Cold calling, a phone call to a prospect who hasn’t asked to hear from you, still works in 2026: the data published between 2024 and 2026 shows a low yield per dial, but a channel that keeps producing meetings. What no longer works is high-volume calling on an unverified list, at the wrong time, with a generic pitch.
The question comes back every year, often raised by vendors of email or automated outreach tools. Rather than answer it with an opinion, this article pulls together the published numbers, with their source, date, sample and exactly what they measure. The studies don’t all say the same thing, and that is the interesting part.
Why do people say cold calling is dead?
The claim rests on three observations, all of them checkable.
Prospects pick up less. Smartphones screen unknown numbers, carriers flag numbers that dial a lot, and a caller ID labeled as spam sees its answer rate collapse. We cover that in our article on caller IDs marked as spam.
Buyers research on their own. Gartner published a survey on June 25, 2025 of 632 B2B buyers interviewed in August and September 2024: 61% of them prefer a rep-free buying experience. The same survey found that 73% actively avoid suppliers who send irrelevant outreach.
Regulation is tightening for consumers. Several countries restrict calls to private individuals more and more; France, for instance, moves consumer telemarketing to prior consent from August 11, 2026. B2B calling is treated differently almost everywhere, as explained in is cold calling legal.
All three observations are accurate. They kill a certain way of calling, not the phone itself. The numbers below show why.
What does the most recent data say?
Here are the main published studies, from most recent to oldest. Values were checked on the publishers’ pages on November 5, 2026. For the full set, sorted by question and with the popular figures we could not trace, see our cold calling statistics for 2026.
| Metric | Value | Source, date, sample |
|---|---|---|
| Average success rate | 2.7% (2025 data), 2.3% (2024 data) | Cognism and WHAM, 2026 report, 200K+ calls |
| Average cold call length | 82 seconds (2025 data), 93 seconds the year before | Cognism and WHAM, 2026 report |
| Attempts to reach a prospect | 1.55 on average (2025 data), 2.9 the year before | Cognism and WHAM, 2026 report |
| Connect rate (answered calls / dials) | 5.4% (average), 13.3% (top quartile) | Gong Labs, July 18, 2024, 300M+ calls |
| Dials per conversation | 19 (average), 8 (top quartile) | Gong Labs, July 18, 2024 |
| Conversations that become a meeting | 4.6% (average), 16.7% (top quartile) | Gong Labs, July 18, 2024 |
| Email reply rate | 3.44% with calls, 1.81% without | Gong Labs, July 18, 2024 |
| Buyers who prefer a rep-free experience | 61% | Gartner, published June 25, 2025, 632 buyers surveyed Aug.-Sept. 2024 |
| Buyers who avoid irrelevant outreach | 73% | Gartner, same survey |
| Share of pipeline generated by phone | more than half | Orum, State of Sales Development, Sept. 14, 2023, methodology not published |
| Buyers who accepted a call from an unknown provider | 69% | RAIN Group, Top Performance in Sales Prospecting, 2018, 488 buyers |
| C-level and VP buyers who prefer the phone | 57% (vs. 51% of directors, 47% of managers) | RAIN Group, 2018 |
How many cold calls actually connect?
In its analysis published on July 18, 2024 across more than 300 million calls, Gong Labs defines connect rate as calls answered by prospects divided by dials made by the rep. The result: 5.4% for the average rep, 13.3% for a top-quartile rep. In practice that means 19 dials per conversation on average, and 8 for the best.
That looks low, and it is, because it includes everything: wrong numbers, switchboards, unverified numbers, dead time slots. Our benchmarks for B2B lists, detailed in cold call connect rate benchmarks, run higher on verified direct dials. The two don’t contradict each other: one measures the average of a huge pool of calls, the other what you get once the list is clean.
The most telling part of Gong’s data is the spread: the top quartile connects about 2.5 times as often as the average. Those reps aren’t calling magically different prospects. They call better-verified numbers, in better time windows, from caller IDs that aren’t flagged.
Cognism and WHAM, for their part, found on 2025 data that it takes 1.55 attempts on average to reach a prospect, against 2.9 the year before. The same report names 10 to 11 AM as the most effective window, followed by 2 to 3 PM, with Thursday as the best day.
What share of conversations turn into meetings?
Once the prospect is on the line, the phone still performs. Gong Labs measures that 4.6% of conversations turn into a meeting for the average rep and 16.7% for the top quartile.
Cognism and WHAM report a “success rate” of 2.7% on 2025 data, up from 2.3% on 2024 data. The report before that gave 4.82% and defined it as the share of conversations that ended with a meeting booked. The 2026 report page doesn’t restate the definition: we assume it is unchanged but can’t confirm it.
