What Is Grey Hat SEO?
The interesting question is not what the term means. It is why the category exists at all. There is a small genuine middle ground. There is also a much larger space where the word is being used to make something sound more acceptable than it is.
Why The Category Exists
Grey hat describes practices that are not obviously prohibited and not obviously safe. That is the definition. What matters more is the job the phrase usually does in a conversation.
Where you will have met it. In a proposal.
Somebody describes an approach, senses hesitation and reaches for the word. It arrives as reassurance: not black hat, just a bit grey, everybody does it. That framing is the reason the term is worth examining rather than simply defining.
What the word is doing there. Supplying permission.
People want a vocabulary for things they suspect are wrong but would like to do anyway. Grey is that vocabulary. It converts a decision about risk into a decision about shading, which feels considerably more comfortable.
Why that matters to you specifically. It shifts the question.
Once something is grey, the conversation becomes how grey rather than whether to do it. The second question is the one worth asking and the word quietly removes it.
What this page does about it. Gives the middle ground its due first.
Some practices genuinely sit between. A page pretending otherwise would deserve to be ignored. That comes next, then the harder part afterwards.
What Genuinely Sits In The Middle
A real middle ground exists. It is smaller than the term suggests. It is made up of practices that are legitimate in one form and problematic in another, where the difference is intent and volume rather than the practice itself.
The first genuine case. Legitimate at one scale, not at another.
Plenty of ordinary marketing activity is unremarkable when done occasionally for a reason and becomes something else entirely when industrialised. The activity did not change. The purpose behind it did.
The second. Where published guidance has not caught up.
New formats and new platforms appear faster than policy is written about them. There are periods where nobody can say with confidence how something will be treated. That is a real uncertainty rather than a manufactured one.
The third. Where reasonable people disagree.
Some practices are within published policy and still considered poor form by experienced practitioners. That is a professional judgement rather than a compliance question. The two are worth separating.
What is not in the middle. Anything failing the tests in the next block.
Deception and fabrication are not shades of anything. They are the two things the black hat pillar is built around and they do not become grey by being described that way.
The Test That Actually Works
Would you be comfortable explaining this to the client, to a competitor and to the search engine in the same sentence. Anything failing that is not grey.
Why three audiences rather than one. Each catches something different.
A practice can be defensible to one and indefensible to another. It is usually the second audience that exposes it. Testing against all three at once is what makes the question difficult to argue around.
The client. Would they proceed if they understood it fully.
Anything relying on the buyer not asking the next question is not a method, it is a gap in their knowledge being monetised.
The competitor. Would you describe it at a conference.
Practices people are proud of get talked about openly. Practices people are not stay in private conversations. That distinction is remarkably reliable.
The search engine. Would you describe it in a review request.
If explaining the method would make things worse rather than better, the method was the problem. This is the sharpest of the three, because it is the only one that ever gets tested in practice.
Why the test beats a list. Nothing expires.
Techniques change constantly. A question about whether you could say it out loud applies to anything anybody invents next year.
Scale Turns White Into Grey
A thing done once for a good reason and the same thing done a thousand times automatically are not the same practice. Volume is usually what moves something across the line. Automation is usually what supplies the volume.
Why volume changes the nature of it. Purpose becomes visible.
Something done occasionally, for a specific reason, with a person deciding each time, is a judgement. The same thing done at scale without anybody deciding is a process. A process aimed at producing a signal is the definition of manufacturing one.
Why automation is the usual trigger. It removes the deciding.
The moment a practice no longer requires anybody to consider whether this particular instance is worth doing, the thing that made it legitimate has been taken out of it. That is the shape of it. It applies far more widely than to any one technique.
Where you will see this argument used badly. In both directions.
Some suppliers defend industrialised activity by pointing at the legitimate small-scale version. Others condemn perfectly reasonable work by pointing at what it resembles at scale. Both are avoiding the actual question, which is how much and decided by whom.
The question to ask. Two parts.
How many of these will be produced, then who decides each one. Vague answers to either are the finding.
What Grey Hat Actually Means In Practice
In most conversations, grey hat means somebody has judged the risk acceptable. That is worth hearing plainly, because it turns a vague description into a specific claim you can examine.
