is proposing a trade on the basis that both sides gain equally, they are describing an accounting entry, not a search engine.
Detection
Can Google detect reciprocal links?
Yes, and it is not close. Finding every pair where A links to B and B links back to A is a routine query against a link graph, and Google has the graph. Your own side of it is a matter of recording what each partner agreed to. Any strategy whose safety depended on that being hard was never safe.
The better question is whether Google ignores them, and there the answer is no, not as a category. It could not, because reciprocal relationships occur naturally everywhere. Treating them all as spam would break the results.
So reciprocity gets combined with the things that actually carry signal: relevance, placement, anchor distribution, timing, site quality, and whether the same group of participants keeps appearing together. A link between two obviously related businesses does not need any special handling. A cluster of unrelated domains that all link to each other does not need much analysis either.
Which makes hiding the pattern a poor use of anyone's time. Stop optimizing for whether the reciprocity can be seen, because it can. Optimize for whether the links hold up once it has been.
The threshold question
How many reciprocal links are too many?
There is no published number. Not five percent, not ten, not thirty, and every article quoting one either guessed or repeated somebody else's guess with more confidence attached. We are not going to invent one here to seem more helpful.
Real link profiles are lumpy. Quotas are smooth. The smoothness is what looks wrong.
The number is also the wrong variable. Ten reciprocal links across ten real industry relationships is a safer object than five obviously manufactured ones, and any ratio that treats those as comparable is measuring the thing that does not matter.
What is worth looking at is shape. Real link acquisition is lumpy: a piece of research picks up thirty links in a fortnight and nothing for a year, a tool page collects one a month from people who found it. Manufactured acquisition is smooth, because a quota is smooth. If your reciprocal links arrived at a steady rate, in pairs, from unconnected domains, the regularity is the tell rather than the count.
And if you are asking because somebody set you a monthly link target that can only be hit through trades, the target is the problem. The full argument on where excessive begins covers the policy side of that.
Comparison
One-way backlinks versus reciprocal links
A one-way link goes in one direction and never comes back. It is generally the stronger signal, and the reason is worth understanding rather than repeating: it is harder to get, so it is better evidence that somebody chose to give it.
| Property | One-way backlink | Reciprocal link |
|---|---|---|
| What it signals | An independent decision to cite you | A decision that may or may not have been independent |
| How hard it is to get | Hard, which is what gives it weight | Easy, if you are willing to trade |
| Typical relationship | Often none, which is the point | A partnership, or an arrangement |
| Failure mode | Scraped directories and paid placements dressed as editorial | Trades at scale with thin relevance |
| Referral traffic | Varies with the page it sits on | Often higher, because partners share an audience |
| Durability | Survives as long as the page does | Dies when either side cleans house |
Where this gets misused: a one-way link from a scraped directory is worth less than a reciprocal link from an integration partner whose docs send you customers. On average one-way wins, and average is doing a lot of work in that sentence. Direction is a weak proxy for quality, not a substitute for judging it.
The workaround
ABC and three way exchanges
The usual response to everything above is to route around it. You link to site B, and site C links back to you, so no two participants point at each other and the pair a reciprocity check looks for never exists.
It is worth knowing one thing about that move before you consider it, and it is not the thing people expect. An ABC link is invisible to your own audit. The intersection method further down this page finds reciprocal links precisely because you linked to the site that linked to you. Break that and your own exposure stops being measurable, six months before anyone else has an opinion about it.
The full comparison, including the three shapes these deals actually arrive in and what the third hop costs you, is its own page: ABC link exchange versus reciprocal link exchange.
Reading an offer
Should you exchange links with this site?
Most of the time you can answer this from the email and one look at the page, plus a mechanical check on the domain. These are the signals worth weighting, in both directions.
Signs it is a real relationship
- You would link to them anyway, without anything in return
- Your audiences overlap in a way you could describe out loud
- There is an existing business connection: a product, a project, an event
- They left the anchor text up to you
- The link would sit inside content rather than on a list
- The destination is a page a reader would be glad to land on
- They are relaxed about whether you link back at all
- You would be comfortable explaining the link to a client
Signs to decline
- They specify exact match commercial anchor text
- The inbound link is conditional on yours going live first
- The offer covers hundreds of sites, or arrives with a domain list attached
- Their niche has nothing to do with yours
- The pitch leads with DR or DA and mentions nothing they published
- Placement is handled automatically by a platform
- The page holding your link exists only to hold links
- Any guarantee about rankings appears in writing
If several of the right hand column are true at once, you can reply in one line and move on. The awkwardness of declining is smaller than the awkwardness of explaining the link later.
Practice
Reciprocal linking best practices
For the cases where two sites do have a real reason to reference each other, recorded in something like the free exchange tracker. None of this is about staying under a threshold, because there is no threshold. It is about the links being defensible on their own terms.
- 1
Link because the destination earns it
The whole guide compresses into this. If you would put the link there with no backlink coming your way, put it there. If you would not, the reciprocity is the only reason it exists and you already know what that makes it.