Combine Gong’s averages and you need about 413 dials per meeting (19 dials per conversation divided by 4.6%). For the top quartile, about 48. Gong turns that gap into monthly meetings: assuming 800 dials a month, 2 meetings for the average rep and 18 for a top-quartile rep.
These very broad averages explain why our own math in how many cold calls to book a meeting, built on cleaner data, comes out far more favorable. Average cold calling works poorly. Well-prepared cold calling works.
Do buyers still want to be called?
Buyer-side data is older, but it is still the most detailed. RAIN Group’s Top Performance in Sales Prospecting study (2018) surveyed 488 buyers representing $4.2 billion in purchasing, and 489 sellers.
What it found:
- 82% of buyers accept meetings at least sometimes with sellers who reach out to them.
- 69% have accepted a phone call from a provider they had never worked with.
- 57% of C-level and VP buyers prefer to be contacted by phone, against 51% of directors and 47% of managers.
- 71% of buyers who accept meetings want to hear from sellers early, while they are still looking for ideas.
- 56% give a new provider 5 to 10 minutes on the phone to show they can be valuable.
Gartner’s 2025 survey seems to say the opposite: 61% of buyers prefer to buy without a rep. The two findings fit together. A buyer can prefer to research alone and still take a call that brings something they didn’t know. The most useful Gartner number for a rep is actually the other one: 73% avoid suppliers whose outreach is irrelevant. The phone isn’t being rejected; sloppiness is.
Cold call or cold email: do you have to choose?
No. Gong Labs found that reps who call get a 3.44% email reply rate, against 1.81% for reps who don’t, even when the call doesn’t connect. A name seen on a missed call or heard on a voicemail makes the next email feel familiar.
We ran the full cost-per-meeting comparison of both channels in cold call vs cold email. The short version: email is cheaper per touch, the phone gets you a conversation faster, and sequences that combine both beat either one alone.
What has really changed in the last ten years?
Several things, and all of them work against volume-first calling.
| What changed | Effect on cold calling | What offsets it |
|---|---|---|
| Unknown-caller screening on smartphones | Fewer pickups from unrecognized numbers | Local numbers, monitored caller ID reputation |
| Carrier spam labeling of high-volume numbers | A flagged number barely connects | Volume spread out, numbers checked regularly |
| More self-directed buyers (Gartner, 2025) | Less patience for a generic pitch | Researched calls with a specific reason to call |
| Consumer consent rules in several countries | Consumer cold calling shrinks | B2B is treated differently, but still regulated |
| Easier access to contact data | Direct dials and mobiles easier to find | Lists verified before calling |
The common thread: list quality and preparation matter more than they did ten years ago, and raw volume matters less. That matches the spread Gong observed between the average and the top quartile.
How do you know whether cold calling works for you?
Studies give you orders of magnitude. Your answer is in your own numbers, as long as you measure them with clear definitions.
- Measure connect rate over at least a few hundred dials, separating real conversations, voicemails and invalid numbers. A very low rate points to the list or the time slot, not the rep.
- Measure how many conversations turn into meetings. A low rate points to your opener, your objection handling or your discovery questions.
- Compare against cost. Divide a rep’s fully loaded cost by the number of meetings held. Our volume benchmarks are in how many cold calls per day.
- Test long enough. A few days of calls won’t settle it, especially while a rep is still learning the pitch. Call reluctance also skews early results.
In Skipcall, the dashboard automatically sorts every attempt into one of seven statuses (connected, voicemail, wrong number, wrong country, cancelled, not connected, failed) and shows two separate conversion rates: on connected calls, which measures the pitch, and on all calls, which measures the list and targeting. That is exactly the split you need to know which lever to pull. Our article on call metrics and KPIs covers these measures in detail.
What to remember
Cold calling isn’t dead, and recent data shows it: 5.4% of dials connect for the average rep, 13.3% for the top quartile (Gong Labs, July 2024), and the success rate measured by Cognism and WHAM climbed back to 2.7% on 2025 data. On the buyer side, most still accept a relevant call, and senior executives are the most phone-friendly group.
What died is generic, high-volume calling. Buyers research on their own, screen unknown numbers and avoid irrelevant suppliers. The teams that get results call less at random: verified lists, the right time windows, healthy caller IDs, a specific reason to call.
So the real question isn’t whether the phone works, but what separates your calls from the top quartile’s, and then fixing the weakest link first: the list, the time slot, the opener or the conversion. To see how long each call should last depending on how it ends, read our article on how long a cold call should last.
