Cold calling statistics: connect rates, success rates and dials per meeting

This page is for SDR managers setting activity targets, reps checking their own numbers, and founders deciding whether to add calling to their outbound. Every figure comes from the original report or dataset, opened in October 2026, with its sample and how it defines "connect" and "success". Most cold calling data is vendor research, because calling platforms and outbound agencies are the ones who see millions of dials. Each source is labelled.

Cold calling statistics at a glance

Why "cold calling success rate" means different things

The same phrase covers at least four ratios:

• Meetings per conversation. Cognism's 2.3% for 2025 and 4.82% for 2024, and Gong's 4.6% "set rate", divide meetings by conversations.
• Meetings per connected call. Gong's 2018 figure of 1.48% divides held meetings by connected calls, after removing no-answers and phone menus.
• Meetings per dial. Baylor's 330 calls per appointment, about 0.3%, divides by every dial, including unanswered calls and dead numbers.
• Conversations per connect. The 65.6% that some roundups call a "success rate" is only the share of connected calls that became a conversation.

Connect rates differ for the same reason. Gong defines a connect as a call "answered by a prospect", which gives 5.4%. Cognism's 20.5% "connected" isn't defined, and probably includes gatekeepers and wrong people. Cognism also reports 6.7% for its own sales team, against what it calls an industry standard of around 2%; that's a vendor describing its own team, with no sample given.

The cold calling funnel, stage by stage

Put the two largest datasets side by side and the averages land in the same place:

The dials-per-meeting row is our arithmetic: 1 / (0.054 x 0.046) for Gong's average rep, 1 / (0.133 x 0.167) for its top quartile, and 204,698 / 617 for WHAM. Gong's own example matches: 800 dials a month gives an average rep 2 meetings and a top-quartile rep 18.

So a typical calling program books roughly one meeting per 330 to 400 dials. The best reps book one per 45. Gong says those reps did two things differently: they prioritized direct dials, and they targeted the prospects most likely to buy, such as companies with a problem now. Neither is a script change. Both are list work, which is why a better list moves this funnel more than a better pitch. At volume, that research is the slow part; Bebop does it for each account in the list, with the people to call and what's happening at the company.

What works on the call

Gong has published two studies of cold call openers. Both are correlations, not experiments, and they use different baselines.

Two findings hold in both studies. "Did I catch you at a bad time?" was the weakest opener each time. And context helps: in the 2018 data, stating a reason for the call ("the reason for my call is...") went with a success rate 2.1 times higher. Successful 2018 calls were also almost twice as long as unsuccessful ones, but that's more likely a result of a good call than a cause of one. For openers built around a researched reason to call, use the outbound call scripts.

How many times to call, and when

Attempts. In WHAM's 2025 data, conversations fell 73% from the first call to the second and another 60% from the second to the third. Three attempts captured 93% of all the conversations the team had; five captured 98.6%. RAIN Group's 2017 survey found 43% of buyers who accept meetings at least sometimes say it's fine for sellers to try five or more times before getting through. Three to five attempts per contact, spread over two or three weeks and mixed with email and a voicemail, is consistent with both.

Time of day. The sources disagree:

Cognism adds its own caveat: a time that works for WHAM may not work for you. Treat timing as something to test in your own data, after list quality and the opener.

Do buyers still take cold calls?

RAIN Group's survey of 488 buyers (June to July 2017) found about 7 in 10 accept phone calls from providers they haven't worked with, and most of those give the caller 5 to 10 minutes. C-level and VP buyers preferred the phone more often (57%) than directors (51%) and managers (47%). The same survey found buyers prefer email overall, and the data is nine years old.

More recent activity data points the same way. In the Bridge Group's 2025 survey, phone-centric SDR teams averaged 56 dials and 4.6 quality conversations a day, against 28 dials and 3.4 conversations for email-centric teams. For the email side, see the cold email statistics.

The rules that shape a calling program

This is general information, not legal advice.

• United States. Most business-to-business calls are exempt from the FTC's Telemarketing Sales Rule, but the ban on misrepresentation has applied to B2B calls since 2024. Under the FCC's rule at 47 CFR 64.1200, autodialed or prerecorded calls to mobile numbers need prior consent, which matters when enrichment tools return personal cell numbers.
• United Kingdom. The ICO says that for B2B calls you need to screen against both the TPS and the Corporate TPS registers, and respect any business that has asked you not to call.

What these numbers can't tell you

• Platform data reflects platform users. Gong and Orum customers are phone-heavy teams with good tooling. WHAM is one outbound team. Your averages may be lower or higher.
• Definitions are often missing. Cognism doesn't define a connect; Gong's 2024 posts don't define success.
• Some data is old or a different market. Baylor's study is B2C real estate from 2011. RAIN Group's buyer survey is from 2017.
• Openers are correlations. A rep who says "How have you been?" naturally may differ from one reading it off a card.

How to measure your own cold calling numbers

Benchmarks tell you whether you're in the normal range. Your own funnel tells you what to fix. Log every dial with an outcome, and use fixed definitions:

Illustrative example. The company, rep and numbers are invented to show the arithmetic.

An SDR at Cobalt Freight Software (cobalt.example) logs one month:

• 1,100 dials (55 a day for 20 days) and 77 connects: a connect rate of 77 / 1,100 = 7.0%.
• 46 conversations: 46 / 77 = 60% of connects.
• 4 meetings booked: a set rate of 4 / 46 = 8.7%.
• 3 meetings held: a show rate of 75%.
• Dials per held meeting: 1,100 / 3 = about 367.

Her connect rate sits between Gong's average (5.4%) and top quartile (13.3%), and so does her set rate. Two ways to gain a meeting:

1. Better numbers. Lift the connect rate to 10% with more direct dials: 110 connects, 66 conversations, 5.7 booked, about 4.3 held.
2. Better targeting and opener. Lift the set rate to 12%: 46 conversations x 12% = 5.5 booked, about 4.1 held.

Either change adds about one held meeting a month from the same dials. With only 46 conversations a month, though, 4 meetings against 5 is within normal variation, so judge any change over a full quarter before you call it a win.

Widely quoted, but we couldn't trace the original

• "It takes 8 attempts to reach a prospect." Repeated on many pages; we found no original study of call attempts. The nearest traceable figure is RAIN Group's 2017 survey of 489 sellers, which counts touches across all channels to convert a prospect, a different measure (see sales follow-up statistics). WHAM's data suggests most conversations happen by the third call.
• "80% of sales need five follow-up calls." No original found.
• "4 to 5pm is the best time to call," often credited to a 2017 study. We couldn't find it in the original publisher's own posts.
• "Cold calling success rate is 0.3%." This is the Baylor study, but it measured real estate agents calling homeowners in 2011, not B2B sales.

Next step: build your openers and branches with the outbound call scripts, each starting with a checked reason to call.