The trouble with reply rate
Reply rate sounds like progress because it's adjacent to progress. Someone replied! Something is working! But the math falls apart fast: a 30% reply rate to a list of 100 wrong-fit prospects is six times worse than a 5% reply rate to a list of 100 right-fit ones, because every reply on the first list is either a polite no, an angry no, or a meeting that wastes 30 minutes of a closer's day.
The metric you're actually trying to move is meetings sourced from people who became customers, divided by messages sent to people who could have. Reply rate is what's measurable when you can't measure that yet.
If a number can't survive a CFO review, it doesn't belong on your dashboard.
Three funnels, three traps
Here's what we saw inside three customer accounts last quarter — same metric, very different stories.
Customer A — high reply, low conversion
27% reply rate. Looked great. When we cut by ICP fit score, replies from the bottom quartile were 78% of total volume — and zero of those produced a meeting that converted. The team was burning time on people who'd reply 'unsubscribe please' politely and consistently.
Customer B — low reply, high conversion
4.8% reply rate. Looked broken. But every reply came from accounts in the top ICP quartile, and 34% of replies became sourced opportunities. The 'low reply rate' was actually the right reply rate — they were filtering hard.
Customer C — middle reply, middle conversion
11% reply rate. Hidden inside that average: a single sequence variant was responsible for 80% of the converting replies. The others were padding the topline and burning the sender domain.
A better hierarchy of metrics
The order we ask our customers to track:
- Sourced opportunities per 1,000 messages — by sequence variant, by ICP quartile.
- Meetings booked → meetings held — show-rate is a deliverability and routing problem.
- Reply rate, segmented by ICP score — only useful if you cut it.
- Sender reputation deltas — open rates are noise but sender deliverability isn't.
- Time-to-first-reply — leading indicator of inbox placement.
How to track it in your CRM
You don't need a new tool for this. Three custom fields on your contact object — sequence_variant, icp_quartile, first_reply_at — plus a weekly report grouping by the first two and filtering by the third, gets you most of the way there. We have a HubSpot and Salesforce template we ship to new customers; ask your CSM for it.
Takeaways
- Reply rate is a vanity metric unless you segment it.
- Sourced opportunities per 1,000 messages is the number to defend.
- If your reply rate goes up but conversion doesn't, you have a fit problem, not a copy problem.
- Track sender reputation as a separate axis — it's a leading indicator of everything else.
If you want to see how this looks inside the LeadGen AI Suite, our platform overview shows the report we ship by default. Or come talk to us — we'll show you yours.