You get about ten seconds. In that window the prospect decides whether this is a conversation or an interruption to be ended, and almost nothing you say later reverses that decision.
Three things make an opener survive it: you name yourself, you name the fact that this is a cold call, and you ask for something small and specific.
What the call data says about opening lines
Two published studies give success rates for named openers. Both are from Gong Labs.
| Opener | Success rate | Against baseline | Source |
|---|---|---|---|
| "Heard the name tossed around?" | 11.24% | baseline not stated | Gong Labs, 2024 |
| Permission-based: context, own the cold call, ask permission | 11.18% | baseline not stated | Gong Labs, 2024 |
| "How have you been?" | 10.01% | 6.6x the 1.5% baseline | Gong Labs, 2018 |
| "How's your day going?" | 7.6% | baseline not stated | Gong Labs, 2024 |
| "How are you?" | 5.2% | 3.4x the 1.5% baseline | Gong Labs, 2018 |
| "The reason for my call is..." | not published | 2.1x baseline | Gong Labs, 2018 |
| "Did I catch you at a bad time?" | 0.9% (2018), 2.15% (2024) | 40% below baseline (2018) | Gong Labs, both |
Sources: Gong Labs, "Cold call opening lines", 90,380 cold calls, success counted as a booked meeting (archived 2026-09-30). Gong Labs, "The best and worst cold call openers", Armand Farrokh, 2024, stated as 300M+ cold calls (archived 2026-09-30).
Read the table with three limits in view. The studies are six years apart and do not share a definition: the 2018 study counts a booked meeting as success, and the 2024 article never says what success means or what its baseline is. Gong sells conversation-intelligence software, so these are a vendor's figures about its own customers' calls, which lean towards funded B2B software teams. And neither study publishes how many calls used each opener. They are still the largest per-opener figures MRR Nerds found published, and they agree on the one point that matters most: the permission and context openers sit at the top, and "Did I catch you at a bad time?" sits at the bottom of both.
The permission openers
These work because they turn the prospect from someone being talked at into someone who made a decision. People who agree to give you thirty seconds listen to the thirty seconds. Gong's 2024 version of this pattern scored 11.18 percent.
- "Hi Sarah, it's Alex from Northwind. I know this is a cold call. Can I have 30 seconds to tell you why I called, and then you can decide?"
- "Sarah, Alex from Northwind. This is a cold call. Do you want to hang up, or do you want the 30-second version?"
- "Hi Sarah, Alex here from Northwind. Is now a terrible time?"
- "Sarah, it's Alex from Northwind. I'll bet you don't know me. Thirty seconds and you can decide if it's worth more?"
Number three is Gong's worst performer with one word changed. No published dataset MRR Nerds found measures it separately, so what follows is reasoning, and you should test it before trusting it. The exaggeration makes a flat yes feel out of proportion: a mildly busy prospect says "no, go on" to terrible and "yes, actually" to bad.
The rule that makes these work: honour the thirty seconds. At thirty seconds, stop and check in. A rep who blows through their own stated limit teaches the prospect in the first minute that nothing they say is reliable.
The social-proof opener
- "Sarah, we work with Acme and Globex on their outbound. This is Alex from Northwind. Have you heard the name tossed around?"
This is the structure Gong's 2024 analysis ranked first at 11.24 percent: lead with peer companies the prospect will recognise, then introduce yourself, then ask whether they have heard of you. Either answer moves the call forward. A yes gives you a reason they already have to listen, and a no invites the one-sentence explanation you wanted to give anyway. Only name customers who have agreed to be named.
The specific-research openers
These only work when the fact is specific and checkable. Used with a generic fact they are worse than the permission openers, because they promise relevance and then fail to deliver it.
- "Sarah? Alex from Northwind. I called because you're hiring three AEs this quarter and I have a specific reason to think your ramp is going to hurt. Can I take a minute?"
- "Hi Sarah, Alex from Northwind. You opened a second location in March. I called about one specific thing that usually breaks at that point. Worth 30 seconds?"
- "Sarah, it's Alex from Northwind. I saw your post about [thing] last week, and it's the reason I'm calling instead of emailing."
- "Hi Sarah, Alex from Northwind. [Competitor] switched to us in June for one reason, and I wanted to check whether that reason applies to you."
Openers six to nine all carry a stated reason for the call, the element Gong's 2018 data credits with 2.1 times the baseline success rate.
Cap your research at sixty seconds per prospect. One specific fact is enough. Two is showing off, and a rep who researches for ten minutes per call makes twelve dials a day and learns nothing from a sample that small.
The pattern-interrupt openers
Higher variance. They work well with senior titles and heavily called markets, and land badly with people who dislike being handled.
- "Sarah, this is a cold call. Would you like to hang up now or in thirty seconds?"
- "Hi Sarah, Alex from Northwind. I'll level with you: I have no idea if this is relevant to you. One question and we'll both know."
- "Sarah? Alex. I've been putting off making this call all morning because I hate cold calling as much as you hate receiving them. Can I get it over with?"
- "Hi Sarah. I'm calling out of the blue and I know that's annoying. Thirty seconds?"
