TL;DR:
- Cold calling is a sales call to someone who has never interacted with your company. It still books meetings in 2026, and the teams it works for treat it as a process rather than a personality test.
- The average cold call success rate, measured dial to booked meeting, sits at 2.7%. It takes 8 to 10 dials to reach one decision maker.
- Only 19 out of 100 dials on a company line reach the target. 41% get absorbed by gatekeepers and 27% go straight to voicemail.
- Switchboard lines connect at 3% to 5%. Verified mobile numbers connect at 11% to 14%, and skip the gatekeeper 84% of the time.
- 82% of buyers say they would accept a meeting from a cold call, as long as the context is relevant to them.
- US Do Not Call rules largely exempt B2B calls. TCPA autodialer rules, state laws, GDPR, and PECR do not.
Cold calling gets argued about more than it gets measured. One camp says the channel died in 2015. The other posts a screenshot of a booked meeting and calls it evidence.
Neither side looks at the funnel. A cold call has four multipliers in it. How many numbers reach a real person. How many of those people answer. How many conversations turn into a meeting. And how many of those meetings happen. Move any one and your pipeline moves. Move the first and it moves a lot.
This guide covers the full process, the scripts, the objections, the voicemail problem, the legal rules, and the numbers to track. I'll also be blunt about where cold calling stops working, because that matters as much as where it does.
What is cold calling?
Cold calling is a phone call to a prospect who has shown no prior interest in your company. No demo request, no downloaded guide, no reply to an email. The rep starts from zero.
It sits inside the wider B2B telemarketing family, though the modern version looks nothing like the mass dialing that name suggests. The goal of a B2B cold call is almost never a sale. It's a meeting. Reps who try to close on the first call push prospects away, because nobody buys software from a stranger in four minutes.
Cold calling versus warm calling
A warm call reaches someone who has already touched your brand. They visited a pricing page, joined a webinar, or replied to a colleague. The gap between cold calling and warm calling is prior signal, not prior relationship.
Here's where I'd push back on the standard framing. Take a researched cold call to a verified mobile at a company that just posted a relevant job opening. It beats a warm call to someone who downloaded a guide by accident. Signal quality matters more than the label.
How cold calling changed
Three shifts reshaped the channel, and each one changed what good looks like.
Remote work broke the office line. Desk phones ring in empty buildings. Mobile numbers became the only reliable route to a decision maker, which turned data quality from a nice-to-have into the whole game.
Carrier spam filtering arrived. US carriers now score inbound calls and label suspected spam before the phone rings. Your prospect sees "Spam Likely" and never hears from you.
Buyers got harder to surprise. They already know your category. The pitch that worked in 2018, where you explained what your product does, now sounds like a waste of their afternoon.
Does cold calling still work in 2026?
Yes, and the honest answer has conditions attached.
SMARTe surveyed 1,200+ sales professionals across North America and Europe and asked them to break down what happens on a dial. The picture is less about rejection than teams assume.
What happens on 100 dials to a company lineShareBlocked, screened, or absorbed by a gatekeeper41%Straight to voicemail with no live answer27%Reached the target directly19%Reached, then immediately deflected13%
Roughly 6 of those 100 dials become a real conversation. That's where the 2.7% success rate comes from, and none of it is a script problem.
Then look at what changes with a verified mobile number. Reps dialing verified mobiles skipped the gatekeeper layer in 84% of attempts. Connect rates moved from the 3% to 5% range on switchboard lines to 11% to 14%. Same prospect, same message, different number.
The full B2B prospecting statistics go deeper, but that one comparison carries the argument. The channel isn't dying. The gap between average and elite is widening, and it tracks almost entirely to data.
The math that decides your workload
Work backward from a meetings target. Say a rep needs 10 meetings a month, and roughly 5% of real conversations turn into a booked meeting. That's 200 conversations.
Now the connect rate sets the workload:
Same rep, same script, same target. One version demands 250 dials a day. The other demands 83.
And it compounds. At 250 dials a day nobody researches anything, so the conversation-to-meeting rate slides from 5% to maybe 3%. That pushes the dial requirement past 8,000 and removes any prep time that was left. The list problem eats the skill problem.
At 83 dials a day a rep has three minutes per prospect to find something real to say. That's the difference between a call that sounds researched and one that sounds like a robocall.
Let me qualify that. Those connect rates sit at the midpoint of the switchboard and mobile ranges, and your segment will land somewhere else. Seniority, region, and industry all move it. Run the model on your own numbers before you trust mine.
