How the grades work
This page grades all thirty-four tactics against four risk grades, with the one-line reason for each. It exists so you can answer one question quickly: is the thing I am about to do, or the thing somebody wants to sell me, defensible?
I want to be clear about what this page is. Several of the tactics graded high risk below are actively sold, right now, by agencies with good websites and real clients. Publishing a grade on them is a public and unhedged judgment made under my own name, and I am not going to soften it with the usual line about how every situation is different. Every situation is different. The mechanisms are not.
The standard each grade is measured against
A grade here is an answer to a specific question, not a general vibe. The question is: if you had to describe exactly what you did, in detail, to the search engine whose rankings you are trying to influence, would you have anything to explain away? That is the test. It is deliberately not "would you get caught," because enforcement priorities change. And it is not "does it work," because plenty of things work briefly. Two questions modify the answer: does the tactic still function mechanically at all, and does the risk scale with volume or is it fixed? The second is what separates a situational grade from a high-risk one.
Safe
Defensible on its face under Google's published link spam policies. The link is the result of somebody else's editorial decision, or of a legitimate request you would be comfortable having read aloud. Doing more of it does not change its character: a hundred broken link building emails are no more suspect than one, provided each is a genuine, qualified, relevant ask. Safe does not mean effective and does not mean easy — several safe tactics are slow, expensive and produce few links. The grade is about defensibility only.
Situational
Safe when executed well, in proportion, by somebody exercising judgment — and a problem at volume or executed carelessly. This is the largest and most interesting grade, and it is where most genuine disagreement in the industry actually lives. Nobody argues about whether directories are worth doing in principle; they argue about which directories and how many. The defining feature is that the risk is a function of your execution rather than of the tactic itself. A single sponsorship of a local charity you actually support is not a link scheme. Forty sponsorships bought purely for links, from an intermediary, on sites that exist to sell sponsor listings, is one. Same tactic name, different thing entirely. On a situational profile, the "when it works and when it does not" section is where the line is drawn.
High risk
In violation of published guidelines, reliably ineffective, or both. The "or both" matters — some tactics land here not because they are dangerous but because they combine a guideline violation with results close to zero, which is the worst possible trade. What these share is that the risk does not depend on execution quality. A carefully built private blog network is still a private blog network. If the mechanism only works because the search engine has not yet identified it, the grade is high risk however well it is run.
Obsolete
Not risky. Gone. The mechanism that made the tactic work has been discounted, removed or automated into worthlessness, and doing it now produces nothing at all. These tactics are documented because they are still recommended in old articles, still bundled into cheap packages, and still bought by people with no way to tell that a link which does nothing looks identical to a link that does something. Grading them high risk would have been inaccurate, and inaccuracy would have made this whole page less trustworthy.
Grades change
A grade is a judgment about a mechanism at a point in time, and mechanisms move. Where a grade depends on something that could plausibly shift, the tactic's own page says so. If you think a specific grade is wrong, the profile gives you the reasoning to argue with — which is more than most link building content offers.