What does a healthy anchor-text distribution look like for an investor's backlink profile?

REI Spark is a B2B SEO platform run by a licensed California real estate agent (DRE #02006033). The operator has tuned anchor profiles across cash-buyer campaigns from 2018 to 2026 — campaigns that triggered manual review for over-optimization, campaigns that recovered after rebalancing, and a current Yuba Home Buyer profile audited live for this article. Most anchor-distribution advice online was written for SaaS and e-commerce sites with hundreds of thousands of monthly visitors. Cash-buyer sites operate at a fraction of that link supply for their target queries. The risk is higher. This post walks the framework after the vendor and site selection step before any anchor decision and within the entity-based SEO framing for investor sites.

What is anchor-text distribution?

Anchor-text distribution is the percentage breakdown of words and phrases other sites use to link to a given page. Search engines model link profiles as patterns, not individual links, so the distribution matters more than any single anchor a buyer chooses.

Five anchor categories cover the natural language of how the web links to a business. Branded anchors contain the company name or a clear variant — "Yuba Home Buyer," "YHB," "yubahomebuyer." Naked URL anchors are the bare domain or full URL — "yubahomebuyer.com," "https://yubahomebuyer.com." Generic anchors are non-descriptive linking phrases — "click here," "visit website," "this site." Partial-match anchors include some target keywords without the exact phrase — "we buy houses in California," "cash for inherited homes." Exact-match anchors reproduce a target query word-for-word — "we buy houses Yuba City," "sell my house fast Marysville."

Distribution beats any single anchor because search engines do not evaluate links individually. They evaluate the pattern a site's anchor profile forms. A profile with one exact-match anchor among fifty branded anchors looks unmanipulated. A profile with twenty exact-match anchors among thirty other anchors looks bought, regardless of where each individual link sits on the donor side.

The math runs at two levels. The domain-level view aggregates every external anchor pointing anywhere on the site. The URL-level view isolates a single page. Both matter. A homepage that looks healthy at the domain level can hide a service page that has been pushed past the natural ratio. The URL view surfaces what the domain view averages away. Distribution analysis becomes meaningful once a profile reaches roughly one hundred referring domains — below that, percentages move with each new link and the signal is noisy.

What ratio do natural investor sites have?

A natural investor-site profile typically shows branded anchors 40–55%, naked URLs 15–25%, generic anchors 10–20%, partial-match 10–15%, and exact-match 1–5%. These ratios are descriptive of unmanipulated profiles, not a published Google threshold.

Branded carries the largest share because real-world citations — directories, forum mentions, news pickups, business listings — overwhelmingly use the brand name. Naked URLs sit second because copy-paste links from emails, social posts, and citation aggregators preserve the URL verbatim. Generic anchors come from forum signatures, footer links, and "visit our partner" placements. Partial-match anchors emerge from editorial mentions where a writer describes what the business does without quoting it. Exact-match anchors occur naturally only in narrow contexts — a comparison article listing direct competitors, a glossary linking to a defining example.

Investor sites differ from e-commerce because the natural link supply is structurally smaller. An e-commerce site selling running shoes has thousands of bloggers, comparison sites, deal aggregators, and review platforms naturally generating links across the full anchor spectrum. A cash-buyer site operating in Yuba County has a few dozen local citations, occasional press mentions, real estate forum threads, and not much else. The same five-category distribution applies, but the absolute volume is two or three orders of magnitude lower. That changes how visible any single anchor becomes.

The 1–5% exact-match figure comes from observed natural distributions across investor-niche profiles, not a published Google threshold. There is no Google announcement that says "exact-match must stay under five percent." The number is derived from looking at hundreds of unmanipulated investor profiles and noting where the natural ceiling lands. Treating it as a hard rule is wrong. Treating it as a descriptive benchmark — this is what unmanipulated profiles look like — is right. The framework breaks if a buyer reads the percentages as a target to engineer toward rather than a pattern to preserve.

How do exact-match anchors hurt local cash-buyer sites?

Exact-match anchors hurt local cash-buyer sites because city-modifier queries have small natural link supply. A handful of bought exact-match anchors push the profile past the natural ratio, triggering classifier devaluation that silently ignores the new links.

