Search for the cost per appointment of cold email and you will find a number. Then you will find another one that is five times higher, and a third that sits somewhere in between, and all three will be presented as a benchmark. Search for the same figure for cold SMS and you will mostly find articles about return on marketing texts sent to people who opted in — a completely different activity from texting a business owner who has never heard of you.
The reason the published numbers disagree is not that some of them are lying. It is that cost per appointment is not a property of a channel. It is a quotient, and three of the four numbers that produce it belong to your market rather than to the channel you picked. Two competent operators can run identical campaigns, spend identical money, and end up an order of magnitude apart because one was selling to a category with fifty thousand reachable businesses and the other was selling to a category with nine hundred.
So this is not another benchmark table. It is the arithmetic, the part of it that people leave out, and an honest account of when cold email genuinely beats cold SMS on cost — because it sometimes does, decisively, and you should know which side of that line you are on before you spend anything.
Why every published number is someone else's
Cost per appointment is total spend divided by held appointments. The trouble is what sits inside those two terms.
Expand the denominator and you get a chain: contacts on your list, times the share you can actually reach, times the share of reachable contacts who reply, times the share of replies that become a booked meeting, times the share of bookings actually attended. Five fractions, each specific to your category, your geography, your offer and your speed of follow-up.
Only one link in that chain is really about the channel — reachability. The rest are about whether your offer is any good and whether the people on your list were the right people. This is why importing a benchmark from an article is close to meaningless: it tells you what happened when a company you have never met sent messages to an audience you do not sell to.
There is a second problem, which is who publishes the figures. Almost every cost-per-meeting benchmark in the cold email space comes from a company selling sending software or sending services. That does not make the numbers false, but it does determine which costs got counted — and the line item that consistently vanishes is labor, which for cold email is often the largest cost of all.
The two channels have opposite cost shapes
Here is the thing the comparison articles miss entirely. Cold email and cold SMS do not merely have different prices. They have different shapes, and shape is what determines who wins.
Cold email is fixed-cost heavy. Before the first message goes out you need secondary domains, a set of inboxes, a warmup period measured in weeks, and a sending tool. After that, each additional send costs approximately nothing. The consequence is that your cost per appointment falls continuously as volume rises — the fixed base gets spread thinner and thinner. Cold email becomes genuinely cheap at scale, and it is expensive, sometimes absurdly so, when you cannot reach scale.
Cold SMS is marginal-cost heavy. There is almost nothing to build up front, but every message costs a real fraction of a cent, every AI-handled reply costs something, and — the part people forget — every record has to be validated as a working mobile number before you can use it at all. None of those costs get cheaper with volume. The consequence is a cost per appointment that is roughly flat: about the same on your five hundredth message as on your fifty thousandth.
| Cold email | Cold SMS | |
|---|---|---|
| Cost before message one | High — domains, inboxes, weeks of warmup, tooling | Low — registration and a sending number |
| Cost per additional contact | Effectively zero | Real and constant — per message, per reply, per validated record |
| Shape of the curve | Falls steeply with volume | Flat at any volume |
| Cheapest when | Your market is deep enough to amortize the setup | Your market is shallow, or you need results before a warmup ends |
| Where the cost hides | Ongoing deliverability labor | Data validation on records you end up discarding |
| What breaks it | Reputation damage; recovery takes weeks | Bad targeting; every wasted message is billed |
Read that table as a decision rule rather than a scorecard. If your filtered market runs to tens of thousands of businesses with valid business email addresses, email's fixed cost gets amortized into irrelevance and it will very likely win on cost per appointment. If your filtered market is a few thousand owner-operated local businesses, that fixed cost never amortizes, and the cost per appointment you compute from a benchmark article — which assumed the deep market — is a fiction.
Get the math for your niche, not a benchmark
On a free 15-minute demo we pull the actual businesses in your category and geography, show you the real messages, and work the profit math with your numbers. At least 3 booked appointments in your first 30 days or you don't pay.
Book your demo →The input everyone leaves out: reachability
Both channels are usually costed on messages sent. Both should be costed on people reached, and the gap between those two numbers is where most of the disagreement in published benchmarks actually lives.
On the email side, a scraped list of local businesses contains a large share of generic addresses that route to a receptionist, a shared mailbox, or nobody at all, plus a slice that no longer resolves. On top of that sits deliverability: filtering, promotions tabs, and reputation losses. Nobody sends you a report saying which messages were never seen.
On the SMS side the loss is blunter and easier to measure. The number on a business listing is frequently a landline or a virtual number, not a mobile, and a text sent to it simply does not arrive. Validating line type before you send turns that into a known quantity instead of a silent tax — the same reason purchased lead lists underperform fresh, validated data so consistently. You pay to check, and you pay for records you then throw away, but you pay once and you know your real denominator.
This is the honest version of the comparison: SMS makes you pay visibly for reachability up front, email makes you pay for it invisibly in a lower reply rate that looks like a copywriting problem. If your two channels appear to have the same reply rate, one of them is being measured on a denominator that includes people who never received anything.
