Anchor Text Analyzer
Paste the anchors pointing at a page and see the distribution: branded, naked, generic, partial and exact match, with a warning when the mix stops looking natural.
Export the anchors column from any backlink tool and paste it in. Nothing leaves your browser.
The five anchor text types
Every anchor falls into one of a handful of buckets, and the ratio between them is what a reviewer reads first. This is how the tool classifies them.
| Type | What it looks like | Healthy share |
|---|---|---|
| Branded | Your brand name, on its own or with a word or two. | 30–60% |
| Naked URL | The raw address, linked as text. | 10–30% |
| Generic | "click here", "read more", "this website". | 5–25% |
| Partial match | Contains the keyword inside a longer phrase. | 5–25% |
| Exact match | The target keyword and nothing else. The risky one. | 0–10% |
| Image / empty | No anchor text — an image link or a bare element. | 0–20% |
| Other | Everything that fits none of the above. | 0–40% |
How to read your distribution
- 1
Look at exact match first. Above roughly 10% is the number that gets profiles flagged. It is the single most diagnostic figure on the page.
- 2
Then check branded. A profile under 30% branded is unusual for a real brand, because most people linking to you will simply write your name.
- 3
Look for one phrase dominating. A single anchor used in a quarter of your links is a stronger pattern than the type mix, and it is the thing manual reviewers spot immediately.
- 4
Compare against a competitor who ranks. Export their anchors, paste them in, and use their ratio as the benchmark instead of a rule of thumb from a blog post.
- 5
Fix by adding, not removing. Disavowing links to change a ratio is almost always the wrong move. Build branded and naked-URL links until the exact-match share dilutes on its own.
Why the ratio matters more than any single link
No individual anchor gets a page penalised. What gets noticed is a distribution that could not have happened by accident. When a hundred different sites link to you, the anchors vary wildly, because a hundred different writers phrased it a hundred different ways. Half of them just used your name. Several pasted the URL. A few wrote "this guide" or "according to".
When links are bought or negotiated, the variance collapses. Everyone was handed the same phrase. That flatness is the signal, not the phrase itself, which is why this tool warns on concentration as well as on type share.
Fixing an over-optimised profile
The instinct is to disavow. Resist it. Disavowing is for links you would be embarrassed to explain, not for links whose anchor text you would rather change, and it removes whatever value those links were passing along with the risk.
The reliable fix is dilution. Build branded links, get naked-URL citations, earn mentions from people who will phrase it their own way. A profile at 25% exact match does not need those links removed, it needs three times as many links that are not exact match. The suggestions the tool produces above are aimed at exactly the buckets your profile is short on.
Frequently asked questions
What is a healthy anchor text ratio?+
There is no official number, but profiles that survive scrutiny tend to sit around 30 to 60% branded, 10 to 30% naked URLs, and under 10% exact match. The reason is not a rule in an algorithm, it is that this is what happens when people link to you because they want to, rather than because you asked them to use a phrase.
What counts as an exact-match anchor?+
An anchor that is your target keyword and nothing else. "link building tools" pointing at your link building tools page is exact match. "the best link building tools we tested" is partial match, which is far safer and reads more naturally in a sentence.
Is over-optimised anchor text still a real risk?+
Yes. Anchor text remains one of the clearest signals that links were built rather than earned, and Google's link-spam systems are explicitly built to detect it. The risk is not that one exact-match anchor hurts you, it is that a page with 40% exact-match anchors looks like nothing that occurs naturally.
Should I ask for specific anchor text in outreach?+
Asking for an exact-match keyword is the fastest way to build the pattern you are trying to avoid, and most good editors will refuse anyway. Ask for the link, suggest the brand name or the article title, and let the mix stay varied on its own.
Where do I get my anchor list from?+
Any backlink tool exports it: in Ahrefs it is the Anchors report, in Search Console it is Links, and Moz and Semrush have equivalents. Export the anchors column and paste it here. This tool does not fetch links itself, which is why it stays free and instant.
Does the tool send my data anywhere?+
No. Classification runs in your browser. The CSV export is generated locally too.
Fix the mix by earning better links.
An over-optimised profile is usually a symptom of buying anchors instead of earning links. LinkForce is where teams find relevant prospects and run outreach that gets editorial placements.
Start free