Matching the winners' ratios: did convergence help?
A 2024 case tested a tempting strategy: measure the anchor distribution of the top 10 for a target term, compute the median, and engineer your profile to match it rather than to any fixed rule.
The top-10 median in this SERP was ~14% exact-match, 41% branded, 22% naked URL, 23% generic. The site started at 39% exact-match and rebuilt toward the median over six months by adding branded and generic links.
Reported outcome: from position 16 to position 6 for the head term over eight months.
The logic is appealing — calibrate to the realized distribution of what already ranks, not to a heuristic. But it conflates correlation with prescription. The top 10 may rank despite their anchor ratios, with the ratio being a byproduct of brand strength you can't replicate by adding links.
On one hand, SERP-relative calibration is more empirical than fixed percentages. On the other, you may be cargo-culting the visible output of factors you can't see.
Limitation: single SERP, single site, eight-month window overlapping other changes.
Open question: does matching the winners' anchor distribution cause convergence in rank, or just correlate with it?
Anchor Theory
@AnchorTheory
Matching the winners' ratios: did convergence help?
Этот пост опубликован в Telegram-канале Anchor Theory. Подписаться можно по ссылке: @AnchorTheory.