Blog comment links

By Bill Hartzer · Hartzer Consulting · 25+ years in search, domains and expert witness work

What blog comment links are

A blog comment link is a link to your own site placed in the comment section of somebody else's post. In the polite version you write a paragraph responding to the article and your name links to your site through the website field of the comment form. In the crude version software posts a generic sentence - something agreeable and content-free - across thousands of blogs, with a keyword phrase in the name field so the anchor text reads as a commercial term rather than a person.

For a period in the mid-2000s this genuinely worked, and worked well enough to break the format. Comment sections filled with automated spam, publishers spent their evenings deleting it, and a great many simply turned comments off. That history is why the tactic is on this reference at all: it is the clearest case in link building of a tactic that succeeded so thoroughly it destroyed its own inventory.

I have graded this obsolete rather than high risk, and the distinction matters more than it might appear. High risk on this site means against published guidelines, or reliably ineffective, or both - a live tactic with a price list and a downside. Obsolete means the thing is simply gone. Nobody is getting a manual action in 2026 for comment links, and grading them dangerous would be inaccurate. They do nothing. That is the finding, and accuracy about it is the entire point of grading tactics at all.

How it worked, step by step

Worth describing in the past tense, because every stage of the chain now has something in the way of it.

1. Find blogs on a subject that accept comments. Once trivial with a few search operators against the footprints that blogging platforms leave in their URLs and page furniture.
2. Post a comment carrying a link. Either through the website field attached to the commenter's name, or as a raw URL in the body where the platform allowed it.
3. Optimize the name field. Instead of a person's name, a commercial phrase, so that the resulting anchor text was the term being targeted. This is the single most recognizable artifact of the era - a profile full of links anchored on service phrases, each attached to a two-line comment saying great post.
4. Scale it with software. Comment posting tools worked against lists of thousands of blogs, with spun text to vary the wording.
5. Chase the plugins that rewarded commenters. For a while, certain popular plugins gave commenters a followed link to their most recent post as an incentive to participate. Those plugins are gone, and their disappearance took the last followed comment inventory with it.

Each of those steps now fails. Search operators still find blogs, but most no longer take comments. Comment forms are behind moderation queues and spam filters that catch automated posting with near-total reliability. And the link, if it survives all of that, carries an attribute that stops it passing anything.

What it costs in time and effort

Nothing, in money. That is precisely the problem, and it is worth stating as an economic principle rather than a complaint: a link anybody can give themselves for free has no scarcity, and a signal with no scarcity carries no information. Search engines cannot treat a self-issued link as an endorsement, because the endorsement would be from you.

Automated comment services still exist and cost very little, sold by the thousand, usually as fodder for tiered schemes. The residual manual version costs real time: reading an article properly, writing something a publisher would want to approve, and doing that consistently enough that the author starts recognizing your name. That is a worthwhile activity, but it is a relationship activity whose eventual payoff is being thought of - not a link tactic, and it should not be budgeted as one.

The one genuine cost is on the other side of the form. Publishers still pay for moderation in attention and in spam filtering - a small ongoing tax levied on the whole web by a tactic that stopped working two decades ago.

When it works and when it does not

It does not work, and it fails at four separate points.

The attribute. The rel="nofollow" attribute was introduced in 2005 by the major search engines for the express purpose of neutralizing comment spam, and blogging platforms applied it to comment links by default almost immediately. It was later joined by rel="ugc", which marks user-generated content specifically. Comment links have therefore been explicitly excluded from passing ranking credit for the entire working life of most people reading this.

Moderation. Default configurations hold first-time commenters for approval, and spam filtering catches automated posting at rates that make bulk campaigns pointless. The comment that gets published is the one a human read and approved, which by definition cannot be produced at scale.

The inventory. Comment sections closed. Discussion migrated to social platforms and messaging apps, and a large share of publishers removed comments entirely rather than keep moderating them. The targets are not there.

