Cold Email Reply Rate Benchmarks 2026: Why Interested Replies Are the Only Metric That Matters

A good cold email reply rate in 2026 is 0.5 to 1.0 percent, and anything above 1.5 percent is really good. But raw reply rate only tells you whether you are landing in the inbox. The metric that actually predicts meetings is the interested reply rate: if 10 percent or more of your replies are genuinely interested, the campaign is good, and at 20 percent or more it is great.

We run cold email campaigns full time for B2B clients, sending up to 50,000 emails a month per client. Every benchmark in this post comes from that live client base, not from a survey or a scraped roundup. And the single biggest lesson from all that volume is this: most of the industry is staring at the wrong number.

What is a good cold email reply rate?

Raw reply rate, meaning total replies divided by emails sent, is a deliverability signal and nothing more. Getting replies, even negative ones, proves you are landing in the primary inbox. Here is how we grade it across our client campaigns:

One important caveat: judge reply rate against the mailbox mix of your list. Sending to every mail provider, around 0.5 to 0.6 percent at volume is average. A Google-only list can run around 1 percent. Enterprise lists heavy on Outlook and security gateways like Mimecast run lower still. A low reply rate against the wrong baseline is not a broken campaign.

Notice what is missing from that table: any mention of meetings, pipeline, or revenue. That is because raw reply rate has almost nothing to do with them.

Why is raw reply rate the wrong metric?

Raw reply rate is the wrong metric because it counts noise as signal. A reply can be an out-of-office message, an unsubscribe request, an angry "take me off your list", a referral to someone else, or a vendor pitching you back. None of those are pipeline. All of them inflate the number that most guides tell you to optimise.

We learned this the hard way inside a single client campaign. Four copy variants were running at comparable volume, and the variant with the most replies pulled a 0.81 percent reply rate, the best of the set. It was also the worst variant we had: only 2.5 percent of those replies were interested. The winning variant pulled a lower 0.63 percent reply rate, but 22.6 percent of its replies were genuinely interested. If we had ranked on raw replies, we would have scaled the loser. A high reply rate on a low interested base is copy that provokes, not copy that sells.

This is why we split the measurement into two separate signals. Reply rate answers "are we landing in the inbox?". Interested reply rate answers "does anyone want what we are offering?". Conflating them is the most common analytical mistake in cold email.

What is a good interested reply rate?

An interested reply is a genuine positive: a yes, a request for the thing you offered, or a real question about how it works. Measured as a share of total replies, 10 percent or more interested is good, 20 percent or more is great, and below 5 percent means the offer is the problem. Here are the full tiers we run our client campaigns against:

We also track the same signal as a density metric, and the cleanest way to express it is emails per interested reply. Under 500 sends per interested reply is really good, which works out at better than 0.2 percent of sends. Five hundred to 1,000 sends per interested reply is good. Over 1,000 means the variant is underperforming, and over 2,000 sends with zero interested means kill it or significantly rework it. Lower is better, and it is the number we use for every head-to-head variant comparison and every volume decision. The offer side of that decision is its own discipline, which we cover in our copywriting and offers playbook.

A healthy reply rate with a zero interested rate is the clearest diagnostic in cold email: your infrastructure works and your offer does not.

How do auto-replies and unsubscribes pollute reply rate?

Auto-replies and unsubscribes pollute reply rate because most sending platforms count every inbound message as a reply by default. Out-of-office responders alone can be a meaningful share of inbound on a large B2B send, and none of it is engagement. We configure every campaign to exclude auto-replies from stats, and we still do not trust the numbers blindly.

The bigger problem is over-counting on the interested side. Automated classifiers routinely mark referrals ("speak to our head of X instead"), polite declines, and confused questions as interested. In one weekly review, a client dashboard showed 9 interested replies. When we read the actual reply text, the real number was about 4: one was a glitch, and two were soft objections dressed up as questions. Every offer decision rests on that one number, so we verify every interested classification by hand against the reply text before acting on it.

