Search engines discover and rank pages by following links. When Googlebot crawls a page, it parses the HTML for anchor tags, can add the href targets to its known URL set, and records the link relationship. This web-graph discovery is how new or updated pages enter the index without a manual submission. Once discovered, the algorithm evaluates each link for context. Anchor text—the clickable words inside the <a> tag—signals what the destination page is about. A link with anchor text 'commercial roofing Ottawa' suggests the target page covers that topic. Surrounding sentences and the linking page's own topic add further context. If a construction industry blog links to a roofing contractor's service page, the algorithm infers topical alignment. Domain-level trust also matters. Links from sites that already rank well, have clean spam histories, and align with the target's niche carry more weight than links from unrelated or low-quality domains. The algorithm doesn't count links as equal votes; it models authority transfer along edges of the web graph, adjusting for factors like editorial placement (in-content versus footer), rel attributes (nofollow, sponsored, ugc), and link velocity patterns that might indicate manipulation. Crawlers also respect robots.txt and meta directives. A page blocked by robots.txt won't be crawled, so its outbound links won't be discovered. A nofollow attribute tells the algorithm not to pass ranking credit, though Google has treated nofollow as a hint rather than a directive since 2019. Evaluate whether a source page is accessible, relevant, editorially credible, and useful to the audience instead of relying on a single third-party score. Avoid manufactured sitewide placements and prefer links that make editorial sense in their surrounding content. Evaluate a link in context: why it exists, who may use it, how closely the source and destination relate, and whether it sends qualified referral visits. Track patterns across discovery, traffic, and search visibility rather than assigning one placement a guaranteed effect. This keeps link work focused on editorial value and useful distribution instead of third-party scores, fixed counts, or promised ranking timelines.