What it does
Paste any HTML containing anchor tags (<a href>) and this tool extracts every link, classifies it by anchor type, and shows the full distribution. You see at a glance whether your internal linking, content page, or migrated competitor analysis has anchor patterns Google would call natural — or patterns that look like over-optimization.
The tool runs entirely in your browser using the standardDOMParserAPI. The HTML you paste never leaves your device, there is no server hop, no rate limit, no signup. Paste an entire page's HTML or just a snippet of nav, body content, or an outreach pitch list.
Why anchor distribution matters
Anchor text is one of the oldest and still one of the most informative signals Google uses to understand what a page is about and whether it has been gaming the system. Two facts shape modern anchor strategy:
- Anchor distribution at the campaign level signals naturalness.Pages with mostly branded and naked-URL anchors look like organic citations. Pages with mostly exact-match anchors look like paid placements. Google's spam-detection systems weight this distribution heavily.
- The page-level mix is also a topic signal. When you internally link from one page on your site to another with descriptive anchor text, you're telling Google what the destination page is about. Generic anchors ("click here") waste this signal entirely.
The healthy distribution depends on context — a brand-new page should not look the same as a 5-year-old page; an internal navigation pattern is different from a backlink profile — but the bands of healthy ratios are well-established. This tool surfaces both the distribution and the risk flags.
How to use this analyzer
- Add your brand keyword and primary keyword. These tell the classifier what counts as "branded" vs "exact-match" vs "partial-match" for your specific case. Without them the tool can still classify naked URLs and generics, but the most important categories need this context.
- Add your domain so internal vs external links can be split correctly.
- Paste your HTML.Either "view source" on a page and paste the whole thing, or paste just the body content you want analyzed (a single article, a list of backlink anchors, etc.). Anything containing
<a href>tags works. - Read the distribution panel. Each anchor type shows count and percentage with a bar visualization. Risk flags surface automatically when ratios exceed safe bands.
- Scan the full anchor list at the bottom of the output. Each entry shows its classification, internal / external status, and nofollow flag.
Anchor type reference
- Branded.Anchor text contains your brand name. "Ankore", "Ankore's blog post", "according to Ankore". Should typically be the largest single category in any natural link profile (40–65% is common).
- Exact-match.Anchor text exactly matches your primary keyword (or its singular/plural variant). "link building", "guest posts". Should stay under 5–8% of total anchors pointing to a single page. Higher than that and Google's spam-detection systems start to flag the page.
- Partial-match.Anchor text contains the primary keyword among other words. "the best link building service", "link-building strategies that work". Generally healthy at 15–30%.
- Naked URL.Anchor text is the URL itself. "https://ankore.io", "ankore.io". Common and natural in citation contexts. 8–15% is typical.
- Generic."Click here", "learn more", "this article", "here". SEO-wasted in most cases — provides Google with zero topic signal and tanks accessibility for screen reader users. Most natural profiles have 10–20%; aim lower on internal links.
- Image. Anchor wraps an image with no text. Schema treats the image alt text as the anchor. Always set meaningful alt text on linked images.
- Empty / no text. Anchor has no text and no image alt. Always a bug. Fixes itself when you add content.
Anchor strategy best practices
- Branded anchors should dominate. 40–65% of anchors pointing to your site (across all referring pages) should mention your brand. This is the single strongest naturalness signal.
- Exact-match anchors stay under 5–8%. The specific threshold varies by niche — competitive SaaS markets tolerate slightly higher exact-match rates than less-monitored niches — but 5–8% is the universal safe ceiling.
- Internal links can be more aggressive. The 5–8% exact-match cap is for external backlinks. Your own internal linking can use more descriptive (partial-match) anchors aggressively because Google understands you control both ends.
- Strip generic anchors from internal navigation. "Click here" or "Read more" on internal links wastes the strongest topic signal you have. Use descriptive anchors for internal links.
- Track anchor distribution at the campaign level, not per-link. Each individual link does not need to fit the perfect distribution. The aggregate of every link you build over a quarter or year is what matters.
- Image alt text doubles as anchor text. When an image is wrapped in a link, Google reads the image alt attribute as the anchor. Always write meaningful alt text on linked images.
- Watch the long tail. A page with 100 backlinks where 95 are branded and 5 are exact-match is healthy. A page with 10 backlinks where 2 are exact-match (20%) is over-optimized. Distribution math is more sensitive on small samples.
Common anchor mistakes
- Stuffing exact-match into outreach campaigns. The most common single anchor mistake. Cheap link-building providers default to exact-match because clients (incorrectly) think it gives the strongest ranking signal. It actually triggers spam filters. Pages that ranked at #5 drop to #20+ after exact-match concentration crosses ~10%.
- Generic anchors on internal links. "Read this guide" on a link inside your blog post wastes the highest-value internal signal you have. Use the destination page's topic as the anchor.
- Empty anchors on logo links. Common bug:
<a href="/"><img src="logo"></a>with no alt text. Both an accessibility violation and a wasted top-of-page anchor signal. - Hyperlinked phrases that mean nothing. "Click here to learn more" underlined as one anchor tells Google the destination is about "learn more". That is not a topic.
- Over-correcting toward branded. A profile that is 100% branded looks artificial too — natural profiles include some partial-match, naked URL, and even occasional exact-match. The distribution should be diverse, not single-category.
Frequently asked questions
Does this analyze backlinks pointing to my site?
No — this tool analyzes anchor text within HTML you provide. To analyze backlinks pointing TO your site (the more common use case), you need a backlink data source (Ahrefs, Semrush, Majestic) that maintains a crawl of the entire web. That requires paid data. This tool is for analyzing on-page link patterns from any HTML you can access.
What's the ideal anchor text distribution?
Across all backlinks pointing to a page: roughly 40–65% branded, 15–30% partial-match, 8–15% naked URL, 10–20% generic, and under 5–8% exact-match. Internal linking distributions can be more aggressive on partial-match since you control both ends.
How is 'branded' classified?
If the anchor text contains the brand keyword you provide (case-insensitive). 'According to Ankore' counts; 'link building agency' does not (unless you provide 'link building' as the primary keyword, in which case it would be partial-match).
How is 'exact-match' classified?
If the anchor text equals the primary keyword you provide, or its singular/plural form. 'link building' = exact-match if your keyword is 'link building'. 'best link building services' = partial-match. 'link-building' (with hyphen) is also matched if the keyword has the same form.
What happens if I don't fill in the brand or keyword?
The classifier still works for naked URLs, generic anchors, and image-wrapped links — but the most important categories (branded, exact, partial-match) won't classify correctly. For meaningful output, provide both a brand keyword and a primary keyword.
Will this work for non-HTML input?
Only HTML. The tool uses DOMParser to find <a> tags. Markdown links won't parse. If you have markdown, paste it through a markdown renderer first, then paste the HTML output here.
Why is the exact-match risk band 5–8% specifically?
Multiple peer-reviewed studies of de-indexed and penalized sites have found exact-match concentrations above 8% strongly correlate with manual actions. 5% is the conservative ceiling we recommend; 8% is the upper edge of the 'usually safe' band. Above 10% and the risk becomes material.
Will this tool log my pasted HTML?
No. Everything runs in your browser via DOMParser. The HTML never leaves your device — there is no server, no logging, no analytics on your input.