- 2
Let the anchor be whatever the sentence needed
Brand names, the page title, a bare URL, the word here. Real citations are messy. If a partner sends you the exact wording they want, especially wording with a commercial keyword and a city in it, that request tells you more about the arrangement than anything else in the email.
- 3
Put the link in the content, not on a list
A partners page is not automatically a problem, but a page whose only content is outbound links is the exact thing Google's policy names. If you keep a partners page, give it copy a reader would actually want, or fold the links into the pages where the relationships are described.
- 4
Never make the return link a condition
The moment reciprocity becomes the price of entry, both links stop being editorial decisions. It also poisons the relationship: a partner who feels obliged is a partner who removes the link the next time they clean up their outbound links.
- 5
Keep exchanges as a small share of acquisition
Not because of a threshold, because of concentration risk. A profile whose growth depends on trades is exposed twice over, once to a policy change and once to every partner who redesigns. Neither of those is under your control.
- 6
Judge the site, not the third party metric
Does it rank for anything. Does the archive have a history. Does the content read like it was written for a person. How many unrelated outbound links sit in the footer. Domain rating is an estimate of link strength from a third party crawler, and it is the easiest number in this industry to inflate.
The part nobody does
Find your own reciprocal footprint first
Ask an SEO team how many reciprocal links their site has and you will usually get an estimate, delivered with a slight pause. Ask how they arrived at it and the pause gets longer.
The reason is structural rather than lazy. Every backlink tool shows you who links to you. Crawlers show you who you link to. Almost nothing shows you the intersection, and the intersection is the only thing that answers the question. So the number stays unknown, and an unknown number is impossible to have an opinion about.
It is a couple of hours of work to fix, once, and then it is a report you can rerun. Here is the sequence we would use.
How to find and judge your reciprocal links
Two exports, one intersection, then five minutes per row on the ones that survive.
- 1
Export your referring domains
From Ahrefs, Semrush, Majestic, or the Links report in Google Search Console. Search Console is free and it is the one dataset that is definitely Google's own view, though it lags and it truncates. If you only have one source, use two: no single backlink index sees everything.
- 2
Export your own outbound links
This is the half nobody does, and it is why the whole exercise gets skipped. Crawl your own site with Screaming Frog or your tool of choice and export external outlinks. You now have two lists: who links to you, and who you link to.
- 3
Intersect the two lists on domain
Match on registered domain, not on URL, or you will miss every case where they link to your blog post and you link to their pricing page. The overlap is your reciprocal footprint. For most sites it is a much shorter list than people expect, and seeing the actual number settles a surprising amount of anxiety on its own.
- 4
Sort by relevance, not by domain rating
Go down the list asking one question per row: does a reader on their page plausibly want what we published, and the other way around? Domain rating tells you nothing here. A DR 70 general news site can be a worse neighbor for a link exchange page than a DR 25 tool your customers actually use.
- 5
Open the page each link sits on
Not the domain, the page. A link inside a paragraph reads differently from the fortieth entry on a partners wall. You are looking for pages whose only content is outbound links, and for your link sitting in a footer that renders across the whole site.
- 6
Read all the anchor text at once
In a single column, sorted. One exact match commercial anchor means nothing. The same phrase across eleven partners is a pattern, and it is usually the clearest evidence of how those links were acquired, whether or not you were the person who acquired them.
- 7
Decide per link and write down why
Keep, ask for a change, or remove. Record the reason beside each one. It takes an afternoon and it is the difference between having an answer and having an opinion the next time somebody asks why your profile looks the way it does.
Steps one to three are a data problem: knowing who you link to, what each partner agreed to, and whether the return link is still live today. That is the part LinkSwapy handles, and the part a spreadsheet stops handling somewhere around a hundred placements. Steps four to seven are judgment, and no tool does those for you. See how return link tracking works.
By business type
SaaS companies, agencies, and bloggers
The general rule is the same for everyone, whether you are in-house or agency side. What changes is how many natural reciprocal links you already have before anyone starts building any, and that changes what a reasonable footprint looks like.
SaaS companies have the most, by a wide margin. Integration partners, technology partners, agencies in the ecosystem, comparison pages, tool recommendations, customers who publish case studies. A mature SaaS site can have dozens of mutual links without one of them being a trade, which means a raw reciprocal count on a SaaS domain is close to meaningless. Two unrelated SaaS products swapping links for rankings is a different thing entirely, and it looks different too.
Agencies sit in the middle. Real relationships with developers, designers, hosting companies, consultants and clients, all of which produce reasonable mutual links. The failure mode is specific to the business model: turning the client roster into a linking network, where every client site carries a footer credit and every credit is a link. That is a sitewide, templated, low relevance link repeated across a portfolio, which is a pattern before it is anything else.
Bloggers and publishers have the fewest structural reasons for reciprocity and the most exposure to the soft version of it: I will mention your piece if you mention mine. The problem is not the link, it is that the editorial decision moved. Once you are citing something because of what you get back, the citation stopped being a citation, and readers can usually feel it even if they cannot name it.
