B2B Outbound Benchmarks 2026: Reply Rates, Meeting Rates and Cost per Meeting by Channel

In 2026, a well-run B2B outbound programme should expect a 3–5% cold email reply rate, a 28% LinkedIn connection acceptance rate, a roughly 10% cold-call connect rate per dial, and a 3–4% response rate on physical mail — with the gap between average and top-quartile performance driven by data quality and targeting rather than volume.

Benchmarks reviewed September 2026. Every figure below links to the report it came from, with its publisher and period. Where sources disagree, the range is shown rather than an average of the two.

Key facts

• The average cold email reply rate is 3.43%, the top quartile 5.5%, measured across billions of interactions between 1 January and 18 December 2025 (Instantly Cold Email Benchmark Report 2026).
• LinkedIn connection requests are accepted 28.5% of the time; post-connection messages reply at 10.4%, across 13,218,869 requests from 13,302 accounts, May 2025 to April 2026 (Expandi, 2026).
• Cold calls connect on 9.9% of dials, and 4.6% of the resulting conversations produce a booked meeting, across 175,000+ dials (Belkins, 2026).
• Direct mail responds at 3.63% overall — 5.30% to a house list and 2.90% to a targeted prospect list (Focus Digital, 2026).
• A LinkedIn Ads lead costs $15 to $350 depending on region, averaging $120 in EMEA and $230 in North America (Tamarind's B2B House, 2026).

What is a good cold email reply rate in 2026?

Three independent 2026 sources put the average between 3.1% and 3.43%, and all three put the top decile above 8%.

Two operational findings matter more than the headline averages.

Data quality roughly doubles the outcome. Verified lists achieve about 2x the reply rate of unverified lists and 5–6x the rate of purchased lists, with bounce rates under 1.2% against 7.8% (Cleanlist, 2026).

Follow-ups are worth 42% of your replies. The first email generates 58% of all replies; everything after it produces the remaining 42% (Instantly, 2026). A sequence that stops at one touch throws away two fifths of its own results.

Best use case: repeatable offers to a defined ICP where you can verify addresses and afford a multi-touch sequence.

What is a good LinkedIn acceptance and reply rate in 2026?

The largest public dataset for 2026 covers 13.2 million connection requests.

All figures: Expandi, 2026, May 2025 – April 2026.

One trend is worth planning around: the connection-note reply rate fell from 3.5% in May 2025 to 2.2% in April 2026, a 37% relative decline, while acceptance rates held steady (Expandi, 2026). The connection request still works; the note attached to it is losing power.

Best use case: senior buyers who read LinkedIn but filter email, and any motion where the reply matters more than the volume.

What connect and meeting rates should a cold calling team expect?

Two 2026 sources, measuring different things, which is why the ranges are wide.

The per-dial and per-prospect numbers are the ones teams most often confuse. A 9.9% connect rate per dial and a 24.5% connect rate per prospect are the same dataset: persistence across attempts is what closes the gap, and 93% of conversations happen by the third call (Prospeo, 2026).

Best use case: high-ACV deals, verified direct dials, and markets where a conversation is worth more than a click.

Does physical mail still work in B2B?

It responds an order of magnitude better than email — and costs an order of magnitude more per touch, which is why it belongs at the top of a named account list rather than at the top of a funnel.

Note the spread between a targeted list at 2.90% and an untargeted one at 0.90%: the same channel performs three times better on a list somebody thought about. That ratio is the whole argument of this article, and it repeats on every channel below.

Best use case: a short list of named accounts where the buyer is unreachable digitally.

What does a paid social lead cost in B2B?

A paid lead is not a meeting. Applying the same discipline as the channels above, a $120 EMEA lead that converts to a meeting one time in three is a $360 meeting — which is the number to compare, not the CPL.

Best use case: creating recognition ahead of outbound against the same named accounts, so the first email is not the first contact.

What does a meeting actually cost on each channel?

