Sales persistence: what attempt data shows

The number everyone quotes is from 1942

Nearly every article on persistence in sales carries the same statistics on how many follow-ups a sale needs and how many reps quit after the first attempt, credited to an American sales association with an official-sounding name. SMEI, the organisation that inherited that association's legacy, wrote in 2021 that the origin is a survey the Long Island chapter ran among its own members in 1942, fewer than 40 people, never published as a dataset. In 2014 VentureBeat searched the registers of the US tax authority and the Better Business Bureau and found no such association, even though the statistic was appearing on more than 300 web pages. So there is no study. What there is instead is four real datasets, and they tell a different story from the folklore.

Three calls capture 93 percent of the conversations

Cognism's 2025 cold calling report, built on dialer data together with We Have A Meeting, counted 41,936 connected calls and 27,513 conversations. Conversations fell 73 percent from the first attempt to the second and a further 60 percent from the second to the third. After three attempts, 93 percent of all the conversations that would ever happen had already taken place. After five, 98.6 percent. The previous year the same company's funnel showed what a single attempt is worth: 55,701 dials produced 9,247 connected calls, 5,265 conversations and 254 booked meetings. Ring the same number a seventh time and you are chasing the last percentage point, paying with time that would have produced more conversations on fresh numbers.

One detail in the same data explains why the second and third attempts are still needed: 26.85 percent of people who call back after a missed call can be reached, well above a cold first attempt. So leave a reason to call back, try again at a different time of day and on a different weekday, and stop after the third.

Six attempts for leads who came to you

Inbound leads follow a different curve, because the person has already shown interest. Velocify's 2013 study The Ultimate Contact Strategy, based on inbound web leads in mostly consumer-facing sectors, found that 93 percent of the leads that ever converted had been reached by the sixth call attempt. The curve flattens after six. What matters most with inbound leads, though, is not the number of attempts but the speed of the first one. The Lead Response Management study, built on three years of data from six companies with more than fifteen thousand leads and over a hundred thousand call attempts, found the odds of qualifying a lead dropped 21-fold when response time slipped from five minutes to thirty. Chili Piper mystery-shopped B2B vendors in 2022 and measured an average response time to demo requests of four hours and fifty minutes.

Five to eight touches, across every channel

Count every channel together and several sources land close to one another. RAIN Group's prospecting study, with 488 buyers representing 4.2 billion dollars in purchases, found that a first meeting took eight touches on average and that top performers managed it in five. Orum's platform data from more than a billion dials shows the best SDRs averaging 5.7 touches per prospect before connecting. Salesloft's 2026 US benchmark, from 500 revenue leaders, measured 35.2 touches per qualified opportunity, which includes every prospect who never answers. And seven in ten buyers in RAIN Group's study said they want to hear from sellers when they are looking for new ways to improve results. Persistence is not the problem. The wrong kind of persistence is.

The Swedish data points to a different problem than too many attempts: too few. Säljmotorn analysed 9,506 cold B2B emails to unique recipients, sent between 8 June and 28 July 2026 in real commercial campaigns. The reply rate was 6.6 percent, and 25 percent of the replies only arrived after a follow-up email. Yet only 46 percent of recipients got any follow-up at all, and 16 percent got two. The dataset is small and comes from a vendor, but the direction matches Gong's analysis of more than 28 million cold emails: the average rep has to send 344 cold emails to land one meeting, while top performers get 4.2 times the reply rate, partly by not pitching in the email, which otherwise cuts reply rates by as much as 57 percent.

The stopping rule

Put the four datasets together and you get a rule that fits on a sticky note. It applies to a prospect who has not answered and has not said no.

1. Week one: three call attempts at different times and on different weekdays, with a voicemail that states the reason, plus a short email on the day of the first call. Gong's data from 300 million calls shows a call attempt nearly doubles the email reply rate, from 1.81 to 3.44 percent, even when nobody picks up.
2. Week two: a follow-up email with a new reason, not a reminder, and a touch on LinkedIn.
3. Week three: a final email that makes it easy to say no. Then the contact is parked for 60 to 90 days.
4. Re-entry only on an event: a new role, new funding, a contract running out, a post that shows the problem exists.

That is seven to eight touches in three weeks, which is RAIN Group's average and close to Orum's 5.7 for the best. Every figure in the rule is Anglo-American or comes from a Swedish vendor with a small sample. Measure your own curve: what share of your conversations happens on attempt one, two and three? If the curve falls the way Cognism's does, the rule holds. If it falls more slowly, add an attempt.

What persistence cannot fix

Gong's review of 300 million calls shows where the real difference sits. The average rep connects with 5.4 percent of the people they dial, the best with 13.3 percent. Of the conversations, 4.6 percent turn into a meeting for the average rep and 16.7 percent for the top. Worked through, the average rep needs about 19 dials per connect and the top rep 8, and on 800 dials a month that is 2 meetings against 18. None of those numbers moves because you call a fourth time. Orum's data says the same per audience: admins answer 7.8 percent of the time, directors and managers 4.8, VPs 3.8 and the C-suite 4.0. Connect rate is decided by the list, the number and the timing. Meetings per conversation are decided by what the rep says in the first few minutes.

HubSpot's 2025 survey of 379 sales professionals shows what most reps actually do: 55 percent of daily cold callers make three to five attempts before moving on, 22 percent make one or two, 13 percent make six to eight. Most are already close to the stopping rule. What separates 2 meetings from 18 is not more attempts but better calls, and that can only be trained by having a lot of them. In Salesprep's Cold call module you practise the opening and the reason for the call against an AI buyer, and in the Follow-up module the second attempt against a buyer who did not answer the first time. Every call is scored on seven metrics with a written comment per score. Three calls in the Cold call module are free, no card required, and Follow-up is available on Pro and up.