Where cold calling doesn't work
Arguments over whether cold calling is dead usually skip this part.
Three situations kill it. The buyer has no phone presence. The deal size can't justify a rep's time. Or the product needs a long technical evaluation before anyone will take a meeting. Selling a $30-a-month tool to developers? Skip the phone. Selling a compliance platform to a CFO at a mid-market bank? Pick it up.
The cold calling process, step by step
This is the phone prospecting process I'd hand a new rep. Nine steps, and the first four happen before anyone dials.

1. Set one objective for the call
Decide what success looks like before you dial. A booked meeting, a qualified next step, or one piece of information you didn't have.
One objective, not three. Reps who try to qualify, pitch, and book in the same four minutes do all three badly. And never raise pricing on a first call. You haven't earned the right to discuss budget, and bringing it up early hands the prospect an easy reason to end the conversation.
2. Build the list before you build the script
Your cold call list sets your ceiling. The script decides how close you get to it.
Start with your ideal customer profile. Industry, headcount, tech stack, funding stage, region. Then check your last ten closed-won deals against it. If the list doesn't match the deals, the list is wrong and no script will save it.
Name three roles per account, not one generic decision maker. The person holding the budget rarely feels the problem first.
3. Research the account in five minutes
I've watched reps spend forty minutes building a call plan for one prospect. That's not research, that's avoidance.
Give yourself five minutes and look for four things:
- A recent announcement, funding round, or product launch
- How long the prospect has held the role
- One position their team is hiring for right now
- A peer company you already sell to in their orbit
Any one of those gives you a first sentence that proves you didn't pull the number off a list.
There's a smarter version of this that senior reps use and juniors rarely try. Before calling the VP, call two people below them. A sales development representative or a manager will tell you what's broken in the team, in their own words. You then walk into the VP call carrying a problem their own staff described. That beats anything you'd find on LinkedIn.
4. Get mobile numbers, not switchboards
Any cold calling database is a perishable asset. People change jobs. Companies rotate switchboards. Mobile numbers follow the person while your CRM record stays with their old employer.
B2B direct dials used to be the goal. With hybrid and remote work, mobile numbers matter more, because a direct dial to an empty desk rings out the same way a switchboard does.
The gap is measurable. Reps working verified mobile numbers reach the decision maker 3x more often than reps on any other number type. And 67% of cold calls that end in a booked meeting used a direct mobile on the first or second dial.
Mobile numbers also age slower. Business email decays at roughly 2.3% a month because it dies the day someone offboards. Mobile numbers decay at roughly 1.1%, because the phone leaves with the person. When a contact changes jobs, their old email is a void and their mobile still rings.
Three checks take a minute each and save hours:
- Confirm the title is current. Tracking job changes matters here, because a move makes the record wrong even when the number works.
- Confirm the number is a mobile or direct line, not a main reception desk.
- Confirm someone verified the record recently. B2B data decay runs faster than teams expect, and a list bought six months ago is not the list you tested.
Watch your caller reputation. US carriers run authentication and spam scoring on inbound calls. A number that dials hundreds of times a day, collects hang-ups, and never receives a call back looks exactly like a robocall to that system. Four fixes:
- Spread volume across multiple outbound numbers instead of hammering one
- Register those numbers with the verification programme your dialer supports
- Match the area code to the region you're calling
- Cut wasted dials, because fewer dead numbers means fewer hang-ups
That last one protects the number you dial from, not just the number you dial.
5. Write a script you can abandon
A cold calling script is a structure, not a paragraph to recite. Four parts: opener, one qualifying question, one proof point, one ask. Everything else you improvise.
Good scripts get shorter over time. If yours has grown a paragraph explaining your product architecture, cut it. The prospect is deciding whether to keep listening, not whether to buy.
Rehearse it out loud. Reading a script silently and delivering it on a live call are different skills, and the gap shows in the first ten seconds.
6. Open the call
You get about eight seconds. Say who you are, say why you're calling this specific person, then stop talking.
Cold calling opening lines that work name something concrete. Openers that fail lead with a permission question like "did I catch you at a bad time?" which hands the prospect a free exit.
One approach I like more than I expected to: name the call for what it is. "This is a cold call. Want to give me thirty seconds to explain why?" It disarms people because it's honest, and honesty is rare enough on the phone to be interesting.
