DANA: Here’s one. A site has about forty blog posts, and the owner wants every single internal link pointing to the main “online yoga classes” page to use that exact phrase as the anchor text. The logic is that repetition tells Google what the page is about. Forty links, one anchor, every time. Do we sign off?
PRIYA: The goal makes sense, the method doesn’t. You want that page to own “online yoga classes,” fine. But forty identical anchors isn’t a signal of relevance, it’s a signal of manipulation, and that’s the opposite of what they’re after.
NOAH: Let me put a number on the instinct, because the instinct isn’t random. Across those forty posts, how many naturally would say “online yoga classes” in a way that fits the sentence? My guess is a handful. The other thirty-something are being forced. So the plan isn’t forty natural links, it’s a handful of real ones and thirty bent into shape.
ELENA: And the bending is where it breaks. An anchor is supposed to describe what’s on the other end for the reader. If a post about morning stretches links the words “online yoga classes” mid-sentence where it doesn’t belong, the reader feels the seam. The link stops being navigation and starts being plumbing.
MARCUS: You’re all agreeing again, so let me take the owner’s side properly. Exact-match anchors do carry relevance weight, that part isn’t a myth. A page with anchors that describe it does tend to rank for those words. So the owner isn’t inventing the mechanism, they’re just dialing it to a hundred when it works at thirty. The question isn’t “does anchor text matter,” it’s “what happens past the point of diminishing returns.” Anybody actually answer that, or are we just calling it spam and moving on?
HANNAH: Fair, and that’s the part I won’t assert without grounding. The general principle is documented, search engines use anchor text to understand link targets and a varied, natural link profile reads as more trustworthy than a uniform one. The exact threshold where helpful becomes harmful isn’t a published number, so I won’t pretend there’s a clean “thirty percent” line. What I’ll stand on is the direction: uniformity is a risk signal, variety is not.
GRACE: There’s a reader cost nobody’s pricing in, separate from rankings. Forty posts, and every link out to the same page reads with the identical four words. By the fifth post a regular reader has seen “online yoga classes” linked so many times it goes invisible. The link loses its pull. Variety isn’t just safer for the algorithm, it keeps the link alive for the human.
SOFIA: That’s exactly my angle, the click. An anchor is a tiny piece of persuasion. “Online yoga classes” tells the reader nothing about why to click right now. “Follow a beginner flow you can do before work” tells them what they get. Different anchors aren’t just risk reduction, they let each link sell the destination from the angle that fits its own post. Same target, forty reasons to click.
THEO: Step back to the process, because the request has a hidden assumption. The owner believes the page underperforms because Google doesn’t understand it, so more identical anchors will clarify it. But that’s a guess about cause. Maybe the page ranks poorly for a different reason entirely, thin content, slow load, weak topical support. Pouring forty matching anchors at a page that has a content problem doesn’t fix the content problem. It just adds a manipulation pattern on top of it.
AIKO: And the systems note, which isn’t about this page. “Use the same anchor everywhere” is a rule of thumb the owner picked up somewhere and applied wholesale. The real fix isn’t correcting forty links once, it’s giving them a simple principle they can apply themselves going forward. Something like “link with the words that describe what the reader will find, vary them naturally, and only link where it genuinely helps.” That outlasts any single cleanup.
MARCUS: That’s better, but it’s still soft. Give me the actual distribution. What does “varied naturally” look like in practice, or are we handing them a vibe?
ELENA: Then here’s the concrete version. A healthy profile to that page looks like a mix, not a monoculture. Some links use the exact phrase where it reads naturally. Some use partial variations, “online yoga sessions,” “guided classes at home.” Some use descriptive phrases tied to the specific post, “a flow for tight hips.” And some are plain contextual, “this class.” No fixed percentages, because the right blend depends on what each sentence actually needs. The rule is that the anchor serves the sentence first, the target second. Forty links, several shapes.
HANNAH: And every one of those still points at the same page, so the relevance the owner wanted is intact. They just stop concentrating it into one repeated string that reads as engineered.
DANA: Okay, decision. We don’t do forty identical anchors, and the reason isn’t “spam,” it’s that uniformity works against every goal the owner has, relevance, clicks, and reader trust at once. Here’s what we tell them. Keep the exact phrase where a sentence genuinely calls for it. Everywhere else, let the anchor describe what that specific post is sending the reader toward. Link only where it helps the reader, not to hit a count. And before we touch anything, one question back to them, what made them think the page was misunderstood in the first place, because if the real issue is the page itself, anchors won’t move it. Theo’s right that we’re guessing at cause until they answer that.
PRIYA: Which keeps the relevance goal alive while dropping the method that would have buried it.
DANA: Right. The target stays the same on every link. The words change to fit the reader in front of them.