Link building, scored against the algorithms that evaluate it.
An inbound link is a vector with several axes the algorithm reads separately. The natural baseline for an inbound anchor profile runs 70 percent or more branded, URL, topic, and natural-variation anchors. Exact-match commercial anchors are evaluated per URL in real time. Link velocity gets compared against measured decay. SpamBrain reads the link-graph network. The methodology that survives the next algorithm update routes through all four signals.
Four signals the off-page profile is scored against.
The algorithmic detection layer compounds every update. Penguin 4.0 evaluates anchor distribution at the URL level in real time. SpamBrain reads the link-graph network at network density. TrustRank scores proximity to a verified seed set. Link velocity is read against measured decay. We scope every engagement against all four.
Link quality is a vector, not a number.
A single inbound link carries weight on multiple axes: source domain authority, source topical relevance, anchor text, link position in the source page, dofollow status, surrounding co-citation context, and the link-graph neighborhood of the source. The Domain Rating metric collapses these into one number for tool convenience. The algorithm evaluates them separately. A DR-30 link from a topically-coherent vertical publication often outperforms a DR-60 link from a generic news aggregator, because co-citation with peer brands inside the same topic cluster reinforces topical authority surfaced in the May 2024 Content Warehouse documentation leak.
Anchor-text distribution is scored per URL.
The September 2016 integration of Penguin 4.0 into Google's core algorithm shifted anchor-text evaluation to a real-time, URL-specific basis. A natural inbound anchor profile across a domain runs 70 percent or more branded, URL, topic, and natural-variation anchors. Commercial-intent exact-match anchors run between 5 and 15 percent. The Penguin discount triggers when commercial exact-match concentration on a specific URL exceeds the per-vertical baseline, even when the root domain looks natural. The taxonomy breaks into exact match, partial match, branded, naked link, generic, and image anchors where the alt attribute serves as the anchor text.
Link velocity gets compared against decay.
Google evaluates link velocity to distinguish natural accumulation from manipulation. Sustained velocity from topically-relevant editorial sources reads as the natural pattern. Sudden spikes from low-quality sources trigger algorithmic discounting via Penguin and SpamBrain. The other side of the velocity equation is decay: acquired links exhibit a natural 10 to 20 percent annual attrition rate as publishers edit pages, remove content, or flip dofollow to nofollow. A retainer that doesn't model placements against this attrition leaks signal over time. The scope target is net of decay, not gross delivery.
SpamBrain reads the link-graph network, not just the destination.
SpamBrain is a neural detection system Google confirmed in 2022. It reads link schemes from the link-graph network rather than just the destination page quality. A campaign that pitches the same vertical from the same outreach mailbox across 80 publishers in 30 days fingerprints as scheme-coordinated regardless of placement quality. The March 2024 spam update intensified SpamBrain-driven enforcement. Surviving the next algorithm update means the campaign architecture renders the same fingerprint as organic editorial outreach: mailbox identity segmented per campaign cohort, outreach paced against the 30-day moving average the link-graph evaluates over, and prospect networks vetted against TrustRank seed-set proximity.
The natural baseline we scope to.
Penguin 4.0 evaluates anchor-text distribution at the URL level in real time. A single page accumulating exact-match commercial anchors past the per-vertical baseline gets those anchors discounted the next time Google crawls them, even if the root domain looks natural. We allocate anchor text across the link profile to keep each landing page inside the natural baseline.
Methodology questions we get during the audit conversation.
What is link building?
Link building is the proactive acquisition of inbound links and brand mentions from third-party domains. It sits inside the broader off-page SEO services surface alongside citations, digital PR, and brand-mention work. The work is editorial outreach to publishers whose audience overlaps the target site's topical cluster. The placement either lands on the merits of the asset and the pitch, or it doesn't. White-hat link acquisition operates on manual outreach and relationship-led placement; it does not operate on rented inventory, private-network footprint, or any other link-scheme apparatus.
Why does anchor text matter so much?
Anchor text is the load-bearing signal Google's Penguin algorithm scores. Penguin 4.0 evaluates anchor-text distribution at the URL level in real time. A landing page that accumulates exact-match commercial anchors past the per-vertical baseline gets those anchors discounted the next time Google crawls them. The natural baseline runs 40 to 55 percent branded, 15 to 25 percent topic, 10 to 15 percent URL, 5 to 10 percent generic, and 5 to 15 percent exact-match. Outreach copy and placement requests get allocated to keep each landing page inside this baseline at the URL level, not just the root-domain average.
What separates link building from link buying?
The distinction is editorial choice vs. paid placement. Editorial placement is a publisher choosing to link to a page because the content fits an editorial decision and no payment changes hands. Paid placement requires the rel="sponsored" attribute on the link (Google's 2019-09-10 introduction, treated as a ranking hint since 2020-03-01) and an FTC §255 material-connection disclosure (2023 Endorsements Guides revision). Suppressing either compliance layer to preserve dofollow signal exposes the buyer to an "Unnatural links to your site" manual action and the publisher to an "Unnatural links from your site" manual action. White-hat methodology meets both compliance layers without trying to launder one through the other.
How does SpamBrain detect link schemes?
SpamBrain detects link schemes through the link-graph network rather than destination quality. The neural detection system reads shared C-class IP and ASN clustering across a network, whois-history overlap (the same registrant emails), content-template overlap (repeated themes, plugin sets, footer structures), and link-graph anomalies where network sites interlink at unusual density. Footprint-fingerprinting at the rendering level catches shared analytics codes, AdSense IDs, and custom CSS. Attempts to defeat detection via diverse IPs, different themes, or Web 2.0 hosted blog networks fail because the link-graph network remains the structural signal SpamBrain reads.
What is link decay and why does it matter?
Link decay is the natural attrition of acquired inbound links over time. Publishers edit pages, remove content, redesign sites, and flip dofollow to nofollow as editorial policies update. The industry-standard baseline runs 10 to 20 percent annual decay. Retainer economics scope monthly placement counts against this attrition. A campaign that delivers 24 placements in a quarter against an 18 percent annual decay rate maintains a net-positive signal trajectory; a one-off pack of 24 placements without a maintenance retainer reverses out over 12 to 18 months and the inbound profile returns to the pre-campaign baseline.
What is the role of branded queries in off-page SEO?
Branded query volume serves as a mathematical anchor for inbound link velocity. Google's Panda patent (U.S. Patent 8,682,892) details a ratio between a site's inbound link volume and the volume of search queries related to the site's brand. The May 2024 Content Warehouse documentation leak confirmed brand-strength scoring tied to this signal. Domains acquiring high inbound link volume without generating branded queries flag the link profile as unnatural and trigger algorithmic demotion. Digital PR and brand-mention work generate the branded query volume that mathematically supports the inbound link velocity, so the two surfaces work as a paired engine.
The exact-match ceiling is 15 percent per URL. The natural baseline is 70 percent branded.
The audit pulls the inbound profile, segments by anchor category and topical-cluster proximity, models the URL-level distribution against the per-vertical baseline, and names the placements Penguin 4.0 is discounting in real time. Inside two weeks.