Resolve CEO Labels Domain Authority Decisions Flawed In Link Building Audit

Domain Authority and Domain Rating are third-party metrics, not Google signals, and they are easy to inflate. That statement came from Michael Johnson, CEO and founder of Resolve, during the October 21 Search Engine Journal webinar "The Link Building Graveyard: What To Bury & What Still Wins," hosted by Loren Baker with a live Q&A.

Johnson offered a keep-or-cut framework for 2027 off-page planning. His central test was simple: a worthwhile link should benefit the brand, the publisher, and the publisher's audience. The test applied to both search rankings and AI visibility. Link count alone was never the goal.

What did Johnson say about domain metrics?

He rejected Domain Authority and Domain Rating as the primary measure of a placement. A high score, he argued, proves no topical expertise, no useful audience, and no sensible reason for a publisher to link. His direct words:

"These are third-party metrics. They're not Google signals, and they're easy to inflate."

He did not tell marketers to discard metrics. He told them to sequence the work. Check topical relevance first. Then move past estimated traffic totals to the keywords a site actually ranks for. A technology publication pulling unrelated celebrity searches deserves scrutiny regardless of its domain score. The clip sits at 20:18 in the on-demand session.

The takeaway: metrics earn a vote only after expertise and audience pass review.

Which tactics belong in the graveyard?

Johnson's cut list was specific:

• Private blog networks
• Link-swap schemes
• Forced commercial anchors
• Generic requests to be added to roundups

He also challenged the habit of aiming most outreach at product pages. A guide, dataset, or tool gives an editor a stronger reason to cite a site. The objection was not absolute. Genuine partners can link to each other. A product page can earn a backlink. The problem is making an artificial pattern the foundation of the program.

In one client example, he converted existing product-related brand mentions into links to the relevant product page. The context of the coverage created the opportunity. No product URL was forced into unrelated content. That example appears at 27:54.

What still wins placements?

Four familiar tactics stayed on his keep list when execution served readers:

• Guest contributions: subject expertise and useful resources, not thin articles built around commercial anchors
• Broken-link and resource-page outreach: help a publisher repair or improve a resource with a fitting asset
• Unlinked mentions: add links where existing coverage already discusses the brand
• Expert commentary: respond to journalist requests only with relevant, substantive expertise

On listicles, Johnson drew a line between asking for a placement and improving a publisher's list. Study the companies already featured. Ask what qualified them. A relevant feature, fresh data, or a helpful tool builds a case for inclusion. Wanting the backlink does not.

Baker asked whether a roundup still holds AI-search value when links run through affiliate programs or to Amazon product pages. Johnson said the brand mention could still matter. He urged earning a legitimate recommendation rather than treating every roundup as equally valuable. He explained why branded mentions count as an AI visibility win at 8:57.

How does digital PR fit?

Johnson favored original research that gives journalists a story unavailable elsewhere. He walked through campaigns built on workplace AI concerns and travel costs. He also demonstrated an interactive news-desert map built from public data and paired with an explanatory guide, which gave journalism programs a resource worth citing.

One caveat followed every promise. He acknowledged that earned campaigns are not guaranteed to succeed. Newsjacking returns run inconsistent. The stronger path is a proactive research program, not reaction to someone else's news cycle. The case for repeating campaigns rests on building a useful system, not on promising every pitch lands.

He also noted that AI-generated prospect lists still required human qualification. Automation can surface candidates. It cannot verify audience fit.

What goes into the 90-day plan?

Johnson sequenced the work into three phases:

1. Audit the existing program, current spending, and opportunities
2. Research, asset creation, and outreach to qualified publishers and resource pages
3. Review placements and develop the next useful asset

The first phase also includes a baseline for relevant AI buying prompts. The phases build on each other rather than running as disconnected campaigns.

His starting decisions:

• Review where the link budget goes; flag placements with weak topical fit or no credible audience
• Build a target list from AI citations, competitor coverage, and audience-specific resource pages
• Choose a dataset, tool, guide, or expert contribution that improves those publishers' coverage
• Track referring domains and linked or unlinked brand mentions alongside AI visibility and organic rankings

His guest-post advice, focused on audience problems, sits at 57:17 in the on-demand session.

The takeaway: the framework converts opinion into a sequenced audit, and the audit forces the budget to justify itself.

Search Engine Journal announced its next webinar, "Live Website Audit: SEO Fixes for AI Search Visibility in 2027," where Baker will audit a real website live. Marketers should watch whether these keep-or-cut claims hold as AI citation data accumulates through 2027.