Automated internal linking, the way we approach it
At Affiliate Traffic Factory, we build affiliate sites that do more than publish content. We structure publisher identity, reviewed entities, schema and trust signals so search engines and AI systems can understand what a site is about. Internal linking sits inside that work. It is part of how a site's structure communicates meaning, not a finishing touch applied after publishing.
What we look at first
We start with meaning rather than keywords. In "How to Automate Internal Linking with AI Embeddings", Niko Alho explains that by converting site content into vector embeddings and calculating the cosine similarity between pages, links can be injected programmatically "based on semantic relevance, not just keyword matching." We agree with the reason he gives: a page discussing "revenue architecture" and a page discussing "profit scaling" might share no keywords, yet be semantically close, and a keyword-based system misses that connection entirely. So the first thing we map is the entities and topics our pages actually cover — the same reviewed entities and schema we already structure for the publisher — so relevance is judged on context.
How we work through it
We treat the site as a graph rather than a list of posts. Alho's aim of a system that distributes authority to high-value assets, instead of leaving orphan pages and wasted crawl budget, is the aim we work to. We work cautiously. As the WordPress.org listing for Pagup's Automatic Internal Links puts it, "A cautious internal linking setup is usually better than an aggressive one. In most cases, a small number of relevant links per page is preferable." We build coverage as a repeatable workflow rather than a series of manual insertions. Moonlit Platform notes that manually weaving internal links through blog posts is slow, easy to miss, and often ends in a scatter-shot web of half-relevant anchors; its tool leans on semantic search and an LLM to place context-rich links. That is the direction we take: context-rich links, applied consistently.
Why we do it this way
Because internal linking is structural, not cosmetic. The Pagup listing calls it "a structural layer of digital readability" that tells search engines and AI systems how pages relate to each other, which pages carry authority, and how content is organised — and warns that without coherent internal links, even well-written content remains structurally isolated. For affiliate publishers, that isolation is the real failure mode: content keeps shipping, but the architecture never tells anyone what the site is about. Automated internal linking, driven by semantic relevance and kept restrained, is how we keep that architecture coherent as a site grows.
References
- How to Automate Internal Linking with AI Embeddings — Niko Alho
- Automatic Internal Links for SEO by Pagup – WordPress plugin | WordPress.org — WordPress.org
- Automate Internal Linking using AI - Moonlit Platform — Moonlit Platform