MCP

Outbound math and economics

Last verified: 2026-08-05

TL;DR

  • Do this calculation before writing copy. It decides whether the campaign is possible at all, and it is the cheapest hour you will spend.
  • Published reply rates range from 0.45% to 3.43% depending on who measured and what they divided by — a sevenfold spread that is mostly definitional, not real.
  • Plan on the pessimistic end. If the campaign only works at the optimistic end, it does not work.

The chain

Every cold campaign is the same multiplication:

contacts → deliverable → replies → positive replies → meetings → opportunities

Each arrow is a fraction well under one, so the result is dominated by whichever fraction you got wrong. Working numbers, all from 2025-26 industry data and all with owners and dates attached:

  • Deliverable. Assume 5% of a fresh list bounces — one 20M-email platform corpus reports 5.1% average, and a 65M-email corpus reports 7.5% (2023) [1, 2]. Verification pulls this under 2% (Email finding and verification).
  • Replies. 3.43% platform-wide across 20M+ emails (2026) [1]. 5-10% is "good", above 10% "excellent" [1].
  • Positive share. Not a published figure — a reply includes objections, referrals and unsubscribes. A third being genuinely positive is an optimistic working assumption.
  • Meetings. Some fraction of positive replies convert, and speed dominates it (Positive reply to meeting).

So: 1,000 contacts → ~950 deliverable → ~33 replies → ~11 positive → maybe 6-8 meetings. That ratio, roughly one meeting per 130-160 contacts, is the number to sanity-check a plan against. Halve it if the list is cold, generic or bought.

The denominator trap

Two reputable 2026 studies report average reply rates of 3.43% and 0.45% [1, 3]. Both are honest. They differ because the second divides replies by total emails sent — every touch in every sequence — while the first reports per-campaign reply rate. A four-touch sequence sends four emails per contact, so the same reality produces a number roughly four times smaller.

This matters more than any tactic on this page. Before adopting a benchmark, find out what it divided by; before reporting yours, say what you divided by. See Benchmarks.

What the arithmetic tells you to change

The chain has one term you can move by an order of magnitude and several you can move by a few percent. Who is on the list is the order-of-magnitude term. The same 2026 corpus that reports a 0.45% average finds companies with 0-10 employees replying at 0.72% and enterprises with 10,000+ at 0.22% — a threefold difference driven entirely by targeting [3]. C-level recipients are 30.2% less likely to reply than non-executives across a million-plus sales cycles (2026) [4].

Copy matters, but it is a multiplier on a base the list decides. This is the empirical core of the argument in Offer and messaging and the reason Audience and lists comes before writing.

Cost per meeting

Once the chain gives you meetings per thousand contacts, cost per meeting follows: data and verification, sending infrastructure, and the hours spent researching and writing. Personalization depth is the term you control most directly, and the trade between time-per-prospect and reply lift is Economics of personalization.

The number worth computing is not cost per meeting in isolation but cost per meeting against your deal size. Outbound is a headcount-and-tooling-heavy channel, and it only makes sense where the deal justifies it — When cold email, and when not.

References

  1. Woodpecker — Cold email statistics (20M+ email corpus, updated Jun 2026)
  2. QuickMail — Cold email statistics (65M journeys and emails, Dec 2023)
  3. Belkins — Cold email response rates (7.5M emails sent in 2025, updated Jun 2026)
  4. Gong Labs — Do execs really reply to cold email? (1M+ sales cycles, Jan 2026)