Is cold calling dead? No, you're just bad at it (and here's the UK law)

The people saying "dead" and the people saying "alive" are both selling something

Search this question and page one is a tidy split. The dialler companies and the data vendors say cold calling is very much alive, and here are seven data-backed reasons. The LinkedIn evangelists and the keynote speakers say it died years ago, and here's what to buy instead. Neither group has sat on a dialler for a living, and not one of the nine pages I read mentions the law you have to follow to do it in this country.

Here's the one honest data point on the whole subject. When the British Chambers of Commerce and Intuit surveyed around 1,500 UK business leaders in 2025 about how they currently win customers, referrals came top at 78%, then social media at 63%, then the website, networking, SEO and email. Cold calling wasn't in the results, because it wasn't even offered as an option. The mainstream has moved on. Which is exactly why a call done well now stands out.

The real numbers, with the thing they're divided by

Every "success rate" you've ever read about cold calling hides a denominator. Success per dial, per answered call, per actual conversation and per meeting are four different numbers, and the vendors pick whichever flatters them. So here they are with the maths showing.

• Gong, 300 million calls (a vendor dataset from 2024): an average rep got a live answer on 5.4% of dials. Of real conversations, 4.6% became a booked meeting. Roughly one meeting per 400 dials.
• Cognism's 2024 report headlines a "4.82% success rate". Their own funnel: 55,701 dials, 9,247 answered, 5,265 conversations, 254 meetings. That's one meeting per 219 dials. The 4.82% is meetings divided by conversations, which is a very different sentence.
• Belkins, 175,000 dials across 2025 (an outbound agency): a live person on 9.9% of dials, 58% of those became conversations, 4.6% of conversations booked. About one meeting per 370 dials.
• The one non-vendor study, Baylor University in 2011: 50 estate agents, 6,264 calls, 28% answered, 19 appointments. One appointment per 330 calls, or one per 93 people who actually picked up.

That's what a numbers game looks like. And it matches the floor. On a dialler, three hundred plus dials a day was normal. The rule of thumb we worked to was that if one in ten of those became an actual conversation the data was good, and you'd book from maybe one in forty of the conversations. The most I ever booked in a day was nine, some of them callbacks, some straight off the dialler. The days that broke people weren't the days of rejection. They were the days of voicemail after voicemail after voicemail, when you couldn't get anyone on the line to reject you.

Two things decided whether the data was good: who bought it, and whether it had been cleaned. Purchased lists rotated in and out of the campaigns, and some of them were rubbish. Hold that thought, because it's also the legal bit.

The stats you've read are zombies

Before the law, a quick burial. Three numbers get pasted into every article on this subject and none of them survive a look at the source.

• "48% of salespeople never follow up, 80% of sales need five follow-ups, 92% give up after four nos." This whole family traces back to a 1942 survey of fewer than forty members of a Long Island sales club. The association it's usually credited to doesn't appear to exist. It's still on Salesforce's and Microsoft's websites.
• "82% of buyers accept meetings from cold callers." The real finding is from RAIN Group in 2018: 488 buyers, 82% said they accept meetings at least sometimes with sellers who proactively reach out. Any channel, not phone, and not a conversion rate. It gets re-dated to 2024 and 2026 with no new fieldwork behind it.
• "Cold calling has a 2% success rate." No publisher, no year, no denominator. It's a rounding of four different studies measuring four different things.

Is cold calling illegal in the UK?

No. But it's regulated, most people doing it have never read the rules, and the Information Commissioner's Office fines small operations, not just the multi-million call scandals. This isn't legal advice, it's the plain-English version of the ICO's own guidance, which you should read before you dial.

The law is the Privacy and Electronic Communications Regulations, PECR, and the bit that matters is Regulation 21. The ICO puts it like this: "In general, you must not make marketing calls to any number listed on the Telephone Preference Service (TPS) or Corporate TPS (CTPS), unless that person has specifically consented to your calls." You don't need consent to make a live business-to-business call. You do need to check first. In the ICO's words: "If you want to make live marketing calls, you must check phone numbers against these registers before you make the calls."

