Cold email metrics: the 7 numbers that predict revenue

Why open rate is not one of the seven

The metric most dashboards lead with is the one to trust least. Privacy features such as Apple Mail Privacy Protection auto-load tracking pixels, so open rate arrives inflated and distorted before you read it. It can settle a subject-line argument, roughly, but nobody pays you for an open. The seven numbers that follow run in funnel order: five that build revenue, and two that protect the sending infrastructure everything sits on.

The five revenue metrics, in funnel order

• Reply rate. Replies divided by leads contacted. We plan capacity on two touches per lead, and treat 2 percent as the realistic mark for a well-built list.
• Positive-reply share. The fraction of replies showing genuine interest, as opposed to objections and unsubscribes. Realistic: 15 percent of replies.
• Meeting rate. The fraction of positive replies that become a booked meeting. Realistic: 25 percent, and it is mostly a function of how fast replies get answered.
• Show-up rate. The fraction of booked meetings the prospect attends. Realistic: 75 percent. High bookings with low show-up means the wrong people are being booked.
• Close rate. The fraction of held meetings that become deals. Realistic: 20 percent, and this one belongs to your sales process and offer fit more than to the campaign.

Track them in that order, because each converts into the next. When the month dips, walking the chain shows you which link broke; a single total shows you nothing.

Benchmarks: conservative, realistic, aggressive

These are the three scenarios our ROI calculator models, with each lead touched twice:

Two things are worth noticing. The gap between the columns compounds at every stage, so small per-stage differences produce very different pipelines from identical volume. And no single row needs to be heroic: hitting the realistic column on all five rows produces more pipeline than hitting aggressive on reply rate and conservative everywhere else.

For calibration against measured reality rather than modeled scenarios: across the 81 qualifying campaigns in our book, the median campaign replies at 2.12 percent of contacted leads, the middle half of campaigns lands between 1.38 and 2.97 percent, and the measured bounce rate is 1.32 percent. The realistic column above is a scenario, not a promise; the measured distribution it has to survive is on the reply rate benchmark page.

The two survival metrics: bounce rate and complaint rate

Bounce rate: 2 percent is the working ceiling. Sustained rates above 5 percent invite throttling and blocks regardless of how good the copy is. The fix is procedural, not creative: verify the list with a live email-verification pass immediately before sending, and never trust a vendor's "verified" flag. In one of our own bounce audits, 35 of 35 dead mailboxes traced back to leads whose vendor validation said valid.

Complaint rate: 0.3 percent is the line. Google and Yahoo hold bulk senders to it, and once you send 5,000 messages a day to either provider, SPF, DKIM, DMARC and one-click unsubscribe stop being best practice and become a hard gate. Placement lost to complaints is slow to recover, which is why these two numbers get checked before any revenue metric. The deliverability guide covers the full setup.

A real 12-month funnel: 1,080,000 emails to 263 qualified sales meetings

Over 12 months, one portal-tracked client campaign sent 1,080,000 emails and 24,000 LinkedIn invites, drew 5,028 replies, and produced 263 qualified sales meetings. Divided through, that is roughly 263 qualified sales meetings over 12 months, and about one across that campaign. On the two-touch cadence those sends reached about 540,000 unique leads, so 5,028 replies sits a hair under the 1 percent conservative benchmark, and the campaign still produced 263 qualified sales meetings across the year. More results, including a client running at 90 or more positive replies per month since 2024, are on the case studies page.

The dedicated 2026 benchmarks page holds the full scenario table and the measured funnel side by side.

What moves each number

• Reply rate moves with list quality and copy. Start with the lead list, then the copywriting, and hold per-mailbox volume conservative - 10 to 12 on Google Workspace, around 1 to 2 on Microsoft 365 - so deliverability never caps it (see how many cold emails per day).
• Positive-reply share moves with offer relevance and how tightly the list matches your ideal customer, which makes it a targeting number more than a copy number.
• Meeting rate moves with reply-handling speed and how easy the yes is: propose times rather than asking for them.
• Show-up rate moves with qualification and reminders.
• Close rate moves with fit and sales process, which is exactly why our pricing is revenue share: we only do well when this number does.
• Bounce and complaint rates move with verification, authentication and volume discipline, all covered in the cold email guide.

Common questions

What is a good cold email reply rate?

On a two-touch cadence: 1 percent is conservative, 2 percent is realistic for a well-built list, and 3 percent is aggressive. Reply rate on its own is incomplete, though. Read it together with positive-reply share, because a high reply rate made of unsubscribes and objections builds no pipeline.

What bounce rate is acceptable for cold email?

Under 2 percent is the working ceiling we run to; sustained rates above 5 percent invite throttling and blocks. The reliable fix is a live verification pass on the list immediately before sending, every time, regardless of what the data vendor claims.

What is the spam complaint threshold?

0.3 percent. Google and Yahoo hold bulk senders to it, and at 5,000 messages a day to either provider the authentication requirements, SPF, DKIM, DMARC and one-click unsubscribe, become mandatory. Complaint damage is the slowest to repair, so this is the number to watch daily.

How many cold emails does it take to get a sales-qualified lead?

In the real 12-month funnel above, 1,080,000 emails produced 5,028 replies and 263 sales-qualified leads: roughly the campaign figures above. Your ratio depends on list quality and offer, and you can model your own version in the ROI calculator.

Why is open rate not on the list?

Privacy features auto-load tracking pixels, so opens arrive inflated and distorted. Open rate can give a rough directional read on a subject line, but it does not predict pipeline, and optimizing to it reliably makes campaigns worse at the seven numbers that do.