City-modifier queries — "we buy houses Yuba City," "cash home buyer Marysville," "sell my house fast Sacramento" — have small natural link supply because the universe of pages that would naturally use that exact phrase is small. A national e-commerce term has thousands of legitimate sources. A local service-plus-city phrase has dozens. When a buyer purchases ten exact-match anchors of a phrase that has fifteen natural exact-match anchors in the entire link graph, the profile crosses the unmanipulated boundary visibly.

Classifier devaluation behaves like the bought link was never acquired. There is no warning, no GSC notification, no manual action message. This is post-Penguin classifier behavior — quiet, automatic, and never surfaced in any operator-facing tool. The classifier reads the anchor pattern as engineered, drops the new links from the rank-influencing signal pool, and the buyer sees nothing happen. Money was spent. Links were placed. The site did not move. This is the framing that matters: over-optimization is not a "penalty risk" — it is wasted spend, plus a profile that now needs rebalancing before further link acquisition will work.

Three signals surface the problem before it compounds. Ranking drops on the city-modifier queries themselves — the exact phrase the anchors targeted starts losing positions while branded queries hold steady. GSC impressions for the exact-match phrase decline week-over-week while broader semantic variants remain flat. Ahrefs anchor share above ten percent on any single URL within ninety days — the velocity threshold above which the classifier weighting flips visibly negative. Any one signal warrants a pause on further exact-match acquisition. Two of three warrants an active rebalance toward branded and naked-URL anchors.

Service-plus-city exact-match anchors are the highest-risk anchor type for cash-buyer sites. The combination is what triggers the pattern detection — service alone or city alone reads naturally; the two welded together reads commercial.

How do you blend city and service anchors?

Blend by buying three branded and naked-URL anchors before any partial-match, never letting service-plus-city exact phrases exceed five percent in a ninety-day window. City-only and service-only partial-match anchors carry relevance signal without exact-match risk.

Anchor types break into four behavioral groups for blending purposes. Branded carries no relevance signal but anchors the profile in unmanipulated territory. City-only anchors — sometimes called local city anchors — like "Yuba County homeowners," "homes in Marysville" — carry geographic context without service intent. Service-only anchors — "cash home buyer," "we buy houses" — carry service intent without geographic targeting. Service-plus-city is the combination that reads commercial, the category that triggers the pattern detection.

The operator's working rule across YHB, FHBC, and SMHFICA is three branded or naked-URL anchors before any partial-match anchor, and exact-match capped under five percent of acquisitions in any ninety-day window. The three-to-one ratio rebuilds the natural floor before any commercial-leaning anchor enters the profile. The ninety-day cap prevents velocity clustering — twenty exact-match anchors spread across a year reads differently than the same twenty placed in a single quarter.

Two-tier link contexts preserve relevance signal without exact-match risk. The donor page provides topical context — a Yuba real estate market analysis page, a California cash-buyer comparison article — and the link anchor uses a city-only or service-only partial-match anchor. The anchor is not exact-match, but the contextual proximity of Yuba on the donor page near a cash buyer anchor passes equivalent relevance signal to the classifier without crossing the over-optimization line. This is the working pattern across REI Spark's wholesale publishing network for cash-buyer profiles entering the rebalance phase.

The five-percent ninety-day cap is a soft floor that holds across operator-controlled profiles. The cap exists not because Google enforces it, but because the cap reverse-engineers the natural ratio that unmanipulated profiles exhibit. Buying anchors that reproduce the natural ratio reads natural to the classifier. That is the entire framework.

What does an over-optimized profile look like in Ahrefs?

An over-optimized profile in Ahrefs shows clustered exact-match velocity in the last ninety days, an exact-match share above ten percent on any single URL, or a partial-match share that overtakes branded. The Anchors-New filter surfaces this earliest.

The audit runs in three steps. Step one: open Site Explorer and navigate to the Anchors view. Sort by referring domains descending. Note the percentage breakdown across the five categories. Healthy profiles show branded as the dominant share. Trouble surfaces when partial-match or exact-match approach branded — that is the pattern shift the classifier reads.

