It is the first question every sales leader asks and the one most benchmarks answer badly: how many cold calls to book a meeting? The answers you find online range from "it takes 209 calls" (a number with no source that everyone quotes) to "cold calling is dead" (from people selling something else). Here is the honest version: a range, the maths behind it, and a calculator so you can set targets from your own numbers within a fortnight.
The short answer
For a decent UK B2B list worked properly, budget 60 to 100 dials per held meeting. That range covers most teams we see. Sharp lists with a strong opener get under 40. Tired bought data with one attempt each can take several hundred, which usually means the list is the problem, not the reps.
The number matters less than the funnel that produces it. Memorise this chain, because every target, forecast and coaching conversation in outbound hangs off it:
Dials → conversations → booked meetings → held meetings.
Three conversion rates connect the four stages: connect rate, meeting rate and show rate. Improve any one and the dials-per-meeting drops. Measure all three or you are managing blind. If you log outcomes as structured data, your dialler should hand you all three for free; if it does not, see our call dispositions framework.
The three rates, with honest ranges
Connect rate: dials to conversations (typical: 8 to 12%)
The connect rate is the share of dials that reach a live human conversation. Voicemails, no-answers, busy lines and gatekeeper blocks do not count. For screened UK B2B data, dialled across several attempts and dayparts, 8 to 12% is the normal band.
Below 5%, stop coaching the opener and interrogate the inputs first, in this order: list quality (disconnected numbers, TPS-listed contacts, job titles that never pick up), attempt count (most teams quit after one or two dials), and number health (flagged caller ID tanks answers; see caller ID and spam reputation). Only then look at talk tracks. Connect rate is a systems metric far more often than a skill metric.
Meeting rate: conversations to bookings (typical: 1 in 8 to 12)
Of the humans you actually speak to, roughly one in ten agrees to a meeting on a reasonable proposition. This is the rate reps influence most: opener quality, personalisation that proves homework, qualification that respects both calendars, and a concrete close ("Thursday 10:15 or Friday 9:40?") rather than "I'll send some info."
Watch the denominator. Teams that count every picked-up hello as a conversation flatter this number and then wonder why bookings lag. A conversation is a substantive exchange with the right person, logged as such.
Show rate: bookings to held meetings (typical: 70 to 80%)
A booked meeting that no-shows cost you the dials twice over. Three in four showing up is normal with a decent confirmation process: calendar invite accepted while still on the call, a reminder the day before, and a same-day nudge. Below 60%, fix confirmation before you buy more dials. Above 90%, you are probably over-qualifying and leaving bookings on the table.
The calculator
Weekly dials needed equals target meetings divided by the product of the three rates:
Dials = target held meetings ÷ (connect rate × meeting rate × show rate)
Worked example for a rep who owes 3 held meetings a week at typical rates (10% connect, 1-in-10 meeting rate, 75% show rate):
3 ÷ (0.10 × 0.10 × 0.75) = 3 ÷ 0.0075 = 400 dials a week, or about 80 a day.
That feels like a lot until you divide it by dialling mode. At 80 dials a day, click-to-call will bury the rep in admin; power dialling with auto-next makes it a normal day. The mode has to match the maths, which is why activity targets set without reference to dialling method always fail.
Ready reckoner for one held meeting a week at three list qualities:
| List quality | Connect | Meeting rate | Show rate | Dials per meeting |
|---|---|---|---|---|
| Sharp (targeted, screened, warm-ish) | 15% | 1 in 7 | 85% | ~35 |
| Typical (decent bought or CRM data) | 10% | 1 in 10 | 75% | ~130 |
| Tired (old data, single attempts) | 5% | 1 in 15 | 60% | ~500 |
The tired row is the point. Nobody should be dialling at those economics; fix the list or buy a better one before hiring another rep. Screening matters twice here: it is a legal duty under TPS and CTPS rules, and screened lists simply convert better because every number is a real, contactable business.
Attempts: where the meetings actually come from
Most booked meetings come after the third attempt. The first dial catches whoever happens to be free; the pipeline lives in attempts three to eight, spread across mornings, afternoons and different weekdays over two to three weeks. Plan 6 to 8 attempts per contact, then retire the record with an explicit disposition so the list stays honest.
Two failure modes to watch. One-and-done calling: reps skim the cream and the list looks exhausted at 20% worked. Favourite-redialling: reps hammer the friendly numbers while hard contacts rot. Both are fixed the same way: attempt caps, retry rules and a queue that serves the longest-waiting eligible contact next instead of letting reps choose. That is what power dialling queues are for; see power dialling versus predictive for why sequential working beats parallel on small teams.
What to measure from week one
You do not need a data warehouse. Four numbers, reviewed weekly per rep and per list source:
- Dials (outbound attempts, nothing fancy).
- Connect rate (conversations ÷ dials).
- Meeting rate (bookings ÷ conversations).
- Show rate (held ÷ booked).
Cut all four by list source and the buying decision gets obvious: double down on sources with cheap meetings, kill or fix sources stuck in the tired row. Within a fortnight you will have replaced every benchmark in this article with your own numbers, which is the entire point. Benchmarks set the starting target; your dispositions set the real one.
The bottom line
Budget 60 to 100 dials per held meeting, work backwards through connect, meeting and show rates to set daily targets, make 6 to 8 attempts the default, and measure the four numbers weekly. Teams that do this stop arguing about effort and start arguing about lists and openers, which is where the meetings are.
If your bottleneck is working the queue fast enough once the maths is set, that is the problem Ringby exists to solve: campaigns with attempt rules, preview and power dialling, every outcome logged, and follow-ups that do not depend on memory. See how it works or check pricing: Solo £29, Team £49 per seat plus call usage.