Most cold email benchmarks come from tool vendors averaging every account they host. That number is fine for a press release and useless for your campaign, because it mixes good lists with bad ones and warm audiences with cold ones.
This post is how we read reply rates on the campaigns we run: what to count, what a normal number looks like, and the changes that actually move it. One of our B2B cold email campaigns reached a 4.7% reply rate, against an industry average of about 1%. The gap came from the list and the angle far more than from clever copy.
How to read the number
Reply rate should count real human replies only. Strip out auto-replies and out-of-office messages, or a big send will look better than it is.
Then track a second number next to it: positive replies, the ones that show interest. The two drift apart more than people expect. A campaign can pull plenty of replies and still lose on interest, because the replies are "not now," "wrong person" and "take me off your list." Raw replies tell you people are reading. Positive replies tell you the offer and the audience fit. Judge a campaign on the second number.
Give it volume before you judge it. A few hundred sends to one audience is the minimum before a rate means anything.
What a normal rate looks like
Our rules of thumb for B2B cold email:
- Under 1 percent after a few hundred sends: something upstream is wrong. Check the list and inbox placement before you touch the copy.
- 1 to 3 percent: a healthy campaign.
- 3 to 5 percent: exceptional. Take it apart, find out why it worked, and reuse it.
These are practices for reading your own numbers, not promises. With the industry average around 1 percent, a campaign that clears 3 percent is worth studying line by line. Cold email to consumers is a different game, and we cover it below.
The angles that earn replies
A few angles keep rising to the top of our campaigns. None of them depend on clever writing.
A question with no pitch. The best performing email we have run asked whether the prospect would be at a conference both companies follow. The body said the sender was going too and would like to meet. One line at the bottom carried the only credential. It read as a logistics question from a peer, so there was nothing to say no to.
Naming their current vendor. "Still working with [their current vendor]?" reads like an internal question about a relationship they already have, so it gets opened. The follow-up that worked best admitted the competitor had won a past deal because of an existing relationship. That candor did more than any claim of being better.
Paying for the call. A $100 to $200 gift card for a 20-minute call is the most reliable mechanic we use. It turns "you are selling me" into "you are paying me for my time." It has a limit: it converts worst when the call is a long demo. Keep the ask short and the time honest.
One fact they can check. Put one signal in the first email that the prospect can verify: an account you both follow, an event you are both attending, a tool they use. Personalization they can check beats personalization that only sounds impressive.
Why cold consumer email rarely pays
Cold email to homeowners and other consumers is where good copy goes to die. We have watched excellent consumer campaigns (a free seasonal guide, before-and-after photos, a checklist, a no-pitch walkthrough offer) barely register on cold lists. The same shop, the same sender and similar copy did many times better on lists with intent: past inquiries and people with a known project.
The lesson for a service company is simple. If a list has no intent signal, fix the list before you write a word. For most home service, remodeling and real estate businesses, the best list you will ever have is the one you already own: past customers, old quotes and inquiries that went quiet. That is where we start, with a reactivation campaign, before any cold sending at all.
What to fix, in order
When a campaign goes quiet, diagnose in this order:
- List fit. Are these the right people, with a reason to care now? Most silent campaigns die here.
- Inbox placement. Seed tests on Gmail and Outlook tell you whether the email is even arriving. If it lands in spam, nothing else matters. Google's email sender guidelines spell out what Gmail checks, and our post on cold email infrastructure covers the setup behind it.
- Subject line. Does it read like a message from a person, or like a promotion?
- Body. Last, and rarely the reason a campaign goes silent.
Two more habits that hold up:
- Use two or three touches when the hook is an incentive or an event, and the full five when the pain is slow-burning.
- Keep the breakup email. A short "last note from me" message tends to draw more replies than any other step in the sequence.
What we do on client builds
Every program starts with the list you already own. In the Foundation Build we clean and segment your customer list, switch on follow-up for new leads, and run one reactivation campaign to past customers and open quotes before any cold sending. When outbound starts, we build the list, set up and own the sending infrastructure, and you see the replies every Monday on your scorecard. How the pieces fit is on how it works, and the programs are on pricing.
FAQ
What is a good cold email reply rate in 2026?
For B2B cold email, we treat 1 to 3 percent as healthy and 3 to 5 percent as exceptional, against an industry average of about 1 percent. One of our B2B campaigns reached 4.7%. Judge a campaign on positive replies, not raw replies, and give it a few hundred sends to one audience before you trust the number.
Why is my cold email reply rate under 1 percent?
In our experience the cause is almost never the body copy. Check list fit first, then inbox placement, then the subject line. A list with no reason to care, or mail landing in spam, will hold any copy under one percent. Fix those before you rewrite a single word of the email.
Do gift cards in cold emails actually work?
A $100 to $200 gift card for a 20-minute call is the most reliable single mechanic we use in B2B campaigns. It reframes the ask as paying for the prospect's time. It does not fix a weak list, and it works worst when the call itself is a long demo.