Rented links and the neighbours they come with

The cheap link package is seductive because the numbers go up first. Rankings lift for a few weeks, everyone is pleased, the contract gets renewed. Then they slide back, past where they started, and the recovery never arrives. That sequence is not bad luck and it is not Google being fickle. It is what the product was always going to do.

This is failure number four in my series on why SEO agencies fail, and it has the longest tail. Most of the others cost you time and fees. This one can cost you the domain. The mechanics are worth understanding, because an owner who understands them can spot the problem in ten minutes without a tool.

What you are actually renting

When a link package costs a few hundred dollars a month and promises dozens or hundreds of links, those links almost always come from a private blog network. A PBN, in the jargon. It sounds technical. It is not. A PBN is a set of websites that exist to sell links to whoever pays.

Here is how one gets built. Websites die all the time. A regional news site folds, a hobbyist loses interest, a company goes under, and the domain expires and goes to auction. Some of those dead domains still carry links from real, respectable places - a council page, a university department, a newspaper article from 2014. Those links are the asset, because Google still sees a domain that trustworthy sites vouched for.

So somebody buys the corpse, puts a generic blog on it, and it looks, at a glance, like a working website. Then they sell link slots. Your plumbing business gets a paragraph on drain unblocking linking to your services page. The next customer gets crypto exchanges. The one after that gets vape juice, an offshore casino, a payday loan site.

That is the neighbourhood problem. You did not choose your neighbours, and you cannot see them from your own site - they are visible only on the pages your links sit on, which is exactly where nobody looks.

Why these networks exist at all

I want to be fair rather than indignant here, because networks exist to solve an honest supply problem. A genuine link - a trade publication mention, a supplier listing you as an approved installer, a local news piece, a sponsorship page for a club you support - takes real work. Someone has to find the opportunity, have something worth pitching, and write to a human being. Most pitches get ignored. A handful of good links is a decent month's work.

Now imagine a package at a few hundred dollars a month promising fifty links. There is no version of that where a person does outreach for fifty individual placements. The maths does not close. The only way to deliver is to buy inventory that already exists, which is what a network sells. An owner holding two hundred expired domains and selling a few slots on each gets the cost of a link down to a few dollars. That is why the package exists at that price.

Nobody can sell you fifty links a month for three hundred dollars and also have spoken to fifty human beings. The price tells you the method.

Some agencies selling this know exactly what it is. Others genuinely do not - they resell a white-label package, the supplier sends a report full of URLs, and nobody ever opens one. I have met both kinds, so I would rather explain the model than accuse anyone.

Why the early lift is real, and why that is the dangerous part

If these links did nothing at all, the trade would have died years ago. They do something, at first, and that is the trap. Google does not evaluate a link the instant it appears and hand down a verdict. It counts it provisionally and refines its view of the source over time, as evidence accumulates about that site and the sites around it. So for a period - often weeks, sometimes months - network links behave like real ones, and on a site with little existing authority the movement can be dramatic.

Look at what that does to the decision-making. Month one, rankings up. Month two, up again. The client is delighted, the agency looks competent, and the obvious move is to buy more. By month six there are not fifty links pointing at the site, there are four hundred. The lift bought the commitment. Then the assessment catches up and it unwinds at once.

That is why the drop is worse than the starting position. The site had been coasting on borrowed authority, so removing it takes you below where you began, and the domain now carries a visible pattern of unnatural linking. Two problems where there was one. It is the same story as paying for SEO and going backwards, and network links are the most common cause of it.

Rented means rented, and discounted means gone

The word rented is literal. There are two ways one of these links stops paying, and most owners have thought about only the first.

• You stop paying. The link sits on someone else's website and stays there exactly as long as the invoice is current. Cancel the package and the links come down, sometimes within days, and whatever ranking they were holding up comes down with them. You were not building an asset. You were leasing one.
• The network gets discounted. This one you cannot control by paying. Google works out that a set of sites is a link operation rather than a set of websites, and the value of every link across it goes to zero at once. Your invoices are current, the links are still visibly there, and they are worth nothing.

Networks get found because they leave a pattern. Hundreds of sites on unrelated topics, all linking out to commercial pages, often sharing hosting, a template, a writing style and a publishing rhythm, with the same client sites appearing across them in the same combinations. Any one of those signals means nothing. Together they describe a footprint, and finding footprints across thousands of sites at once is a machine's idea of a good afternoon.

