I’ve tested link pruning a few times, and the results really depend on why you’re doing it.
When the backlink profile was clearly polluted (spammy PBNs, hacked sites, auto-generated junk), pruning did show measurable gains, usually better keyword stability and slow but steady recovery in impressions within 4–8 weeks. It didn’t cause ranking jumps overnight, but it removed the “drag” holding the site back.
On cleaner profiles, though, the impact was negligible. In some cases, aggressive pruning even slowed growth because borderline but still useful links were removed.
My takeaway: link pruning works as a risk-reduction and recovery tool, not a growth strategy. It’s effective when there’s a real toxicity issue; otherwise, you’re better off focusing on earning stronger links rather than cutting existing ones.