Anchor Text Ratio: What a Natural Backlink Profile Actually Looks Like
Anchor text ratio is the percentage distribution of anchor text types — branded, naked URL, generic, partial-match, and exact-match — across the backlinks pointing to a page. A healthy profile is dominated by branded and natural anchors, with exact-match keyword anchors kept under roughly 5% of referring domains for most pages.
Google reads this distribution as a signal: a profile where 40% of anchors repeat the same money keyword doesn't happen naturally, and Google has known that since Penguin in 2012. This guide covers the five anchor categories, realistic ranges by page type, whether ratios still matter, how to audit your profile, and what over-optimization looks like.
TL;DR:
- Branded and URL anchors should dominate most profiles; keep exact-match under ~5% of referring domains on commercial pages.
- Safe ranges differ by page type — homepages run 60–80% branded, blog posts tolerate 25–45% partial-match.
- There is no universal ratio. The reliable benchmark is the anchor profiles of pages already ranking for your keyword.
- Audit by referring domains, not raw link counts — one sitewide footer link can fake thousands of identical anchors.
- Since Penguin 4.0, over-optimized anchors are mostly devalued silently: the links just stop passing value.
The five anchor text categories
Every backlink anchor falls into one of these buckets. Get comfortable classifying them, because every audit and every link building decision comes back to this.
Branded anchors use your company or website name: "LinkVetted," "according to LinkVetted." These dominate natural profiles because when people reference a site without an SEO agenda, they use its name. Branded anchors are the safest category — you almost can't have too many.
URL (naked) anchors are the raw address: "linkvetted.com" or "https://linkvetted.com/pricing/." These show up in citations, directories, forum posts, and press mentions. Like branded anchors, they signal that a link was placed by someone referencing you, not optimizing for you.
Generic anchors carry no keyword meaning: "click here," "this article," "read more," "source," "website." They look useless from a relevance standpoint, but a profile with zero generic anchors looks curated — real people link lazily.
Partial-match anchors contain your target keyword inside a longer phrase: "this guide to anchor text distribution" or "a solid link building agency we've used." This is where most of your relevance signal should come from, because partial matches pass topical context without the pattern-matching risk of exact repetition.
Exact-match anchors are the target keyword and nothing else: "anchor text ratio," "buy backlinks," "dental implants chicago." These are the most powerful per-link relevance signal and by far the most dangerous at scale. Exact-match abuse is the specific behavior Penguin was built to catch.
Some SEOs split these further — brand + keyword hybrids, image anchors (which pass the image's alt text, or count as empty), natural-phrase anchors with no keyword at all. For auditing purposes, the five buckets above cover what matters; fold empty and image anchors into generic.
Why does Google care about anchor text ratios?
Anchor text has been a ranking input since before Google was a company. The 1998 paper that described Google's architecture, "The Anatomy of a Large-Scale Hypertextual Web Search Engine," states that link text is treated "in a special way" — Google associates the words in a link with the page it points to. That's why anchors move rankings, and why SEOs immediately started stuffing keywords into them.
The countermeasures arrived in stages. In 2005, Googlers publicly discussed using statistical analysis to spot unnatural linking patterns — the moment "anchor text ratios" entered SEO vocabulary. In April 2012, the Penguin update hit sites with keyword-heavy anchor profiles hard enough that entire businesses disappeared from search overnight. And since Penguin 4.0 in 2016, Penguin runs in real time as part of the core algorithm and mostly devalues manipulative links rather than demoting the whole site.
The practical takeaway from that history: anchor distribution isn't a ranking lever you push, it's a spam pattern you avoid. Google isn't rewarding sites for hitting 62% branded anchors. It's discounting — and occasionally punishing — profiles that look statistically manufactured.
What is a good anchor text ratio?
