Anchor Text Distribution Review

Anchor text fails in both directions and the two look nothing alike. Some sites never earn a single descriptive anchor and rank for nothing; others carry a third of their anchors on one commercial phrase because a vendor was paid per placement. No safe ratio has ever been published or confirmed, so anyone quoting a percentage is guessing, and the only defensible comparison is against the pages currently outranking you for the query.

The skill file

---
name: anchor-text-distribution-review
description: Use when a page has plenty of links but will not rank, or when past link buying may have left an unnatural anchor pattern.
---

# Anchor Text Distribution Review

Anchor text fails in both directions and the two look nothing alike. Some sites never earn a single descriptive anchor and rank for nothing; others carry a third of their anchors on one commercial phrase because a vendor was paid per placement. No safe ratio has ever been published or confirmed, so anyone quoting a percentage is guessing, and the only defensible comparison is against the pages currently outranking you for the query.

## What you need first

- anchor text with linking domain and target URL, deduped to one row per domain per target
- the same export for the two or three pages that outrank you for the target query
- acquisition tags from a backlink audit so you know which anchors were paid for
- internal anchor text for the same target pages, from a site crawl

## Method

1. Work per target URL, never per site. A site level profile is dominated by homepage brand links and will happily hide one page where every anchor is the same money phrase.
2. Bucket anchors into brand, bare URL, generic, descriptive partial match and exact match, and take the share of linking domains rather than of links, so a single site-wide footer does not distort the whole distribution.
3. Run the same buckets on the competing pages. If they rank at 5 percent exact match while you sit at 35, the anomaly is yours whatever any published guide claims.
4. Cross reference exact-match anchors against the acquisition tags and their first seen dates. The same share accumulated over five years from mixed sources is normal; all of it from guest posts placed in one quarter is a footprint.
5. Check internal anchors before touching anything external. A commercial page with no descriptive internal anchor anywhere on the site is the free fix, it is fully under your control, and it is common.
6. Where the pattern really is skewed by paid placements, dilute rather than remove. Earning brand and descriptive anchors moves the ratio without asking a paid site to edit a link, a request that often gets it deleted instead.
7. Re-measure only after the next 20 to 30 earned domains. Domain level ratios move slowly and a monthly check reports noise as progress.

## What this produces

A per-target-URL anchor bucket comparison against competing pages, with the skew explained by acquisition source and an explicit dilute, leave or fix-internally decision.

## Where this goes wrong

- quoting a safe exact-match percentage as if it were established fact rather than folklore
- counting links instead of linking domains, so one site-wide footer makes a normal anchor look dominant
- asking a paid placement to change its anchor text and prompting the site to remove the link entirely
- ignoring internal anchors, which are the only descriptive anchors you can add without asking anyone

---

From the QuQi skill library - https://www.quqi.io/skills/anchor-text-distribution-review

What you need first

• anchor text with linking domain and target URL, deduped to one row per domain per target
• the same export for the two or three pages that outrank you for the target query
• acquisition tags from a backlink audit so you know which anchors were paid for
• internal anchor text for the same target pages, from a site crawl

Method

1. 01 Work per target URL, never per site. A site level profile is dominated by homepage brand links and will happily hide one page where every anchor is the same money phrase.
2. 02 Bucket anchors into brand, bare URL, generic, descriptive partial match and exact match, and take the share of linking domains rather than of links, so a single site-wide footer does not distort the whole distribution.
3. 03 Run the same buckets on the competing pages. If they rank at 5 percent exact match while you sit at 35, the anomaly is yours whatever any published guide claims.
4. 04 Cross reference exact-match anchors against the acquisition tags and their first seen dates. The same share accumulated over five years from mixed sources is normal; all of it from guest posts placed in one quarter is a footprint.
5. 05 Check internal anchors before touching anything external. A commercial page with no descriptive internal anchor anywhere on the site is the free fix, it is fully under your control, and it is common.
6. 06 Where the pattern really is skewed by paid placements, dilute rather than remove. Earning brand and descriptive anchors moves the ratio without asking a paid site to edit a link, a request that often gets it deleted instead.
7. 07 Re-measure only after the next 20 to 30 earned domains. Domain level ratios move slowly and a monthly check reports noise as progress.

What this produces

A per-target-URL anchor bucket comparison against competing pages, with the skew explained by acquisition source and an explicit dilute, leave or fix-internally decision.

Where this goes wrong

• quoting a safe exact-match percentage as if it were established fact rather than folklore
• counting links instead of linking domains, so one site-wide footer makes a normal anchor look dominant
• asking a paid placement to change its anchor text and prompting the site to remove the link entirely
• ignoring internal anchors, which are the only descriptive anchors you can add without asking anyone

Use this skill in your own AI

The download is a plain markdown file with the name and trigger in its frontmatter. Where an assistant supports skills it can load itself, that frontmatter is what it reads to decide this one applies.

Questions about this skill

More in Links & authority