Introduction
A placed link is one that exists because you arranged for it to exist. Money changed hands, or a site was built for the purpose, or a paragraph was inserted into somebody's old article so that it could carry your URL. Nobody made an editorial decision to cite you. The link's reason for existing is that you wanted it.
Nine tactics sit in this category. Six are graded high risk and three are obsolete. There is no safe tactic here and there is no situational one, which makes this the only category on the site where the grades are uniform. That uniformity is not editorial laziness; it is the definition of the category doing its work. If a link exists only because you arranged it, the thing that would make it defensible — an independent decision by a publisher — is precisely what is missing.
Why document them at all
I considered leaving this category out. It would have made the site shorter and easier to defend. I kept it for one reason: readers are going to be sold these, and they deserve to understand what they are buying.
Nearly every link building proposal a business receives contains something from this category. It will not be labeled as such. The vocabulary is careful: "authority placements," "contextual link acquisition," "premium publisher inventory," "editorial insertions," "high DA guest post packages." Strip the language away and you find guest posts bought at volume, link insertions into existing articles, or a network of sites owned by the vendor. A buyer who has read these nine pages can ask the two questions that end the conversation: who owns the site the link will be on, and would this article exist if nobody were paying for the link in it?
The second reason is diagnostic. If you take over an established site you inherit its link history, including whatever the previous three vendors did. Recognizing a private blog network in a backlink export, spotting the footprint of a tiered scheme, or identifying a directory blast from six years ago is a real skill, and you cannot learn it from a reference that pretends these things do not exist. Module 3 teaches the detection method; these pages describe what you are detecting.
The third reason is that some of these tactics are simply dead, and being dead is different from being dangerous. Article directories, social bookmarking and blog comment links are graded obsolete rather than high risk because they will not hurt you — they will just do nothing at all, while consuming money. Somebody is still selling all three, because the marginal cost of producing them is close to zero and the buyer usually cannot tell the difference between a link that does nothing and a link that does something.
What the high-risk six have in common
Paid links and sponsored posts, niche edits and link insertions, private blog networks, tiered link building, web 2.0 and forum profile links, and press release syndication. Different mechanisms, same underlying problem in each case.
- The link's existence is not evidence of anything. The entire premise of links as a ranking signal is that a link represents a judgment by somebody with something to lose. A link you arranged carries no such judgment, which is why search engines put so much effort into identifying and discounting them.
- They leave patterns. Anything produced at scale for a purpose looks like it was produced at scale for a purpose. Shared hosting, repeated templates, identical publishing cadences, the same fifty client URLs distributed across the same two hundred sites, anchor text distributions no natural profile ever produces. Detection is a pattern-matching problem, and pattern matching is something search engines are extremely good at.
- The failure is asymmetric. When placed links work, you get a ranking improvement you could probably have achieved another way. When they fail, you can lose the rankings you had, and the cleanup — auditing, removing, disavowing, waiting — costs far more than the links did. You are accepting an unbounded downside in exchange for a bounded upside.
- You are renting. A bought link exists while the arrangement exists. The site gets sold, the owner cleans up, the network gets deindexed, the invoice stops. My own data shows that even honest links decay — the median referring domain stops linking after about 1,080 days, and 49.6% of all links ever recorded on my site are gone — and those were links nobody had a commercial reason to remove. Placed links have that decay plus a commercial reason.
I want to be accurate rather than dramatic here, so: yes, placed links demonstrably move rankings sometimes, in some markets, for some periods. Pretending otherwise would be dishonest and every experienced practitioner knows it. That is not the argument. The argument is that they move rankings on borrowed time, in a system explicitly designed to find and neutralize them, with a downside you cannot size in advance, and with a cleanup cost that lands on whoever owns the site later — often not the person who bought them.
The three obsolete ones
Blog comment links, social bookmarking and article directories were all mainstream tactics once, and all three stopped working for the same structural reason: the mechanism that made them work was cheap enough to industrialize, so it was industrialized, so it was discounted. The tools built to exploit them still exist. The packages are still sold. The results are nil.
They are documented here as accurately as the rest, without the sneering that usually accompanies them, because a reader who arrives from a 2011 blog post that recommended one of them deserves a straight answer about what happened rather than a joke at the expense of someone who asked a reasonable question.
If you are going to do it anyway
Some readers will. People inherit programs, have targets, and work for people who have already decided. So, plainly: the risk is not evenly distributed across these tactics, it scales with volume and with footprint, and it is at its worst when the same purchased pattern is applied to a site's money pages with commercial anchor text. Whatever else you do, keep a complete record of every placement you buy — URLs, dates, vendor, anchor text. When the cleanup comes, and the cleanup usually comes, the audit is enormously cheaper if the list already exists. The sites that suffer most from an eventual link-based problem are almost always the ones that cannot reconstruct what was done to them.
Read the individual profiles for the mechanism, the detection signature and the honest verdict on each. I have tried to be unimpressed without being inaccurate.