• Screen against both registers. The TPS covers individuals, which in practice includes sole traders and often partnerships. The CTPS covers limited companies, LLPs and public bodies. If you're not sure what a business is, screen both. Registrations take effect after 28 days, so screen before every campaign, not once a year.
• A public listing is not permission. Again, the ICO: "Just because a telephone number is publicly available or appears in a telephone directory does not mean it overrides a TPS or CTPS registration." That includes Companies House, Google Maps, LinkedIn and the trade directory you scraped.
• Keep your own do-not-call list. "Not interested, take me off your list" is an objection. Log the number, the date and what they said, and never call it again.
• Show your number, say who you are. Regulation 24: no withheld numbers, give your business name, and give an address or freephone number if asked. The hidden-number cold call is illegal as well as insulting.
• Outsourcing doesn't outsource the blame. The regulation covers anyone who makes calls or "instigates" them. If you pay a call centre to dial an unscreened list, that's your fine.
• No robot calls. Automated, recorded and AI "avatar" calls sit under a different rule and need prior consent whatever the registers say.

The fines are not theoretical. In 2024 the ICO fined WerepairUK £80,000 for 42,688 marketing calls, which is a few weeks of work for one small team. Jacksons Marketing got £130,000 for 232,776 calls to TPS-listed numbers. Elderly Aids, a firm selling call-blocking devices of all things, got £190,000. And the ceiling on PECR fines has now been lifted to UK GDPR levels under the Data (Use and Access) Act 2025, so the numbers only go one way from here.

For what it's worth, the call centre I worked in did all of this. Every list that came in went through the dialler's screening function against TPS and CTPS, and suppressions from other campaigns were carried across. It wasn't optional and it wasn't hard. If a call centre can do it on three hundred dials a day, you can do it on forty.

"Cold calling isn't dead. Lazy cold calling is, and it deserved it."

Why most people are bad at it

The title isn't a provocation, it's a diagnosis. Cold calling gets called dead by people who did it badly for a fortnight and stopped. Here's what bad looks like, in the order I saw it kill campaigns.

• No idea who they're calling. Market to everyone and you're marketing to no one. A list of "businesses" is not a list. A list of 40 firms that fit one profile, with a reason each one might care, is.
• Pitching in the first ten seconds. Nobody has ever bought because a stranger described their product quickly. The first call is for finding pain, not delivering a pitch.
• Features instead of benefits. The thing that made the dialler years work for me was translating something technical into what the owner actually gets. Nobody wants an integration. They want the orders to stop going missing.
• Sounding like a salesman. When the words are scripted, the whole job is tone. Friendly, curious, unhurried. The moment you sound like sales, the shutters come down.
• Taking it personally. The one habit that separated the people who booked from the people who didn't: they didn't let the calls affect them. It's not personal. They stayed locked in, didn't drift, and kept nicking little things that worked from the people around them.
• The TikTok opener. "Hi, what did I catch you in the middle of?" If you're cold calling in the UK, nobody cares what you caught them in the middle of. That stuff is theatre for other salespeople. Say who you are and why you rang.

Cold calling tips from 12 years on the phones

1. Build the list before you touch the phone. Define the ideal customer, write two avatars, then find every local business that fits. A Google Maps scrape into a spreadsheet will do it in an afternoon. Then screen it (see above). The right 20 beats the wrong 200.
2. Say who you are and why, in one breath. Name, business, one line of research on them, one question. Then stop talking.
3. Listen for the big three. Pain, urgency, budget. You're not selling on this call. You're finding out whether there's anything to sell, and whether they care enough to talk again.
4. Watch for the tell. Real buyers get a bit enthusiastic. They ask a question back. Everyone else is polite. Polite isn't a lead.
5. Book, don't pitch. The only outcome you want is the next conversation, in the diary, with a date.
6. Turn "just send me an email". The line that booked more appointments than anything else I said: "I can definitely send you an email, but have you ever had an email that was more valuable than a conversation? Let me get you fifteen minutes with the right person, and everything I'd have emailed will be in the confirmation anyway." Works because it agrees with them first.
7. Turn "not interested" with curiosity. "No problem. Just out of curiosity, what is it you're not interested in?" Half the time they haven't heard what you said. The other half tell you exactly what they'd need.
8. Block the time and defend it. An hour, phone only, no email open. Calling in five-minute gaps between client work is how people conclude it doesn't work.
9. Two minutes at the end of every session. What objection came up most? What line landed? Write it down. That's your call review, and it's also your next piece of content, because if three people asked it on the phone, people are typing it into Google too.
10. Every call is now the first of about twelve touches. The moment you hang up, they look you up. Your website, your LinkedIn, your reviews are the second call. Make sure they say the same thing you did.