What has actually been said. A risk assessment.
Not that the practice is fine. That somebody weighed a probability against a benefit and decided the trade was worth making. That is a legitimate kind of decision and it is being made on your behalf.
The question that follows. Who carries it.
Every risk assessment has somebody holding the consequence. Establishing who that is comes before agreeing to anything. It is almost never the person making the assessment.
Why the word obscures this. It sounds like a property.
Grey sounds like a description of the practice, as though the risk were a feature of the technique rather than something landing on a particular party. Naming who holds it turns the conversation back into a commercial one.
How to ask it without an argument. Plainly.
If this stops working, what happens to us and what happens to you. A supplier with a good answer gives it immediately. A supplier without one changes the subject. That is the next block.
The Risk Sits With The Client
This is the part worth remembering if you remember nothing else from the page. When a risky method stops working, the agency moves on and the business keeps the site.
What the agency loses. A client, eventually.
Their exposure is commercial and recoverable. They have other clients, the method worked for a while and they will describe what happened as an industry change rather than as a decision they made.
What the business loses. Everything downstream of it.
The rankings, the enquiries those produced, the revenue those produced. Then the cost of establishing what was done, the cost of undoing it and a period of trading without any of it while that happens.
The asymmetry is the whole point. Not a criticism of anybody's character.
Somebody proposing a method is assessing a risk they do not carry. That is not necessarily deceitful and it is structurally guaranteed to produce a different answer from the one you would reach yourself.
What follows practically. Ask for the assessment, not the reassurance.
Anybody describing something as low risk should be able to say what the risk is, what would trigger it and what happens to your business if it does. What to look for in a supplier is covered in our material on hiring an SEO agency.
Nothing In This Cluster Is Advice To Try Something
Describing where a line sits is not an invitation to stand near it. This is the page in the cluster most likely to be read as permission, so the position is stated here rather than left to be inferred.
Why the risk of misreading is real. The subject invites it.
A page explaining that some things are less prohibited than others can be read as a map of what you might get away with. That is not what any of this is for. Reading it that way misses what the previous block established about who pays.
What the cluster deliberately does not contain. Method.
No page here explains how any practice is carried out. Not because the information is unavailable elsewhere. Because publishing it would make this material useful to somebody attempting these things rather than to somebody assessing them.
Why we name practices at all. Recognition.
You cannot assess a proposal you have no vocabulary for. Naming a category and explaining its risk is what lets a business owner ask a better question, which is the entire purpose of the cluster.
The distinction in one line. Recognising is not attempting.
Knowing what a practice is and why it is a risk is protective. Knowing how to do it is not, which is why one of those is here and the other is not.
What We Will Not Do
It is easier to judge an agency by what it declines than by what it promises. These are the categories we will not take on regardless of what a client asks for. They are worth stating, because most agencies do not state them.
Anything showing search engines something people do not see. Without exception.
This is the first test in what is black hat SEO. There is no version of it we will build.
Manufactured references. Bought, exchanged or networked.
Links exist to record that somebody chose to point at you. We will not arrange for that to be simulated. We will say so if it is proposed to us.
Pages produced at volume to catch searches. Including location pages.
We build town pages, so this one costs us something to say. A page has to carry information that exists nowhere else. If we cannot write that for a given town, we will tell you the page is not worth building.
Guarantees we do not control. Positions, timescales, recovery outcomes.
Nobody controls these, so promising them is not a commitment. It is a sales technique with a refund attached at best.
Work we cannot explain to you. The one that covers everything else.
If we cannot describe what we are doing and why in ordinary language, we should not be doing it. The published position it rests on is on the Google spam policy page. The full series is on the dangerous SEO practices guide.
Ask what
happens when
it stops working.
A review that tells you what is actually in your site, what carries risk and what does not, written so you can act on it rather than worry about it. We will tell you plainly if we find nothing, because most sites we look at have nothing worth removing.
What a risk review covers:
A risk review is a one-off piece of work rather than a monthly service, so it is scoped against the size of the site and how much history it carries.
Every guide.
One practice.
Black hat, grey hat, link risk, thin content, the spam policies, negative SEO, manual actions, algorithmic penalties, how to check for one and how to recover from one.
