Anchor Text Ratios That Look Natural to Google in 2026
Ask five SEOs for the "correct" anchor text ratio and you'll get five different spreadsheets, most of them presented with far more confidence than the underlying data supports. Anchor text distribution isn't a formula you hit -- it's a pattern that either looks like something a real audience produced organically, or looks like something a link builder assembled on purpose. Here's how to think about it without over-engineering a number that was never meant to be exact.
Why ratios matter at all
When link building was younger, exact-match anchor text was one of the strongest ranking signals available, and it got abused accordingly -- sites built link profiles that were 80% or more exact-match keyword anchors, because that's what worked. Search engines adjusted, and heavily skewed exact-match anchor profiles became one of the clearer tells of manipulation rather than a boost.
What "natural" looks like instead is a profile dominated by branded mentions and generic phrases, with exact-match keyword anchors as a modest minority. That's simply what happens when real people link to things: they use your brand name, they use the page title, they write "check this out," they paste the bare URL. Nobody writing a genuine recommendation thinks in exact-match SEO phrases.
Indicative ranges by anchor type
These ranges are directional starting points for evaluating your overall profile, not a checklist to engineer link by link. Treat them as a sanity check, not a target.
| Anchor type | Indicative safe range (share of total anchors) | Notes |
|---|---|---|
| Branded (your company/site name) | 40-60% | Should typically be your largest single category |
| Naked URL (yourdomain.com or full URL) | 10-20% | Very common in natural profiles, especially from citations and mentions |
| Generic ("click here," "this article," "read more") | 10-20% | Normal in editorial writing where the writer isn't thinking about your keywords at all |
| Partial-match (keyword phrase variations, not exact) | 5-15% | Reasonable and generally low-risk |
| Exact-match keyword | 1-7% | The category to keep smallest and most conservative |
| Image/alt-text anchors | Varies, typically low single digits | Often overlooked in manual ratio calculations |
If your profile is inverted from this shape -- exact-match keyword anchors making up a large share, with your own brand name a minor player -- that's the pattern worth addressing, regardless of the exact numbers you land on.
If you'd rather have this done than DIY it, the placement service runs the same vetting this article describes.
Get link placementsWhy "indicative" is doing a lot of work in that table
These ranges will not be identical for a two-year-old SaaS brand with a distinctive name versus a newer site with a generic, keyword-heavy domain name. A site literally named after its main keyword is going to have naturally higher partial-match anchor rates just from people typing its name normally, and that's fine -- it's a function of the brand, not a sign of manipulation. The point of the table isn't to hit these numbers precisely. It's to notice when you're wildly outside them without a good reason.
The mistake that ratio-chasing creates
Ironically, obsessing over anchor ratios often creates the exact manipulated pattern you're trying to avoid. A link builder who tracks their ratio spreadsheet and deliberately assigns anchor text to hit a target percentage is still manufacturing the pattern -- just a more sophisticated-looking one. A genuinely natural profile isn't the result of hitting a target ratio; it's the byproduct of getting links from a range of sources that each had their own reason to link, phrased in their own words.
The practical implication: stop assigning anchor text as the first step of outreach. Let editors and site owners write the sentence around your link in language that fits their content. If you're placing links yourself -- on your own guest posts, for instance -- default to branded or natural phrasing, and reserve keyword anchors for the minority of placements where it genuinely reads naturally in context.
How to actually audit your own profile
Pull your full backlink list from whatever tool you already use, export the anchor text column, and bucket each entry into the six categories from the table above. Don't overthink borderline cases -- if an anchor could plausibly fit two categories, pick the one a neutral reader would call it first. What you're looking for isn't precision, it's a shape: does branded dominate, is exact-match a small minority, does the rest spread across naked URLs and generic phrases in a way that doesn't look engineered.
Do this exercise once a quarter, not once a week. Anchor ratios shift slowly under normal link acquisition, and checking too often just invites the same tinkering urge that creates unnatural patterns in the first place. If a quarterly check shows the ratio drifting hard toward exact-match, that's your cue to diversify sourcing, not to manually rebalance old links -- there's no clean way to retroactively "fix" anchor text on links you don't control, and trying usually means asking site owners for edits, which draws more attention to the link than leaving it alone would.
A word on new versus established sites
A brand-new site with only a handful of backlinks doesn't have enough data for ratio analysis to mean anything -- five links is a sample size, not a pattern. Ratio thinking becomes useful once you're in the dozens-to-hundreds range, where an actual distribution starts to emerge. Don't let ratio anxiety talk you into avoiding exact-match anchors entirely in the early days out of overcaution; a new site with zero keyword anchors ever is its own (much milder) tell, since real recommendations do occasionally use descriptive language that happens to match a keyword.
My honest opinion
Most anchor-ratio guidance online is more precise-sounding than the underlying reality justifies -- tidy percentages presented as if they were extracted from an algorithm, when they're really just pattern-matching on what natural link profiles tend to look like. Use the ranges above as a gut check, not a spreadsheet formula. If your profile roughly resembles what an organically popular, non-SEO-obsessed site's profile would look like, you're almost certainly fine. If you've built a system to hit precise percentages across hundreds of links, you've built a more elaborate version of the exact problem this whole exercise is meant to avoid.
If you want a second opinion on whether your current anchor distribution looks natural or looks engineered, send us the details through the lead form -- it's a quick read for anyone who's looked at enough link profiles to spot the difference.
Frequently asked questions
Is there an exact 'safe' percentage for exact-match anchor text?
No, and treat any source that gives you one precise number with suspicion. Natural link profiles vary by niche, brand size, and content type. The ranges here are indicative starting points for thinking about balance, not a formula to hit exactly.
Why do branded anchors dominate natural link profiles?
Because most people linking to you organically don't know or care about your target keywords -- they just write your brand name, or link the raw URL, or use generic phrases like 'this article' or 'read more.' A link profile with far more exact-match keyword anchors than brand mentions almost always signals manufactured links.
Should I request specific anchor text when I get a guest post or placement?
Occasionally, and lightly. It's more natural to let the publishing site's editor choose wording that fits their sentence, sometimes suggesting a soft option, and to save any specific keyword anchor requests for a small minority of your placements.
What happens if my anchor text ratio looks unnatural?
There's no single dramatic penalty tied to anchor ratios alone -- it's typically one signal among several that search engines weigh when assessing whether a link profile looks manipulated. An unnatural ratio combined with other red flags (low-quality linking sites, sudden volume spikes) is a much bigger risk than an unnatural ratio in isolation.