Anchor-text ratios tested in real campaigns

Anchor-text ratios tested in real campaigns

Three documented experiments on how anchors were distributed across niche edits.

— A practitioner blog ran 10 edits with 70% branded/URL anchors, 30% partial-match — target gained 5 positions, no fluctuation flagged.
— Counterpoint: an agency case where 8 exact-match anchors on the same keyword preceded a 3-week ranking drop, recovered only after they diluted with a follow-up branded edit.
— Also circulating: a Reddit comment claiming 'topic anchors' (the page's H1 phrase) outperformed exact-match in their 6-edit test.

Editor's note: small samples all, but the over-optimization case is a useful cautionary data point with an actual timeline attached.

Pick of the week: the exact-match drop-and-recover writeup — it's the rare case that documents both the mistake and the fix.