Relevance and endorsement. This is the deepest reason and it would apply even if the other three vanished. A link is supposed to represent a publisher's judgment that a reader would benefit from following it. A comment link involves no judgment by anyone except you. It is not a publisher pointing at you; it is you pointing at yourself on a page you do not own. No amount of volume converts that into an endorsement.

The residual version that has value is unrecognizable as this tactic. Commenting on a handful of sites you actually read, under your own name, saying something worth reading, occasionally leads to the author knowing who you are and citing you later in an article they wrote. That is networking with a link as an eventual byproduct. Do it if you enjoy it. Do not put it in a link plan.

Common mistakes

• Putting a keyword phrase in the name field. It fools nobody, it gets the comment deleted, and it leaves the most identifiable spam artifact in existence in whatever profile survives.
• Buying comment packages as tier fodder. A nofollowed link on an unmoderated dead blog has nothing to pass at any depth.
• Believing followed comment links are worth hunting. Sites that still allow them are almost always abandoned or already swamped, which is why the attribute was never applied.
• Panicking at spam comments pointing to you. Anyone can post a comment naming any site, so search engines must ignore rather than penalize this class of link. Finding thousands of them in your profile is normal and requires no action.

A worked example

Segment a legacy profile - one that has been accumulating links for a decade or more - and look at what the comment-sourced portion actually consists of. The exercise takes an hour and is more persuasive than any argument.

You will typically find the source pages are still live. Almost all of them. That fits the wider pattern: across 645,202 lost links on my own profile, 97.4% disappeared while the source page remained perfectly reachable, and only four were lost to a 404. Comment links go the same way - the post survives, the comment gets purged in a cleanup, and nothing about the page tells you.

Then check three things on the ones that remain:

1. The attribute. Nofollow or ugc, essentially without exception on any maintained platform.
2. Trust Flow. Zero, or close to it, across the set. On my own site, 69.3% of 22,260 referring domains sit at Trust Flow 0 - on a profile where no link was ever bought. Zero-trust links accumulate by themselves in vast quantity. Comment sections are one of the places they come from.
3. Referral traffic. None. Nobody reads the comments under a nine-year-old post, and the ones who do are not clicking through to a stranger's website.

The conclusion is the useful part. The web hands out this category of link for free, in bulk, whether or not you want it. Anything the web gives away at that volume cannot be a ranking advantage, because everybody already has it.

How to measure it

The only measurement worth doing here is exclusion, and it is genuinely worth doing.

Segment comment-sourced links out of your link reporting. If your referring domain count includes several thousand comment links at Trust Flow 0, that number describes the web's background noise rather than your link building, and it will be quoted back at you in a meeting eventually. Report referring domains above a trust threshold, or report the total with the zero-trust portion broken out beside it.

If someone insists on a positive measurement for manual commenting as an outreach activity, measure the right thing: not links, but replies from the author, subsequent email exchanges, and citations in later articles they wrote unprompted. Those are the outputs the activity actually produces. Judged as a link tactic it will always look like a failure, because it is.

And do not build a disavow file out of comment links. There is no problem to solve. The links are inert in both directions, they arrived without your involvement, and a large speculative disavow is more likely to remove something you valued than to fix anything.

The verdict

Blog comment links are obsolete, not dangerous. Nobody is going to be penalized for them in 2026, and if a profile you inherited contains thousands of them, the correct action is none. They were neutralized by an attribute invented specifically to neutralize them, filtered by moderation systems that have had twenty years to improve, and finally starved of inventory as publishers closed their comment sections in self-defense.

The lesson worth carrying out of this page is not about comments. It is about what makes a link worth anything in the first place. A link is a publisher's judgment, expressed publicly, that their reader would benefit. Any tactic that lets you issue that judgment on your own behalf is not producing links in the sense that matters, and it will eventually be recognized as such - which is exactly what happened here, in full public view, over a couple of years.

Comment on sites you read, under your own name, because it is a decent way to be known. Do not put it in a plan.