A cold calling script that isn't a script

Scripts get you fired from the conversation the moment the other person goes off it. What worked on the floor was a skeleton with mandatory lines and free tone. Yours needs four beats and nothing else:

1. Who and why. "Morning, it's Gareth from [firm]. I'm ringing because I noticed [one specific thing about them]."
2. One question about their pain. "How are you handling [the problem you solve] at the moment?" Then shut up. The silence is the script.
3. Reflect and qualify. "So the issue is really [their words]. Is that costing you anything yet, or is it more of an annoyance?" Pain, urgency, budget, in their words, not yours.
4. Book or bin. "Worth fifteen minutes next week to see if we can fix that? Tuesday or Thursday?" If no, thank them, log it, move on. Not personal.

There's a salesman on LinkedIn who has branded himself something like the most hated salesman in Britain, and I watched him land an appointment on video without even giving his name. It was phenomenal, and it was all tone and questions. The words barely mattered.

Who should cold call, and who shouldn't

Anyone can, essentially. It suits people who are willing to talk to strangers, who understand that rejection is coming and don't take it home, and who are a bit interested in sales as a craft. It isn't for the faint-hearted, and if the thought of the fortieth "no" before lunch makes you feel sick, your hours are worth more spent on the inbound engine.

Where I'd reach for it: a small B2B firm with no marketing budget, a clearly defined customer, and a service worth enough that twenty good conversations could produce a meaningful contract. IT support for local businesses is a decent example. Lots of firms need it, it's a known pain, and a screened local list is a morning's work. Where I'd be honest about my limits: I've never seen cold calling run for an accountancy firm, and I don't know how it would go, so I won't pretend. For accountants I'd build organic content and paid search first. The test for everyone is the same. Does an hour of calling make you more gross profit than an hour spent on referrals, networking, warm follow-ups or writing the page that answers the question everyone keeps asking?

And yes, I'd do it today for a new business with no money. The agency I work at is a new business, our sales guy is on the phones as I write this, and I've chipped in. I've lost a bit of my edge from not doing it daily. It came back within an hour. What's changed since my dialler days isn't the phone, it's what happens after the call. Buyers research you before they ring back, so the call is the start of the conversation, not the whole of it.

The honest verdict

Not dead. Harder. Legal only if you screen. Brilliant for the right firm calling the right forty businesses, a waste of a founder's afternoon for everyone else. If that's not you, the answer isn't to give up on outbound, it's to build the lead generation system that makes the eventual call warm. And whichever you pick, read inbound vs outbound: which to run first before you decide.

Sources: ICO guidance on live marketing calls and the Guide to PECR (quotes verbatim); ICO monetary penalty notices for Jacksons Marketing Ltd, WerepairUK Ltd and Elderly Aids Ltd; Data (Use and Access) Act 2025 (penalty ceiling); British Chambers of Commerce and Intuit, The Turning Point for SMEs (2025, 1,500+ UK business leaders); Gong, "Insights from 300M calls" (2024, vendor); Cognism, State of Cold Calling 2024 (vendor); Belkins cold calling benchmarks 2026 (2025 data, vendor); Baylor University Keller Center, "Has Cold Calling Gone Cold?" (2012); RAIN Group, Top Performance in Sales Prospecting (2018, 488 buyers). Named as unreliable and not used: the 1942-derived follow-up statistics, "82% accept cold calls", "2% success rate", "8 attempts to reach a prospect", "6 hours a week cold calling". My own dialler figures are recollection, not records. Nothing here is legal advice.