TL;DR: Anchor text is the clickable text in a hyperlink, and it still influences rankings, but not in the way most 2015-era guides describe. We analyzed anchor text patterns across 12,000+ site audits at SEOJuice (collected between 2024 and early 2026, covering sites from 50 to 500,000+ pages). The findings: over-optimized exact-match anchors are rarer than you'd think among top-ranking sites, branded anchors dominate healthy profiles, and the biggest mistake isn't using the wrong type. It's using generic "click here" anchors for internal links where descriptive text would actually help Google understand your content.
I wrote this because most anchor text guides I've read recently are either rehashing 2012 advice (which will get you penalized) or so vague they're useless. After spending three years looking at real anchor text data across thousands of sites, I wanted to document what actually correlates with rankings versus what the SEO echo chamber keeps repeating. Some of what I found contradicted what I'd been telling clients for years.
Anchor text is the visible, clickable text in a hyperlink. When you see a blue underlined phrase in an article and click it, that phrase is the anchor text. In HTML, it's the text between the opening and closing <a> tags:
<a href="/tools/seo-audit/">free SEO audit tool</a>
In that example, "free SEO audit tool" is the anchor text. It tells both users and search engines what the linked page is about before anyone clicks through. Google has used anchor text as a ranking signal since the beginning. The original PageRank paper from 1998 specifically noted that anchor text often describes target pages more accurately than the pages describe themselves.
That insight still holds. When 50 different websites link to your page using the phrase "internal linking tool," Google takes that as strong evidence your page is relevant for that query. The challenge is that this signal can be manipulated (and was, aggressively, for over a decade), which is why Google's treatment of anchor text has evolved significantly since Penguin first rolled out in 2012.
Not all anchor text is created equal, and the terminology gets thrown around loosely. Here's a clear breakdown with actual examples:
| Type | Example (linking to SEOJuice's audit tool) | Signal Strength | Risk Level |
|---|---|---|---|
| Exact match | "SEO audit tool" | Strongest topical signal | High if overused |
| Partial match | "run a quick SEO audit on your site" | Strong, more natural | Low |
| Branded | "SEOJuice" | Trust signal, weak topical | None |
| Branded + keyword | "SEOJuice's audit tool" | Both trust and topical | Very low |
| Naked URL | "https://seojuice.com/tools/seo-audit/" | Weak (no context) | None |
| Generic | "click here" / "learn more" / "this article" | Near zero | None (but wasted opportunity) |
| Image (alt text) | Alt attribute of a linked image | Moderate (if alt is descriptive) | Low |
The type that matters most depends entirely on context. For external backlinks, you mostly don't control what anchor text people use (nor should you try to — that's how you get Penguin'd). For internal links, you have complete control, and the choice is strategic.
This is where I can share data that most anchor text guides can't, because most guides are written from theory. We've crawled and analyzed over 12,000 websites through SEOJuice's audit tool between 2024 and early 2026. Here's what the anchor text landscape actually looks like.
I'll be honest: before pulling this data, I expected top-ranking sites to have more exact-match anchors than everyone else. I'd been telling clients "get keyword-rich backlinks" for years. The data said otherwise.
When we looked at the external backlink profiles of sites ranking in positions 1-3 for competitive terms (500+ monthly search volume), the anchor text distribution was not what I expected:
| Anchor Type | Median % (Top 3 Rankings) | Median % (Positions 20-50) |
|---|---|---|
| Branded | 38-45% | 22-28% |
| Naked URL | 18-25% | 15-20% |
| Partial match | 15-22% | 12-18% |
| Exact match | 3-8% | 14-25% |
| Generic | 8-12% | 10-15% |
| Other/long-tail | 5-10% | 5-12% |
The headline finding: sites ranking in the top 3 have significantly less exact-match anchor text and significantly more branded anchor text than sites ranking on page 2-5. This isn't necessarily causal (larger brands naturally get more branded mentions, and they have bigger SEO teams doing everything better), but it's a useful benchmark. If your exact-match anchor text percentage is above 20%, you're in unusual territory compared to what we see among top performers.
For internal anchor text (links within the same domain), the pattern reverses. Top-ranking sites use more descriptive, keyword-relevant internal anchors:
| Internal Anchor Pattern | Top Performers | Underperformers |
|---|---|---|
| Descriptive/partial match | 55-65% | 25-35% |
| Generic ("read more", "here") | 10-15% | 35-50% |
| Navigation-style (page titles) | 20-30% | 20-30% |
This makes intuitive sense. Internally, you're not at risk of a Penguin penalty. Google expects you to use descriptive anchors for your own pages. The opportunity cost of "click here" is enormous: every generic internal anchor is a missed chance to tell Google what the target page is about.
I want to be transparent about a limitation here: our data captures what correlates with good rankings, not what causes them. Sites with thoughtful internal anchor text also tend to have thoughtful content, logical site structure, and competent SEO teams. The anchor text might be a symptom of broader quality rather than the direct cause of rankings. Rand Fishkin made a similar point when discussing the 2024 API leak on SparkToro's blog: correlation in SEO data is everywhere, but isolating causation remains nearly impossible when Google uses hundreds of interacting signals. That said, anchor text is one of the cheapest things to fix, so even if the effect size is modest, the ROI is high.
Google's Penguin algorithm (first launched 2012, integrated into the core algorithm in 2016) specifically targets manipulative link patterns. One of its primary signals: unnatural anchor text distribution.