Here's the part most articles get wrong: they hand you one universal percentage table as if a homepage and a money page attract the same kinds of links. They don't. The ranges below are starting points drawn from what unpenalized, well-ranking profiles tend to look like — treat them as guardrails, not gospel.
| Anchor type | Homepage | Money page (service/product) | Blog post |
|---|---|---|---|
| Branded | 60–80% | 30–50% | 20–40% |
| URL / naked | 10–25% | 10–20% | 5–15% |
| Generic | 5–15% | 10–20% | 10–25% |
| Partial-match | 5–15% | 15–30% | 25–45% |
| Exact-match | 0–5% | 1–5% | 1–10% |
Why the ranges differ: homepages collect brand mentions, press citations, and directory listings, so branded and naked anchors naturally dominate. Money pages get fewer organic links overall, and the links they do earn tend to be descriptive ("their link insertion service") rather than branded. Blog posts earn the most descriptive, keyword-adjacent anchors because people link to content by describing what it covers — that's why informational content can carry a higher partial-match percentage without looking manipulated.
The big caveat: natural varies by niche. A local dentist's profile looks nothing like a SaaS company's, which looks nothing like an ecommerce store's. Some niches — legal, local services — run heavily branded because most links come from citations and directories. Others — recipes, tutorials, tools — run keyword-heavy because the content itself is the brand. Before you commit to any target distribution, pull the anchor reports for the top 3–5 sites ranking for your target keywords and see what their profiles actually look like. If every competitor on page one has 8% exact-match anchors and you're holding yourself to 1%, you're not being safe, you're being underpowered. If they're all at 2% and you're at 15%, you're the outlier Google's algorithms are built to flag.
Competitor anchor data is also the honest answer to "what ratio should I use?" — because there is no universal number, only what's normal for the SERP you're trying to win.
Do anchor text ratios still matter?
Some credible voices say no — or at least, far less than the spreadsheet crowd believes. Search Engine Journal has argued that obsessing over ratios is 2005-era thinking, and that if your links come from genuinely natural sources, ratios take care of themselves. Ahrefs' own study of 384,614 pages found only weak correlations between keyword-rich anchors and rankings — exact-match anchors correlated at roughly 0.14, which is barely a signal.
Both points are correct, and neither makes ratios irrelevant. Here's the reconciliation:
Ratios are not a ranking lever. You will not outrank anyone by tuning your branded share from 55% to 62%. The Ahrefs data shows anchor composition explains very little of who ranks.
Ratios are a risk guardrail. The moment you build links deliberately, you are choosing anchors — and repeated commercial anchors are the most machine-detectable footprint in link building. Ratio discipline isn't about impressing Google; it's about not tripping a statistical filter that's been refined for over a decade.
So the skeptics are right that chasing a "perfect" ratio wastes effort, and practitioners are right that ignoring distribution while actively building links is how pages get suppressed. Don't micro-optimize percentages; do stay well inside natural bounds.
How to audit your own anchor profile
You need a backlink index. Ahrefs is the standard — we use it for vetting every site we place links on — but Semrush and Majestic produce comparable reports.
Step 1: Pull the anchors report. In Ahrefs Site Explorer, enter your domain (or a specific page — do both), and open the Anchors report. You'll see every anchor phrase, how many referring domains use it, and how many total backlinks carry it.
Step 2: Classify by referring domains, not total links. One spammy site linking 400 times with the same anchor will wreck your percentages if you count raw links. Count each referring domain once. Sitewide footer and sidebar links are the usual distortion — a single toxic or low-quality domain can generate thousands of identical anchors.
Step 3: Bucket every anchor into the five categories. Export the report, add a category column, and tag each row. For most sites this takes under an hour. Empty anchors and image links go in their own bucket or get folded into generic.
Step 4: Run the same audit per target page. Domain-wide ratios hide page-level problems. Your homepage might be 70% branded while your main service page sits at 35% exact-match — and page-level over-optimization is where penalties and suppressions actually land.
Step 5: Compare against ranking competitors. Run the identical audit on the pages outranking you. That comparison tells you whether your next ten links should be branded (you're over-optimized), partial-match (you're under-signaled), or whether anchors aren't your problem at all and you simply need more referring domains.
Step 6: Recheck quarterly. Anchor profiles drift. New links arrive — including ones you didn't build — and a negative-SEO burst of exact-match spam is worth catching early.
What does anchor text over-optimization look like?
