Warm replies get ~5x more engagement than cold replies on X

The stat. In the State of X Engagement 2026 dataset, replies sent to accounts that had already engaged the author first ("warm" replies — they liked, replied to, or followed the account) averaged about 5x the engagement of "cold" replies sent to strangers in the same period.

How it was measured. Every reply the agent posted was logged, then re-scraped a few days later for likes, replies, and bookmarks. Engagement was scored with a single weighted formula — likes + 3×replies + 5×bookmarks — so that high-intent signals count more than passing likes. Replies were tagged "warm" if the target had a prior interaction with the account on record, "cold" otherwise, and the two groups' average scores were compared. The sample was 983 replies from a single account (@thedeepflux).

The honest caveat. This is observational data from one account, n≈983 replies — not a randomized experiment. Warm targets are correlated with other things (the author already found them interesting, they're often smaller and more reciprocal), so part of the 5x is selection, not pure causation. The exact multiplier will differ by niche and account size. Treat "about 5x" as a strong directional signal, not a precise constant.

The tactic it implies. Spend your reply budget on people who've already shown up for you before chasing strangers. Reply-guy energy aimed at people who liked your last post, replied to you, or just followed you compounds far faster than cold-replying big accounts hoping to be noticed. Practically: work your notifications first, your timeline second. The warm loop is the highest-ROI surface on X, and most people skip it to go chase reach.