Better use of the effort
What to do instead of scaling exchanges
Exchanges scale badly on purpose. Every additional one takes the same amount of negotiation as the first, produces a weaker link than the first, and adds to a pattern that gets less defensible as it grows. That is a poor return curve for something you are doing manually.
The methods below share one property that trades do not have: the output keeps working after you stop. A statistics page earns links for years without another email being sent. A trade earns one link and needs the next email.
None of this is fast. That is the actual trade-off, and anyone telling you otherwise is selling something.
- Linkable assets: original studies, statistics pages, free tools, calculators, templates
- Digital PR, meaning stories and data a journalist can legitimately reference
- Expert commentary, contributed where you genuinely know the subject
- Resource outreach for pages your thing actually belongs on
- Content partnerships where the collaboration is the point and links follow
- Unlinked brand mentions, which are the cheapest links on this list
- Reclaiming links you already earned and quietly lost
The verdict
So, are reciprocal links bad for SEO?
No, not as a category, and the framing does most of the damage here. Reciprocity is a shape in a link graph, not a behavior. What creates risk is trading at scale between sites with nothing in common, on pages built to hold links, with anchors somebody dictated. Strip those away and you are left with two businesses referencing each other, which is what the web has always done.
- Mutual links between real partnersLow
- An occasional negotiated link that belongsLow
- Regular trading with thin relevance
- Exchanges as your main acquisition method
- Automated swaps across unrelated domainsVery high
Link because the destination earns it. When somebody links back for the same reason, that is not a scheme, it is just what a working relationship looks like.
Questions
Questions people actually ask about this
What is a reciprocal link?
Two websites that link to each other. Site A links to site B, site B links back to site A. That is the entire definition, and it describes direction only. It says nothing about whether either link was earned, negotiated, paid for, or placed by a script, which is why the term causes more arguments than it settles.
Are reciprocal links bad for SEO?
Not automatically. Suppliers, integration partners, associations and research collaborators link to each other constantly, and none of that is what Google's spam policies are aimed at. The risk comes from what surrounds the reciprocity: whether the sites are relevant, how many of these you have, whether the anchors repeat, and whether the links would exist if backlinks counted for nothing.
Do reciprocal links pass PageRank?
They can, in the sense that nothing about reciprocity strips a link of its ability to pass signals. What you cannot assume is that the two links cancel out into a fair trade, or that either one is being counted at all. Google weighs a link on relevance, placement, the quality of the linking page and the pattern around it. Reciprocity is one input among those, not a switch.
Are reciprocal links the same thing as a link exchange?
No, and conflating them is the root of most confusion here. A link exchange is an arrangement, meaning somebody agreed to trade. A reciprocal link is a shape, two sites pointing at each other, which can happen with no arrangement at all. Every exchange produces reciprocal links. Plenty of reciprocal links come from nobody agreeing to anything.
How many reciprocal links are too many?
Google has published no threshold, and any percentage you have seen quoted was invented by whoever wrote it. Fifty reciprocal links across fifty real industry relationships is a different object from fifty acquired in a Slack group last month, and no ratio separates them. Judge relevance, placement, anchor spread and how the partners were chosen.
Should I remove my reciprocal backlinks?
Almost certainly not as a blanket action. Stripping legitimate partner links out of a profile to fix a problem you have not confirmed costs you real signals in exchange for nothing. Find them first, score them individually, and only act on the ones you could not explain to the client whose site they point at.
Are one-way backlinks always better than reciprocal ones?
On average yes, because a one-way link is harder to obtain and reads as an independent editorial decision. But average is doing a lot of work in that sentence. A one-way link from a scraped directory is worth less than a reciprocal link from an integration partner whose docs actually send you customers. Direction is a weak proxy for quality, not a substitute for it.
Keep reading
Related reading and tools
The policy side of this question, plus the operational half: qualifying partners, holding them to what was agreed, and knowing the day a placement disappears.
Link Exchange Guidelines: Domain, Placement, Anchor, and Quality Standards
Most teams have standards, in one person's head, and they hold until volume outgrows that person. Here is the version you can hand to somebody else: eligibility, domain and page criteria, anchors, and the verification step nobody writes down.
Read moreLink Exchange vs Listicle Exchange: SEO Value, Risks, and Which to Use
Listicle exchanges get sold as the natural-looking version. They are also the only exchange format where the page makes a factual claim about merit, and where the two sides are almost never trading assets of equal value.
Read moreAre Link Exchange Platforms Safe? How to Judge One Before You Join
How a platform makes money predicts what it will push you toward, more reliably than anything on its homepage. The five models, the incentive baked into each, and what to ask before you hand over a domain.
Read moreLink exchange tracker
Every partnership in one place: what was agreed, what is owed, and what actually went live.
Read moreFree backlink status checker
Paste your placements and see which are still live and still dofollow, with no signup.
Read moreLink building glossary
Plain definitions for anchor text, referring domains, link equity and the rest of the vocabulary here.
Read moreKnow which of your partners actually linked back
Track what each side agreed to, whether the return link went live, and whether it is still live today. No marketplace, no automated placement, no domain lists to trade from.