Over-optimized anchor profiles share recognizable fingerprints:
- Exact-match percentage that no natural process explains. If 25% of the referring domains pointing at your "best crm software" page use the anchor "best crm software," no one believes those links were editorially given — including Google.
- Money anchors from irrelevant or weak sites. A DR 8 lifestyle blog linking to a payday loan page with a commercial anchor is a purchased-link signature. Anchor risk compounds with link quality risk.
- Identical anchors across identical link types. Fifty guest posts, each with one exact-match anchor in the last paragraph, is a pattern. Real editorial links vary in phrasing, placement, and context.
- Commercial anchors pointing at commercial pages, exclusively. Natural profiles send most descriptive anchors to informational content. When money pages hoover up all the keyword anchors, the profile inverts from what organic linking produces.
- Velocity spikes in one anchor. Thirty new "buy backlinks cheap" anchors in three weeks after two years of branded links is a flag regardless of your overall percentages.
The consequence isn't always a manual action. Since Penguin 4.0, Google mostly devalues manipulative links rather than penalizing the site — which means over-optimized anchors often just stop working, silently. You've paid for links that pass nothing. In worse cases, the affected page gets suppressed for its target keyword specifically: it ranks fine for everything except the term you over-anchored, a symptom experienced SEOs recognize immediately.
Fixing it is straightforward but slow: stop building keyword anchors to the affected page, dilute with branded and URL anchors from new links, and disavow only if the profile contains genuinely spammy domains at scale.
How to use anchors when actively building links
If you're building links deliberately — guest posts, niche edits, digital PR — you control the anchor on many of your links, which is exactly why discipline matters. The links you don't control (citations, organic mentions) skew branded on their own; the links you do control are where over-optimization gets manufactured.
Practical rules we apply in our own campaigns:
- Default to branded, partial-match, or natural-phrase anchors. Save exact-match for high-authority, topically relevant placements — the links strong enough to justify the risk.
- Never repeat an exact-match anchor to the same page twice in a row. Vary phrasing every time: "anchor text ratios," "how anchor distribution works," "this anchor text guide."
- Let page type set the mix. Follow the table above: heavier keyword anchors to blog content, conservative branded/partial mixes to money pages.
- Anchor decisions come after site vetting, not before. A perfect anchor on a PBN link is still a PBN link. Distribution only matters across links that pass quality checks — which is why we show Ahrefs data for every placement site before you pay, and why every order at our pricing tiers includes anchor recommendations matched to your existing profile.
FAQ
What is a good anchor text ratio?
There's no single correct ratio. As broad guardrails: homepages skew 60–80% branded with under 5% exact-match; money pages run 30–50% branded with 1–5% exact-match; blog posts tolerate more partial-match (25–45%). The reliable method is auditing the anchor profiles of pages already ranking for your keyword and staying inside the range they establish.
How many exact-match anchors are safe?
For most pages, keep exact-match under 5% of referring domains — often lower for competitive commercial terms where Google scrutiny is highest. Some niches naturally support more. The percentage matters less than the outlier test: if your exact-match share is several times higher than every competitor ranking above you, you're exposed.
Do anchor text ratios still matter for SEO?
As a risk guardrail, yes; as a ranking lever, barely. Ahrefs' large-scale study found only weak correlation between keyword anchors and rankings, so tuning percentages won't lift you. But repeated commercial anchors remain the most detectable link building footprint, so staying inside natural bounds still protects the links you build.
Does anchor text ratio apply to internal links too?
The ratio concern is mainly about external backlinks, because those are the ones Google treats as third-party votes. Internal anchors can be descriptive and keyword-rich without meaningful risk — they're navigation you control, and consistent descriptive internal anchors help Google understand page topics.
Can I fix an over-optimized anchor profile without disavowing?
Usually, yes. Dilution is the standard fix: build new links with branded, URL, and generic anchors to the affected page until the keyword-anchor share drops back into a normal range. Disavow is only worth it when the keyword anchors come from genuinely spammy domains at scale, not from decent sites with aggressive anchors.
Want anchor recommendations based on your actual profile instead of a generic percentage chart? Get in touch — we'll pull your Ahrefs anchor data, show you where you stand against ranking competitors, and prove out every placement site before you pay for a single link.