Number twelve earns a laugh more often than anything else on this page, and the laugh is the permission. It also cannot be faked, so only use it if it is true.
The referral and warm openers
The strongest openers available, and most teams underuse them because asking for the referral is uncomfortable.
- "Hi Tom, it's Alex from Northwind. [Name] at [Company] suggested I call you."
- "Tom, Alex from Northwind. [Name] mentioned you're dealing with the same thing they were. Is that fair, or has [Name] oversold it?"
- "Hi Tom, we met briefly at [event] in March. Alex from Northwind. How have you been?"
Put the name in the first sentence. It is the entire asset and burying it wastes it. Ask permission before using someone's name, because using it without asking damages two relationships at once.
Number sixteen is where "How have you been?" belongs. It scored 10.01 percent in Gong's 2018 data, the highest single figure in either study, and it implies the two of you have spoken before. With a real prior touch that implication is true. On a stranger, the prospect who notices it has a reason to doubt everything that follows.
The direct openers, for very senior titles
- "Sarah? Alex from Northwind. You've got four people doing manual invoice checks. I sell software that does 80% of that. Is that a problem anyone there is complaining about?"
- "Sarah, Alex from Northwind. One question and I'll go: who owns [process] at your company these days?"
Number eighteen is a research call dressed as a sales call, and it is the most underrated opener here. It is one question, it is easy to answer, and it either gets you the right name or tells you the account is not worth working.
The five that kill calls
"Did I catch you at a bad time?" The only opener both Gong studies put last: 0.9 percent against a 1.5 percent baseline in 2018, 2.15 percent in 2024. It invites the easiest yes in the conversation.
"Do you have a few minutes?" Everybody knows a few minutes means fifteen. Ask for thirty seconds and mean it.
"I'm calling to see if you'd be interested in..." The prospect has no basis for interest yet, because you have not established that they have a problem.
"We help companies like yours to..." The prospect hears a broadcast. Every vendor says this.
"Am I speaking with the person responsible for..." Reads as a screening question from a robot. Ask for the person by name, or ask who owns the process.
Earlier versions of this page listed "How are you today?" here as the worst opener of all. Gong's data does not support that: "How are you?" booked at 5.2 percent in 2018 and "How's your day going?" at 7.6 percent in 2024, both above the 2018 baseline. It still trails the permission openers, so it is a weak choice. It is no longer on this list, and the change is recorded on /corrections.
Testing your own
Change one variable at a time, run each version for at least a hundred dials, and measure the share of connects that survive past twenty seconds.
Meeting rate is affected by everything downstream of the opener, so over a few hundred dials it is too noisy to judge an opening line. Survival past twenty seconds measures exactly the thing the opener does. Gong's studies could use booked meetings because they had tens of thousands of calls; a single rep's test has hundreds.
Record the calls. The gap between what reps believe they said and what they actually said is large, and it is only visible on a recording.
Frequently asked questions
- What is the best cold call opening line?
- In Gong's 2024 analysis the top two openers were "Heard the name tossed around?" at 11.24 percent success and a permission-based opener at 11.18 percent. The permission version is the easiest to learn: "Hi [name], it's [your name] from [company]. I know this is a cold call. Can I have 30 seconds to tell you why I called, and then you can decide?" It names you, names the cold call and offers an exit.
- Should you ask 'did I catch you at a bad time'?
- No. Both of Gong's studies put it last. Across 90,380 cold calls analysed in 2018, "Did I catch you at a bad time?" booked meetings 0.9 percent of the time against a 1.5 percent baseline, which is 40 percent worse than average. Gong's 2024 analysis put it at 2.15 percent, the lowest of the four openers it validated. It invites a yes, and a yes ends the call.
- Should you say 'how are you' on a cold call?
- Gong's data says it does less harm than its reputation. In Gong's 2018 study of 90,380 cold calls, "How are you?" booked meetings at 5.2 percent, 3.4 times the 1.5 percent baseline, and "How's your day going?" scored 7.6 percent in Gong's 2024 analysis. Both still trail a permission-based opener at 11.18 percent. This page used to call it the worst opener, which the data does not support; the change is logged on our corrections page.
- Does 'how have you been' work on a cold call?
- It posted the highest single number in Gong's 2018 study: 10.01 percent against a 1.5 percent baseline across 90,380 cold calls, 6.6 times average. Gong attributes it to a pattern interrupt. The line implies you have spoken before, so MRR Nerds' advice is to use it only with a real prior touch: an event, an earlier email thread, a mutual contact. On a true stranger, a prospect who notices the borrowed familiarity has a reason to doubt the rest of the call.
- How long should a cold call opener be?
- Under ten seconds before the prospect speaks, which is roughly twenty-five words. The opener's only job is to buy permission for the next thirty seconds, and every extra word spends goodwill you have not earned yet. If your cold call opener already contains a value proposition, cut it and save the value proposition for after the prospect says yes.
How to cite this page
Abhi Chawla. "The best cold call opening lines, ranked by what happens next." MRR Nerds, 2026-09-30. https://www.mrrnerds.com/guides/cold-call-opening-lines. Accessed 2026-10-16.
The underlying vendor data is published at /data under CC BY 4.0. Cite the dataset for numbers, this page for the analysis.