7. Ask questions that surface the problem
Once you're past the opener, stop pitching and start asking. The rep who talks less learns more.
Two techniques worth stealing. Mirroring: repeat the prospect's own words back as a question. If they say "our team finds all their leads manually," you say "all of them, manually?" Things that sound normal in your own mouth sound absurd coming back at you. That pause is where a prospect starts questioning their own process.
Open questions over closed ones. "How does your team handle territory coverage today?" gets you somewhere. "Do you have a solution for that?" gets you a yes or a no and ends the thread.
This is also where you qualify. If there's no fit, say so and get off the phone. A rep who books meetings with people who'll never buy is stealing from their own AE.
8. Handle the objection
Cold calling objections are not rejection. An objection means the prospect cares enough to push back. That puts you ahead of the 68% of dials that never reach a live person at all.
The framework worth learning is Feel, Felt, Found. Acknowledge how they feel, tell them others felt the same, then share what you found. It works because it removes the argument from the conversation.
There's a simpler version I prefer for cold calls: ask a question back. When someone says "we already have a tool for that," respond with "what does it not handle well?" You've turned an exit line into a discovery question.
9. Ask for the meeting, then close
If the call is going well, ask directly. Reps lose more meetings to vagueness than to objections.
Three closes that hold up:
- The demo swap. "Rather than me describing this, would you be against seeing it for fifteen minutes?"
- The specific time. "Thursday at 10, or does Friday morning work better?" Two options beat an open question.
- The freebie. Offer something they keep whether or not they buy. A data sample, a benchmark, a short audit.
If they're not ready, book the follow-up on the call. "I'll call you the first week of next month" is not a follow-up. A calendar invite is.
10. Review the call afterwards
Log the outcome, the objection wording, and the next step while it's fresh. Then listen back to at least two of your own calls a week.
Reviewing your own recordings is uncomfortable and it's the fastest way to improve. You'll hear the sales call mistakes nobody catches live: talking over people, rushing the ask, explaining things nobody asked about. A manager sitting ten desks away never hears any of it.
Cold calling scripts and examples
Every script below is a structure to adapt, not a line to read aloud. Swap the bracketed parts for your own market. Each one runs 45 to 90 seconds if the prospect stays on.
The four-part frame behind all of them
If you build your own, keep this shape:
Notice what's missing from all four scripts. No product description. No feature list. No "we're the leading platform for." A cold call is a filter, not a demo.
The six objections you'll hear every day
Rehearse these. Improvising at 3pm on a Thursday goes badly.
Build your own version of this table from your call recordings. Do the same for the sales objections your AEs hear later in the deal. Your prospects use specific wording, and the wording matters. A response written for "I'm not interested" lands differently against "we're not looking at anything right now."
What to do when you hit voicemail
27% of dials to a company line go straight to voicemail. Reps treat that quarter of their day as dead time, which is exactly why it stays dead.
A voicemail that works does three things and stops:
- Names the prospect and yourself inside the first four seconds
- Gives one specific reason for the call, tied to their company
- Says you'll follow up by email today, then names the subject line
That last part is the one reps skip. A voicemail plus a matching email lands twice, which is the practical answer to the cold email versus cold call argument: run both.
Keep it under 20 seconds. A long voicemail gets deleted at the point where it starts explaining your product, and that point arrives faster than reps think.
Do not leave the same voicemail twice. If attempt one hit voicemail, attempt two should reference it. "Second time trying you, I'll stop after this one" beats a repeated pitch. It's honest, and it sets a real deadline.
How to get past the gatekeeper
41% of dials get blocked, screened, or absorbed before they reach the target. Receptionists and executive assistants are doing their job, which is filtering people who waste their boss's time. Reps who treat getting past the gatekeeper as a battle lose here.
Four things that work:
- Be direct about who you are and why you're calling. Evasion is what they're trained to catch.
- Ask for help rather than access. "Can you point me in the right direction?" gets further than "can you put me through?"
- Ask them who owns the problem. They often know the org chart better than the website does.
- Call outside standard hours. Executives pick up their own phone at 8am when nobody is screening.
The tricks that circulate in sales forums stopped working years ago. Assistants compare notes, and a rep who lies to get through burns the account for everyone behind them.
Verified mobile numbers sidestep the layer entirely, and skip it in 84% of attempts. That's another reason the list work pays for itself.