Published sources rarely state cost per meeting directly. Two that come close, plus the arithmetic they imply:

The calculated rows are arithmetic on published inputs, not observed results, and they assume a median-performing rep. Treat them as an order of magnitude: a meeting in B2B outbound costs low hundreds of dollars when the system works, and the entire cost base when it does not.

What the benchmarks miss

Every table above averages single channels across thousands of accounts that have nothing in common. Three things that changes.

Averages hide the targeting effect. The same direct mail piece responds at 2.90% to a targeted list and 0.90% to an untargeted one (Focus Digital, 2026); verified email lists reply at roughly double the rate of unverified ones (Cleanlist, 2026). A benchmark tells you what the median list achieves. It cannot tell you what yours would, because the list is the variable.

Signal-based targeting is not in the denominator. Benchmark datasets pool campaigns sent to static lists. When the trigger is a signal — a hire, a funding event, a tender, a technology change — the same channel performs differently, and no public dataset separates the two. In our own work, the ICP for Kastner Frankfurt came out of more than one million data points analysed into a single dynamic DACH list with buying signals attached, which is not a population any benchmark measures.

Single-channel numbers cannot describe a multi-channel motion. For TektonWorx we ran email, LinkedIn, Facebook and the website against the same audience in parallel, then scaled only the channel that worked — into 400+ qualified sign-ups across four markets. The right comparison was never "is email better than LinkedIn"; it was "which one reaches this buyer".

The strongest illustration we have of the gap between an average and a targeted system is the investor outreach we ran for Oceyon AG: 500+ positive investor responses and 300+ booked meetings, adding up to €25M+ in investor pipeline. Those are not channel benchmarks. They are what one channel does when the list, the signal and the message are decided before anything is sent. The full workflow is public in our workflow library, and the engagement is written up in our case studies.

How should you use these benchmarks?

Three rules that keep a benchmark honest.

Compare like for like. A per-dial connect rate and a per-prospect connect rate differ by a factor of two and a half in the same dataset. Read the denominator before you read the number.

Use ranges as guardrails, not targets. If your cold email reply rate is 0.5%, the benchmark tells you something is broken — probably the list. If it is 6%, the benchmark tells you not to blow it up in search of 12%.

Measure the step you can change. Reply rate is downstream of the list. Meeting rate is downstream of the reply. Cost per meeting is downstream of everything. Fix the earliest broken step, then re-measure, and see the Sales System for the order we work in.

Frequently asked questions

What is a good cold email reply rate in 2026?

Between 3% and 5% is a well-run campaign; 5.5% puts you in the top quartile and 8–12% in the top decile. The measured average is 3.43% across 2025 data (Instantly, 2026) and 3.1% in a separate 2026 dataset (Cleanlist, 2026).

How many cold calls does it take to book one meeting in 2026?

Between 100 and 370 dials, depending on data quality and persistence. Top-performing teams book inside 100 dials, the average sits near 180 (Prospeo, 2026), and an end-to-end model across 175,000+ dials implies roughly 370 (Belkins, 2026).

What is a good LinkedIn connection acceptance rate?

28.5% is the platform average across 13.2 million requests, ranging from 19.9% in apparel to 36.5% in staffing and recruiting (Expandi, 2026). Post-connection messages reply at 10.4% on average.

Is cold email or LinkedIn better for B2B outbound in 2026?

They answer different questions. Cold email reaches more people per hour at a 3.43% average reply rate; LinkedIn reaches fewer at a 10.4% post-connection reply rate. The decision should be a test against your own audience, not a benchmark comparison — running channels in parallel and scaling the winner is the method we use.

How much does a B2B meeting cost?

Published data puts an in-house US SDR meeting at $766 (Bridge Group via Alba Talent, 2024). Applying the 14.6-meetings-a-month median to published outsourced capacity pricing of $2,500–$6,500 per month (CIENCE, 2026) implies roughly $171–$445 per meeting. Both figures assume a median-performing rep and exclude management time.

Cite this article

"B2B Outbound Benchmarks 2026: Reply Rates, Meeting Rates and Cost per Meeting by Channel", Saverflow AI, 2026. https://www.saverflow.ai/articles/b2b-outbound-benchmarks-2026/