Reply rate
Reply rate is the share of delivered emails that get a human response. It is the most reliable measure of cold outreach, because unlike opens it cannot be faked or miscounted.
Reply rate is replies divided by emails delivered. It is the number worth managing in cold outreach, for two reasons: it cannot be inflated the way open rate can, and it is the input mailbox providers weight most heavily when deciding where to put your next message.
Open rate stopped being trustworthy when privacy proxies began pre-fetching tracking pixels, which registers an open nobody performed. Depending on the audience a meaningful share of reported opens are machine artefacts, so a campaign can show healthy opens and no human interest. A reply requires a person to have read the message and decided to respond, which is why it is both harder to move and worth more.
What counts as good depends on how narrowly you are targeting. Broad sending to a bought list runs well under 1%. Carefully aimed outreach to a tight profile, referencing something specific about each recipient, commonly runs 5% to 15%, and a very narrow campaign into a strong signal can go higher. If a campaign is under about 2%, the problem is nearly always relevance or timing rather than wording, and sending more of it makes things worse: low reply rates and rising complaints are the same underlying fact seen from two sides.
Count positive and negative replies separately. A clear no is a useful outcome and a healthy sign the message reached the right person; counting it alongside interest hides whether anything is working.
In practice
- 12% on a hundred emails to companies that had just posted a matching job.
- 0.4% on four thousand emails to a bought list, with complaints rising through the send.
- A 6% reply rate where half the replies are declines, which is still a working campaign.
Questions
What is a good cold email reply rate?
For carefully targeted business-to-business outreach, 5% to 15% is a normal band, and above that usually means either a very narrow list or a strong timing signal. Under 2% indicates the list or the timing is wrong rather than the copy. Broad sending to unverified lists typically comes in under 1%, which is why volume strategies need enormous numbers to work at all and damage the sending domain while doing it.
Why not measure open rate?
Because it is no longer a measurement. Privacy features that pre-fetch images register opens that no person performed, so the number includes machine activity you cannot separate out. It can still be read as a rough directional signal on subject lines within a single audience, but it should not be used to judge whether a campaign is working, and it should never be used to decide whether to scale one.
Should I count a negative reply?
Count it, and keep it separate. A no means the message reached a real person who read it, which is information about deliverability and targeting both. Tracking positive and negative separately tells you whether a low reply rate is a reach problem or an interest problem, and those have completely different fixes.
From the blog
Related terms
Last updated