The best time to cold call
The best time to cold call shifts your odds, though less than list quality does. SMARTe analysed over 8 million call attempts by hour and weekday, and the pattern holds well enough to plan around. It's still smaller than the gap between a switchboard number and a mobile.
Treat any published window as a hypothesis. Field service managers answer early. Engineers answer late. Finance teams disappear at month end. Your own last 500 dials outrank anyone's benchmark.
What surprises teams more is the attempt count. It takes 8 to 10 dials to reach one decision maker once you factor in gatekeepers. So call the same prospect at different times of day before you write them off. Building a sales cadence around repeat attempts beats one well-timed dial.
The metrics that tell you what's broken
These five SDR metrics each tell you something different. Averaged together they hide the problem.
- Connect rate. Live conversations divided by dials. Under 10% points at your data, not your reps.
- Conversation-to-meeting rate. Meetings booked per real conversation. Low here means targeting or pitch.
- Attempts per contact. How many times you dial before quitting. Two is too few.
- Show rate. Booked meetings that happen. High booking with low show usually means reps pushed for a soft yes.
- Meetings to opportunity. Whether the discovery call that follows turns into real pipeline.
Two benchmarks worth holding yourself against. 4 to 6 quality conversations per rep per day. And 18 to 24 meetings per rep per month at the top quartile. If your team is dialing hard and landing well under that, the input is wrong, not the effort.
Conversation intelligence software records and scores calls so coaching runs on evidence instead of memory. Without it, a manager coaches from a rep's summary of a call the rep already misremembers.
Is cold calling legal?
Short section, because the rules are simpler than sales teams fear and the penalties are worse than they assume.
In the US. Federal Do Not Call registry rules largely don't cover business-to-business calls, with narrow exceptions. That exemption gets misread as a free pass. TCPA restrictions on autodialed and prerecorded calls still apply, and the FCC's telemarketing rules set out where consent applies. The Telemarketing Sales Rule bans deceptive practices regardless of who you're calling. State do not call laws add their own restrictions, and some reach further than federal rules.
In the UK and EU. The ICO's B2B marketing guidance confirms that GDPR and PECR both apply to business calls using personal data. A named contact at a company is personal data. The rules on GDPR and cold calling deserve a proper read before you scale into those markets. UK callers also need to screen against the Telephone Preference Service and its corporate equivalent.
By industry. Financial services carry extra rules. In the UK, FCA regulations ban cold calls about some investment products outright.
Do the compliance work at the list stage, not the dial stage. Screening a list once is cheap. Unwinding a complaint is not.
Dealing with rejection and call reluctance
Nobody talks about this part honestly, so here it is.
Cold calling involves being told no dozens of times a day by strangers. That wears on people. Call reluctance is real, it affects experienced reps as well as new ones, and telling someone to "just push through" helps nobody.
What does help:
- Separate the outcome from the identity. A no is information about timing, not a verdict on you.
- Track leading activity, not outcomes. You control dials and attempts. You don't control who picks up.
- Block calling time. Reps who dial in scheduled blocks make more calls than reps who dial whenever they feel ready, because nobody ever feels ready.
- Roleplay the hard calls. Practising the objection you dread removes the reason to avoid the phone.
The reps who last aren't the ones who stopped feeling nervous. They're the ones who dial anyway.
Where SMARTe fits
SMARTe won't write your opener or handle your objections. It changes one variable in the math at the top of this guide: how many of your numbers reach a live person.
Coverage runs across 289M+ verified B2B contacts and 66M+ company profiles. US mobile and direct dial coverage sits above 75%, with global direct dial coverage above 50% across EMEA, APAC, and LATAM. Verification happens in real time rather than on a batch schedule, so the number you dial on Thursday reflects Thursday. Job change tracking flags a moved decision maker before your rep finds out the hard way.
I'll be honest about the limits. SMARTe doesn't dial for you, doesn't record your calls, and won't fix a targeting problem. If your connect rate is healthy and your meetings still aren't landing, the data isn't your bottleneck. Spend the money on coaching instead.
See how SMARTe finds verified mobile numbers in your target accounts
Conclusion
The reps who last at this job aren't the ones with the smoothest delivery. They're the ones who stopped treating the phone as a test of personality and started treating it as a system with inputs they control.
You control who's on the list. You control how many attempts each contact gets. You control whether you rehearsed the objection answers or you're inventing them mid-call. You don't control whether any given person picks up.
Fix the inputs, run the volume, and let the arithmetic do what it does.