Based on what we've observed across penalized and recovered sites in our user base (roughly 40 sites that experienced clear Penguin-related drops between 2016 and 2024, then subsequently recovered after cleanup), the rough danger zone looks like this:
I'll caveat this heavily: these numbers come from a relatively small sample of penalized sites, and Google's thresholds are almost certainly contextual. A new brand with fewer backlinks might naturally have lower branded percentages without triggering anything. A niche with limited vocabulary (think "plumber in Dallas") might tolerate higher exact-match naturally. Use these as directional guidance, not as hard rules.
These are fundamentally different games with different rules, and conflating them is one of the most common mistakes I see.
For links pointing to your site from other domains, you mostly don't control the anchor text. And that's fine. Attempts to control external anchor text at scale (through link buying, PBNs, or aggressive outreach specifying exact phrases) are exactly what Penguin was built to detect.
What you can do:
Internal links are entirely yours. Use them strategically:
If 40% of your backlinks use the exact phrase you're trying to rank for, you have a problem. This was the primary link building strategy from 2005-2012, and it still works in the short term, which is why people still do it. But Penguin catches up. The recovery process after a Penguin-related demotion takes 6-18 months in my experience, and there's no guarantee you'll return to previous positions. I've seen sites that never recovered. Not worth the gamble.
The opposite problem, and frankly the more common one in 2026. Sites built with CMS templates that auto-generate "Read more →" links on every blog post are throwing away one of the cheapest ranking signals available. Every "read more" link could instead be "how to fix orphan pages" or "our complete guide to content silos."
This requires editorial effort (someone has to choose contextual anchor text for each link, and yes, I know that sounds like I'm pitching our tool here, but even doing it manually in a spreadsheet works). It's a one-time fix per link with ongoing ranking benefit. We've seen sites improve average position by 2-4 spots on key pages within 6 weeks of replacing their generic internal anchors. That's not a controlled experiment, so I can't claim pure causation, but the correlation is consistent enough across multiple client sites that I'd bet on it.
When an image is wrapped in a link, the image's alt attribute becomes the anchor text. Many sites have linked logos, linked thumbnails, and linked diagrams with empty or generic alt text. A linked image with alt="" is the visual equivalent of "click here." If the image links to your pricing page, the alt should be something like "SEOJuice pricing plans" rather than "logo" or blank.
This one bit us personally. We had our logo in the header linking back to the homepage with alt="logo" for over a year before someone flagged it. That's hundreds of internal links passing "logo" as anchor text to our homepage instead of our brand name. Embarrassing for a company that sells SEO tools.
This is subtler and harder to catch with automated tools. Using anchor text that describes something different from what the target page actually delivers. Example: linking with the text "anchor text best practices" to a page that's really about general link building. Users feel misled, bounce rates increase, and Google's engagement signals pick up the disconnect. The anchor text is a promise; the target page must deliver on it.
Here's the practical workflow I'd recommend running quarterly:
The internal audit is usually more actionable than the external one. You can fix every internal anchor in a single afternoon (seriously, it's mostly find-and-replace on "click here" and "read more"). External anchors require months of outreach, content marketing, and patience.
Anchor text is the visible, clickable text in a hyperlink. It tells both users and search engines what the linked page is about. Google has used anchor text as a ranking signal since 1998; their original PageRank paper specifically identified it as a strong relevance indicator. In practice, the anchor text pointing to your pages influences which queries those pages rank for.
For internal links: descriptive, keyword-relevant text that accurately represents the target page. "Content decay detection" is better than "click here." For external backlinks: a natural mix dominated by branded anchors (38-45%), with partial-match and exact-match anchors as a smaller portion (15-22% and 3-8% respectively, based on our audit data from top-ranking sites). There's no single "best" (though if I had to pick one rule, it would be: never use "click here" for anything, ever).
Based on our analysis of ~40 penalized-and-recovered sites, exact-match anchors above 25-30% of your total backlink profile enter risky territory. Any single anchor phrase above 15% is unusual for naturally-acquired links. But context matters — new sites naturally have less diverse profiles, and niche topics may have limited ways to describe them. Use these as directional benchmarks, not hard thresholds.
Yes. The 2024 Google API documentation leak confirmed that anchor text remains an active signal in their ranking systems. Cyrus Shepard at Zyppy published a detailed analysis of the leaked attributes showing multiple anchor-text-related scoring fields. However, its relative weight has likely decreased since 2012 as Google has added hundreds of other signals (user engagement, content quality, E-E-A-T indicators). Anchor text is one signal among many: necessary but not sufficient for rankings.
No. Using identical anchor text across all internal links to the same page looks templated and wastes the opportunity to signal relevance for multiple related queries. If your target page is about "content silos for SEO," vary your internal anchors: "content silos," "organizing your content into silos," "silo structure for SEO," "topical clustering strategy." Each variation reinforces a slightly different keyword cluster.
Anchor text is one of those fundamentals that most SEO guides either oversimplify ("use keywords!") or overcomplicate ("maintain a precise 40/25/20/10/5 ratio at all times"). The reality from our data: for external links, aim for a natural profile dominated by branded mentions. For internal links, be descriptive and varied. The single biggest win for most sites isn't a sophisticated anchor text strategy. It's replacing the "click here" links they already have with something that actually tells Google what's on the other end.
If you want to see what your internal anchor text situation looks like, our free internal link finder shows you every generic anchor on your site along with contextual suggestions for what to replace them with. Most sites have 20-50 quick wins hiding in plain sight.
Related reading:
no credit card required