Step two: apply the New filter for the last ninety days. This is where over-optimization velocity surfaces earliest. Cumulative anchor counts at the domain level move slowly. A profile that has accumulated forty exact-match anchors over five years reads differently than a profile that has accumulated twenty exact-match anchors over the last quarter. The cumulative view averages the velocity away. The New filter exposes it directly. If the last ninety days show clustered exact-match acquisitions on a single phrase or a single donor source, the profile is in over-optimization territory regardless of what the cumulative view shows.

Step three: drill into Pages > Best by links and audit URL-level anchor distribution. URL-level over-optimization is invisible at the domain level. A homepage that runs branded forty-five percent across the whole site can mask a single service page running exact-match thirty percent. The classifier evaluates per-URL link patterns, not just domain patterns. A page-level breach shows in Pages > Best by links by clicking into the high-anchor URLs and inspecting the per-URL Anchors view.

Remediation when an audit surfaces over-optimization moves toward branded and naked-URL anchors as the safest acquisition type. The three-to-one rule applies in reverse — three branded or naked-URL anchors per existing exact-match before any new partial-match enters the profile. The rebalance takes ninety days minimum to surface in classifier behavior; faster expectations create wasted spend on top of wasted spend.

What does Yuba Home Buyer's anchor distribution look like in May 2026?

Yuba Home Buyer's Ahrefs profile in May 2026 shows branded near nine percent, naked URL thirty-eight, generic twenty-six, partial-match twenty-six, and exact-match two percent. Those figures were calculated only after stripping forty-two percent of referring domains as parasitic spam first.

The data is from a live Ahrefs audit run on May 3, 2026 against yubahomebuyer.com. The operator owns the site, has logged every paid anchor acquisition since 2018, and pulled the anchor profile for this article. The findings are operator-verifiable.

The audit started with eighty referring domains across thirty-five distinct anchors. Before computing distribution percentages, the audit filtered forty-two percent of referring domains as parasitic spam — fake-testimonial syndication networks pushing third-party SEO services, Telegram bot spam, signature spam from SEO sellers stamping the YHB domain into their own promotional anchor text. Most aging investor domains accumulate this drip. The classifier still evaluates the noise floor, but the operator's distribution percentages have to be computed against operator-influenced anchors only. Otherwise the spam ratio drowns the signal the operator can act on.

The post-filter distribution sits at branded near nine percent, naked URL thirty-eight, generic twenty-six, partial-match twenty-six, and exact-match two percent. Three readings of those numbers matter.

Branded is under-developed at nine percent against the forty-to-fifty-five natural range. That is the operator's next-cycle priority — building out branded acquisitions through citations, forum mentions, and editorial pickups before any further commercial-leaning anchor lands.

Exact-match holds at two percent against the one-to-five natural range. The discipline this article describes — three branded or naked-URL acquisitions before any partial-match, ninety-day cap on service-plus-city exact-match — has been running on YHB since 2018. Eight years of caps produced an exact-match share that sits cleanly inside the unmanipulated boundary. That is the operational proof. The discipline works because the discipline ran for eight years.

The forty-two percent noise floor is normal. Most cash-buyer operators running for five-plus years carry a similar parasitic share. The audit method matters more than the cleanup — the cleanup is a one-time disavow file pass, the audit cadence is bi-weekly. Anchors > New filtered to the last thirty days surfaces incoming spam fast enough to stay ahead of cumulative pollution.

The action: pull your own anchor distribution from Ahrefs or Webmaster Tools. Categorize the top fifty into the five buckets. Filter parasitic spam first — the noise floor distorts every percentage you'd compute against the raw data. Read the post-filter distribution against the natural ranges, identify which category is under-developed, and build the next ninety days of acquisitions against the gap. Return upstream to the vendor-vetting prerequisite for the site-selection layer that runs before any anchor decision, and to the entity-based SEO root post for the topical-authority framework. REI Spark's vetted link marketplace for real estate investors operates with the three-to-one branded-to-partial-match rule built into the recommended-mix logic.

By YK Kuliev, California DRE #02006033 — operating cash-buyer brand sites since 2018, REI Spark since 2025.