How our cold email open rate went from 21% to 31%

This is a plain record of one evolution round - from a 21% open rate to 31% - and what actually moved the number.

Where we started

The open rate sat at 21%, roughly the industry average. The copy was fine; the targeting was broad. We were sending to everyone who might fit, instead of the few who clearly did.

What we changed

• Narrowed the match: instead of "manufacturers in category X," we filtered to those with recent procurement signals. Fewer sends, sharper fits.
• Rewrote the first line: the opener now references the recipient's own product, not a generic pitch. Personalization at scale, driven by the matching engine.
• Compressed the ask: one clear question instead of three. Lower friction, higher reply.

What the data said

Open rate climbed to 31%. But the more telling shift was repeat opens - readers came back. That told us the content now earned a second look, not just a glance.

The lesson wasn't "write better copy." It was "send to fewer, sharper fits and let the engine personalize." Manual tweaking chased the symptom; the system fixed the cause.

Why self-evolution wins

Every send feeds its result back. The next round inherits what worked and drops what didn't - without a human rewriting templates each week. That's the difference between a tool you operate and a system that improves itself.