What Cold Calling Is, and Which Rules Apply to a B2B Call

A rep opens a list, dials a number nobody gave them permission to dial, and asks a stranger for eight minutes. That is the whole of cold calling, and almost every argument about it turns out to be an argument about one of three things: what counts as cold, which rules apply, and what the call is supposed to produce.

The definition itself is short. A cold call is an unsolicited telephone call to a person who has not asked to hear from you, made to start a commercial conversation. The prospect has no prior relationship with the caller, has not enquired, has not downloaded anything, and is not expecting the call. Everything else that gets described as cold calling is warmer than that: calling a lapsed customer, calling someone who filled in a form, calling a referral, calling an account your colleague met at a conference. Those calls have a reason attached that the recipient can verify in their own memory, which changes how they open and how often they convert.

What separates a cold call from the alternatives

The useful distinction is not the temperature word. It is what the recipient is able to do with the interruption in the first four seconds.

A written message waits. Someone reads it when they choose to, and the cost of ignoring it is zero. A call does not wait. It takes the recipient out of whatever they were doing, which is both the entire advantage and the entire problem. The advantage is that you find out today. The problem is that you have borrowed attention rather than requested it, and the opening has to be worth the theft.

That difference has a practical consequence for how the two channels are planned. Written outbound is a volume instrument: the message is fixed, the list is large, and the reply rate does the sorting. Calling is a depth instrument: the list is smaller, the caller adapts in real time, and the same person can be reached again in a way that does not read as pestering, because a phone conversation is two-way correspondence rather than a broadcast. Our own written channels run one message per campaign for exactly that reason. The phone does not carry the same constraint, and pretending it does produces bad advice.

Cold callSpoken, synchronous

• Answer rate is the gate
• You learn the outcome today
• Adapts inside one conversation
• Caller time is the binding cost
• Reaching a small market repeatedly is normal

Cold emailWritten, asynchronous

• Deliverability is the gate
• Outcome arrives over days
• Message is fixed before send
• Infrastructure is the binding cost
• One message per campaign, no bumps

LinkedIn outreachWritten, semi-synchronous

• Connection acceptance is the gate
• Profile context is visible to both sides
• Message limits cap volume
• Account safety is the binding cost
• A second message lands under the first, so we do not send one

Is cold calling marketing or sales?

The question sounds like semantics and decides who owns the budget, so it is worth answering. Cold calling is usually run as a marketing channel and staffed as a sales function. It is a marketing channel in the sense that matters commercially: it is a paid route to a stranger's attention, it competes for budget against email, paid media and events, and it is judged on cost per meeting like any of them. It is a sales function in the sense that the work is a live conversation held by a person who has to think, which no marketing team is set up to manage.

Where it goes wrong is when the two owners disagree about the metric. Marketing counts conversations produced. Sales counts meetings held that turned into pipeline. Those two numbers can move in opposite directions for a quarter before anyone notices, so agree which one the programme reports on before it starts.

Cold calling and telemarketing are not the same job

The two words get used interchangeably and describe different work. Telemarketing in its classic form tries to complete a transaction on the call: a script, a price, a card number, one attempt. B2B cold calling almost never tries to close anything. It tries to establish that a problem exists, that this person owns it, and that a longer conversation is worth putting in a diary. The pitch is a meeting, and the product barely gets described.

That difference shows up in how the two are staffed. A telemarketing floor is measured on calls handled and conversion on the call. A B2B calling team is measured on meetings held and, further down, on whether those meetings turned into pipeline that the account executives agreed was real. Anyone quoting call-centre productivity figures at a B2B calling programme is measuring the wrong job.

Which rules actually apply, and the part most articles get backwards

Cold calling in the United States is governed mainly by the Federal Trade Commission's Telemarketing Sales Rule. The FTC's own compliance guide is unusually direct about business calling, and it says something most cold calling advice never mentions.

"Most phone calls between a telemarketer and a business are exempt from the TSR," the guide states, in the section covering business-to-business solicitation calls. The exemption is broad, and it has a small number of specific holes rather than vague ones. Calls placed to induce the retail sale of nondurable office or cleaning supplies are not exempt, and the guide is precise about what nondurable means: "paper, pencils, solvents, copying machine toner, and ink," anything that depletes when used. Software, copiers, computers, mops and buckets are treated as durable and stay outside the carve-out.

The National Do Not Call Registry follows the same line. The FTC guide states that the prohibition on calling numbers on the Registry "does not apply to business-to-business calls." The Registry is a consumer protection, and a company's main line is not a consumer.

