Link building : what to bury and what still works

Link building : what to bury and what still works

Michael Johnson, CEO and founder of Resolve, detailed in a webinar the link building tactics (acquiring links from other sites) he would abandon and those he would fund in 2027. Hosted by Loren Baker on October 21, the discussion is based on a simple test: a useful link must benefit the brand, the publisher, and the publisher’s audience.

  • Michael Johnson juge qu’un lien utile doit profiter à la marque, à l’éditeur et à son audience.
  • Il rejette les métriques de domaine comme critère principal, car elles sont faciles à gonfler.
  • Il abandonne les réseaux de blogs privés, les échanges de liens et les ancres commerciales forcées, mais garde les contributions d’invités, le netlinking sur liens cassés et les commentaires d’experts bien réalisés.
  • Il privilégie les études originales et les outils utiles, qui donnent aux éditeurs une vraie raison de citer la marque.

He believes this criterion applies equally to search engine ranking and visibility in generative AI. In this context, the number of links is not the goal. A report may appear satisfactory while the approaches that produced it give publishers little reason to be interested.

Don’t let domain metrics decide

Michael Johnson opposes making Domain Authority or Domain Rating the primary criterion for placement. A high score establishes neither expertise on the subject, nor a useful audience, nor a serious reason for the publisher to cite a brand.

Asked about these scores, he stated: “They are third-party indicators. They are not Google signals, and they are easy to inflate.” To qualify a site, he recommends first checking thematic relevance, then examining the keywords for which the site actually ranks, beyond the total estimated traffic. A technology publication that attracts searches unrelated to its topic, such as celebrity searches, deserves careful scrutiny in his view, regardless of its domain score.

Ban manipulation, not all known practices

The list of practices to abandon includes private blog networks, link exchanges, forced commercial anchors, and generic requests for inclusion in article roundups. Michael Johnson also criticizes the habit of directing most outreach to product pages, whereas a guide, dataset, or tool would give a publisher a better reason to cite the site.

He does not, however, argue that authentic partners should never cite each other, or that a product page can never obtain a useful link. His objection is to artificial schemes used as the foundation of a program. Several common practices remain on his list to keep, provided they are well executed: guest contributions, which bring expertise and useful resources rather than a thin article built around a commercial anchor; outreach around broken links and resource pages; brand mentions without links, which can be turned into links when it helps the reader; and expert commentary, provided the contributor brings relevant and substantial expertise.

Give publishers a reason to include you

For article roundups, Michael Johnson first studies the companies already present and what got them there. A relevant feature, recent data, or a useful tool can justify inclusion, whereas the desire to get a link is not enough. Asked about roundups that link to an affiliate program or an Amazon product page, he believes the brand mention can still have value, provided it aims for a legitimate recommendation rather than treating every roundup as an equivalent target.

For digital PR, he favors original studies, which offer journalists a topic they won’t find elsewhere. He also presented an interactive map of “news deserts,” built from public data and accompanied by an explanatory guide, which could interest journalism training programs. Johnson also notes that a strategy based on one-off campaigns does not guarantee results: the justification for an approach lies in building a useful system, not in the promise that every outreach will succeed.

Finally, he presents a 90-day organization plan, ranging from auditing the existing program to revising placements, including identifying relevant AI-related purchase queries and creating necessary content. For details of this 90-day execution plan, the full webinar remains the reference.