If you take one operational habit from this post, make it that one. Read the replies. The count in your dashboard is a claim, not a fact.

What does each benchmark tier actually tell you?

Each tier is a diagnosis, not a grade. The whole point of splitting reply rate from interested rate is that the combination tells you which lever to pull:

• Low reply rate, any interested rate. Below 0.5 percent at 1,000 or more sends, suspect deliverability or list quality first. Pause, reverify the list, check your sending settings. Our deliverability playbook covers the hygiene layer in detail.
• Healthy reply rate, zero or near-zero interested. People are receiving and reading, and nobody wants the thing. This is an offer problem. No amount of copy polish fixes it.
• Healthy reply rate, interested replies flowing. The offer works. The only mistake now is tinkering instead of scaling.
• Interested replies but no meetings. The problem has moved downstream to reply handling and booking friction, not the email itself.

There is one hard rule underneath all of these tiers: volume floors. Nothing is conclusive on thin volume. We do not make variant decisions before roughly 1,000 sends per variant, and we do not declare an offer dead before it has taken around 2,000 sends and around 20 replies with zero interested. Below those floors, a bad number is noise. Above them, it is a verdict. The full logic of which lever to pull, and in what order, is in our companion post on why rewriting your copy is usually the wrong fix.

Why does everyone else say 1-5% is a good reply rate?

The generic advice to "aim for a 1 to 5 percent reply rate" survives because it is easy to measure and flattering to report. It comes from blog roundups quoting each other, and from platforms whose dashboards put raw replies front and centre. It fails for two reasons.

First, it bundles noise into the numerator. A 5 percent reply rate that is half out-of-office messages and unsubscribes is not a 5 percent reply rate in any meaningful sense. Second, it says nothing about intent. We have watched campaigns with modest reply rates outbook campaigns with double the replies, because one offer pulled buyers and the other pulled objections. One of our best-performing client campaigns ran a 0.86 percent reply rate, well below the generic target, with 18 percent of replies interested and around 17 percent of those interested replies converting to booked meetings. By the roundup benchmarks it was mediocre. By the only metrics that matter, it was a money printer.

Reply rate tells you the email arrived. Interested rate tells you the offer works. Bookings tell you the whole system works. Optimise them in that order, and never let the first one stand in for the other two.

FAQ

What is a good cold email reply rate in 2026?

A raw reply rate of 0.5-1.0% confirms you are landing in the inbox, 1.0-1.5% is good, and above 1.5% is really good. But raw reply rate only measures deliverability. Judge campaign quality on interested replies: 10%+ of replies interested is good, 20%+ is great.

What is an interested reply rate?

Interested reply rate is the share of replies that express genuine positive intent: a yes, a request for more information, or a real technical question. It excludes auto-replies, unsubscribes, referrals, and polite declines. It is the offer signal, where raw reply rate is only the deliverability signal.

Why is my reply rate high but I get no meetings?

A high reply rate with near-zero interested replies means your emails are landing and being read, but the offer is not pulling. That is an offer problem, not a deliverability problem. Rework the offer with a stronger hook, more proof, or a lower commitment ask.

How many cold emails does it take to get one interested reply?

In our campaigns, under 500 sends per interested reply is really good, 500-1,000 is good, and over 1,000 means the variant is underperforming. Over 2,000 sends with zero interested replies means kill or significantly rework that variant.

When should I pause a cold email campaign?

Pause and rework when an offer has taken roughly 2,000 sends and around 20 replies with zero interested replies. At that point the offer is validated as not working. Continuing to send the same message just burns list and sender reputation for no return.

Do auto-replies count towards reply rate?

They should not. Out-of-office messages, unsubscribes, and vendors pitching you back are not engagement, but most sending platforms count them by default. We exclude auto-replies from stats on every campaign and manually verify each interested classification against the actual reply text.