When it happens the effect is a step change rather than a slope, which is why it looks like a cliff from the outside. And it does not stop at ignoring the links. Association matters: a site whose backlink profile is largely made of links from a network that has just been written off does not revert to where it was before. In the worst cases that becomes a collapse that never recovers on its own, and the long version is in how a cheap SEO provider can permanently damage your domain. It is also why you cannot buy your way back out. Adding good links on top does not dilute the bad ones - there is no ratio where a hundred clean links cancel four hundred dirty ones.

The ten-minute self-check

You do not need a tool for this, or any SEO knowledge. You need a link list and ten minutes.

1. Ask for every link built in the last six months as a plain list: the URL of the page your link sits on, and the page of yours it points to. Any provider doing the work already has it. If the list never arrives, that is its own answer.
2. Open five of those pages at random. Not the five they would pick - five you pick, from different points in the list.
3. Read each for one minute as a human being, not as a client. Ignore the design and ask one question: does this look like something a person would actually read?
4. Scroll the other articles and see whether they belong together. A site carrying posts on tax software, dog grooming, online pokies and skip bins is not a publication. It is inventory.
5. Look for a human being: a named author with a history of posts, an address, a phone number, a staff page. A site nobody stands behind is a site nobody reads.
6. Look for the giveaway pages. A prominent "write for us" or "sponsored posts" page, especially one quoting prices, tells you the site's product is links, not readers.
7. Check whether the article is about anything. Network articles tend to be four hundred words of sourceless generality with your link dropped into the third paragraph.
8. Count your score. If most of the five fail, the money has been buying risk rather than authority - and proportion matters, because a profile made almost entirely of these is different from a handful on an otherwise natural one.

Being fair: when a cheap link is just a cheap link

I do not want to leave you suspicious of every link you have ever received, because that is its own expensive mistake. Cheap does not mean toxic, and a guest post is not automatically a network placement.

Sponsoring the local junior rugby club and getting a link from the sponsors page might cost a few hundred dollars and is completely legitimate - that page exists because the club has sponsors, not because someone is selling link space. A trade association directory, a supplier's approved-installer list, a genuine local news mention, a real industry blog that edits what it publishes: all fine, often cheap, all things I would build for a client.

The distinction is not price, and it is not whether money changed hands. It is whether the site exists for its readers or for its link buyers. A real publication has an audience it would lose if it published rubbish, so it has a reason to say no to you. A network site has no audience to lose, which is why it says yes to everyone. Sites also pick up odd links naturally, so one ugly link is not a crisis. Proportion and pattern are what matter.

Does a disavow fix it?

Google provides a disavow tool. You give it a list of domains and it tells Google to disregard links from those sources when assessing your site. It exists for precisely this situation, so yes, it is worth doing. But I would be misleading you if I called it a reset button.

• You have to identify the right links. Without a full list from the provider you are guessing which of hundreds of domains did the harm, from third-party data that is always incomplete.
• It is slow. Google has to re-crawl and re-assess the affected pages before anything changes. Months, not weeks, with no confirmation that the file was even right.
• It is a request, not a command in your favour. You are asking Google to discount signals, not erasing the site's history.
• It can be overdone. Sweeping in legitimate links because they looked unfamiliar throws away authority you earned.

Sometimes a disavow clears it and the site recovers. Sometimes it does not, and after a year of waiting the honest conclusion is that the domain is not coming back. That is when the conversation turns to walking away from the domain and starting fresh, a bigger decision than it sounds.

If you have already bought them

It depends on volume and how long it has run. Here is the order I would work in.

1. Stop buying more. Adding to the pile makes every later decision harder and more expensive.
2. Get the full list out of the provider while the relationship still exists. The day after you cancel is the worst time to ask, and without it a cleanup becomes guesswork. Ask in writing, as URLs, now.
3. Diagnose before you act. Match when the rankings turned against when the links were built. If traffic fell off a cliff in a particular fortnight, that is the most useful evidence you have.
4. Do not buy good links to cancel out the bad ones. It does not work that way, and it is the most common expensive reflex I see.
5. Take down what you can. Rented links stop when the rental stops. Rankings may fall further first, but you are removing a dependency you were never going to keep.
6. Then consider a disavow, built conservatively from the real list, and give it a defined window - two or three quarters - with a decision made in advance about what happens if nothing moves. Drifting for two years hoping is the worst outcome available.

When you next hire someone, make this a hiring question rather than an audit question. Ask where links will come from, ask to see sites they have placed on for other clients, and open two of them in front of them. Anyone doing real work will enjoy that conversation, which is why it is one of the more revealing things to raise when choosing an SEO provider.

The thing to take away is that this is checkable. You do not have to trust anyone's word on link quality, including mine. Get the list, open five pages, read them as a person. A website that no human reads is obvious to any human who looks at it, and ten minutes of looking is the cheapest due diligence in this industry.