The regulation itself is tighter than the guide and worth reading once. 16 CFR 310.6(b)(7) exempts "telephone calls between a telemarketer and any business to induce the purchase of goods or services", and names exactly two things the exemption does not reach: the nondurable-supplies case above, and "the requirements of § 310.3(a)(2) and (4)". Those two are the anti-deception provisions, prohibiting "misrepresenting, directly or by implication" a material aspect of what is sold and "making a false or misleading statement to induce any person to pay for goods or services". So the residue of the rule that still binds a business-to-business caller is the obligation to tell the truth, which is a better summary of the compliance position than any list of mechanics.

There is one trap inside all of that, and it is the one worth memorising. The guide notes that "telemarketing calls that solicit consumers at their work," meaning calls to a business line that ask an individual employee to buy something for their own personal use, "are not business-to-business solicitations and are not exempt." Calling a facilities manager about a facilities contract is business calling. Calling the same person at the same desk about a personal insurance policy is not.

Is this call inside the FTC's business-to-business exemption?

• ✓Yes: The call is to a business, about something the business would buy
• ✓Yes: The purchase is for the company, not for the individual personally
• ✓Yes: You are not selling nondurable office or cleaning supplies at retail
• ✓Yes: You are not soliciting a charitable contribution from the employee
• ~Depends: State law and non-US jurisdictions are checked separately, because the exemption above is a federal one

For the calls that are covered, the FTC guide sets out mechanics worth knowing even when they do not bind you, because they describe what regulators consider abusive. A call counts as abandoned "if a person answers it and the telemarketer does not connect the call to a sales representative within two seconds of the person's completed greeting." The safe harbour allows abandonment of "no more than three percent of all calls answered by a live person," measured per campaign or per 30-day period, requires letting the phone ring "for 15 seconds or four rings," and requires a recorded message naming the seller when no representative is available. Prerecorded sales pitches fail this test by construction, because a recording is not a sales representative. Calls before 8am or after 9pm are named as abusive, and records have to be kept for two years.

Outside the United States the defaults invert. Several jurisdictions treat unsolicited business calling as opt-out by register or opt-in by consent rather than exempt, so a list that is clean in Ohio is not automatically clean in Manchester or Munich. Check the destination country before the first dial, not after the first complaint.

What a cold call is supposed to produce

The output of a cold call is a booked conversation with a named person, and nothing else. It is not a sale, not a demo watched on the spot, and not a promise to think about it. Teams that judge calling on anything other than meetings held end up optimising for pleasant conversations, which are free to obtain and worth nothing.

Three variables set the result, in this order.

The list decides most of it. A correct mobile number for a person who owns the problem beats any script, and a wrong number is unrecoverable at any level of skill. That is why the phone-number quality problem swallows more calling programmes than technique ever does.

The opening decides the rest of the conversation. Whatever is said in the first ten seconds has to give the recipient a specific reason why this call is happening to them today, and that reason is a property of the research behind the list. The first ten seconds and the list behind them is where most of the improvement available to a calling team actually sits.

The market size decides whether the channel is the right one at all. A team with a few thousand qualifying accounts can afford to reach each of them by hand, and a team with a hundred thousand cannot. That is the reasoning behind running calling as the primary motion when the addressable market is small, and it inverts as the market grows.

Where calling sits next to everything else

Channels differ mostly in how quickly they answer and how much they cost per attempt. Sorting them that way, rather than by which one somebody likes, is the practical version of a channel strategy, and it is worth doing on paper before hiring anyone. Our own comparison of how fast each lead generation channel answers lays that out.

RevenueFlow does not sell cold calling. We build and run cold email and LinkedIn outbound, and we say so plainly on pages like this one because the alternative is writing calling advice with a thumb on the scale. If the phone is the right channel for your market, the honest options are an in-house caller, a specialist firm, or a hybrid where calling handles the top accounts and written outbound handles the rest. The SDR role as it exists after automation is a useful reference point for what you are actually hiring when you hire for this.

What we can say from the buying side is that calling and written outbound fail differently. Calling fails loudly and immediately, in the form of a caller with nothing to do and no meetings on the board. Written outbound fails quietly, in the form of messages that were never delivered. Loud failure is easier to manage, which is a genuine argument for the phone that rarely gets made.

If written outbound is the part you would rather not run yourself, that is the part we do, and our outbound programmes are built around a single message per campaign with nothing scheduled behind it.

Regulatory citations verified against the FTC's own compliance guide as of August 2026. Rules change and vary by state and country. Verify current obligations with the FTC and with counsel before running a calling programme.