<b>The 2023 antitrust testimony reframed clicks as a long-running signal</b>

<b>The 2023 antitrust testimony reframed clicks as a long-running signal</b>

The question: do user clicks influence rankings, and how does that relate to trust?

For years Google publicly minimized click signals. The 2023 U.S. v. Google antitrust trial complicated that. Testimony and exhibits referenced a system, Navboost, that uses aggregated, historical click data — described internally as one of Google's strongest signals — and a long memory of which results users engaged with for a query.

The relevance to E-E-A-T is indirect but real. If aggregated user behavior over time feeds ranking, then a site that earns repeat, satisfied engagement is accumulating something that behaves like trust, expressed through behavior rather than through credentials or links. Conversely, a site whose results are clicked and quickly abandoned is sending the inverse.

This must be stated carefully to avoid the usual overreach. The testimony does not establish a per-page 'click-through-rate dial' that you can game; the systems described use <i>aggregated</i> long-run patterns and are defended against manipulation. Buying clicks or pogo-sticking your own results is not the lesson.

Caveat: trial testimony is adversarial and partial; the exhibits describe systems as they existed at disclosed points in time, not a current, complete architecture. Treat specifics as indicative, not authoritative.

What we still don't know: how click signals interact with the quality and reliability systems — whether they reinforce or can override classifier judgments of low quality.