Recompute one number and most confusion disappears. Take your last campaign and divide replies by contacts you can prove were reached — validated mobiles for SMS, deliverable and monitored inboxes for email — rather than by messages sent. Most people find one channel improves sharply and the other gets worse. That corrected reply rate is the only one worth putting in a cost calculation.
The labor line that never appears in a benchmark
Cold email has an ongoing maintenance requirement with no equivalent on the SMS side. Domains age and need rotating, inboxes get flagged, warmup schedules have to be maintained, bounce rates watched, copy adjusted when filtering shifts. Somebody does that work every week, and if that somebody is you, the hours are invisible in your spreadsheet and very visible in your calendar.
Price it honestly: hours per month, times what an hour of your time is worth, added to the numerator. For a solo operator or small agency this single adjustment routinely doubles the true cost per appointment — and it does not shrink as volume rises, because the maintenance is per-infrastructure, not per-send.
SMS has its own labor, but it sits in a different place: targeting and reply handling. Getting the list wrong on SMS is expensive in a way it never is on email, because every message to the wrong business is billed. One of our own clients spent a week messaging med spas and towing companies he had no intention of ever selling to, and described it exactly right — it was burning through his data. On a channel with a real marginal cost, sloppy targeting is not a quality problem, it is a line item. The filters that prevent it are covered in qualifying filters that keep junk off your calendar.
When cold email genuinely wins
The clearest example we have came from a prospect who talked us out of our own pitch, and he was right to.
He was a sell-side business broker running a stacked cold email operation — an enrichment tool, a database, and a sequencer — and getting roughly forty booked calls a month out of it at a marginal cost he considered trivial, because his infrastructure was already built and paid for. We quoted him a program that would produce a fraction of that volume for more money. He declined in one sentence: the math did not work, and he might as well add more fuel to the cold email fire. He wanted to try SMS and said so twice. He still should not have bought.
That is what a genuine email win looks like, and it has three ingredients. His market was deep enough that he had not exhausted it. His fixed cost was already sunk, so his forward-looking cost per appointment was almost pure marginal. And his buyers — business owners considering selling a company — are people who read email. Where all three hold, adding a second channel to buy incremental appointments is a worse use of money than pushing more volume through the channel already working.
The corollary matters just as much. Each of those three ingredients has an expiry date. Markets get exhausted, and when they do the cost per appointment on a fixed-cost-heavy channel climbs fast, because the same setup cost is now spread over a shrinking pool of people who have not already heard from you. That failure is slow and looks like fatigue rather than arithmetic; the runway math behind it is in how many cold messages you should send per month.
When SMS wins on cost
SMS wins where email's assumptions break, which is more often than the benchmark articles suggest because those articles are written about desk-bound corporate buyers.
It wins when your buyer is an owner-operator. A roofer, a salon owner, a contractor and a med spa director all have one thing in common: the phone is in their pocket and the email address on their listing is checked occasionally, or by somebody else. Reachability collapses on one channel and holds on the other, and reachability is the only term in the chain the channel genuinely controls.
It wins when the market is shallow. If your filtered category and geography yields a few thousand businesses, you will never send the volume that makes a fixed-cost-heavy channel cheap, and a flat cost curve is strictly better than a falling one you cannot ride down.
It wins when you need speed — warmup takes weeks before meaningful volume is safe, while a validated SMS program produces replies in days. And it wins on reply-cycle speed, which shortens the whole funnel: text conversations resolve in minutes rather than the multi-day loops email creates, which means fewer touches per booking. The channel-by-channel breakdown of that behavior lives in our cold email vs cold calling vs SMS comparison, and the message craft in the cold SMS playbook. This article is deliberately only about the money.
How to actually run the comparison
Do not compare your campaign to an article. Compare your campaign to your other campaign, under conditions that make the result mean something.
Take one list and split it in half at random. The two halves must never touch — a business that gets the same pitch by text and by email in the same month concludes you are spam, and you lose the prospect and the comparison at once. Run the same offer on both, in the same weeks, and let both run long enough to produce a double-digit number of replies, because a comparison built on three bookings is noise.
Then count carefully. Held appointments, not booked ones, because the two channels do not have the same show-rate behavior and booked-count comparisons flatter whichever one produces the softer yes. Count total spend including the setup you amortized and the hours you worked. Divide. Now you have two numbers that are actually about you.
Finally, compare both against your ceiling rather than against each other — what a closed customer is worth, multiplied by the share of meetings you close, is the most you can pay per appointment and still make money. A channel that beats the other one but sits above that ceiling has not won anything. The ceiling calculation and the wider market rates are covered in how much appointment setting costs, and if you are a marketing agency running this comparison for your own pipeline, the acquisition context sits on our marketing agency page. Current pricing for our own program is on the pricing section of the site rather than in this article, because it changes and this arithmetic does not.
See both channels costed with your numbers
Fifteen minutes, no charge: the real businesses we'd reach in your category, the messages we'd send, and the cost-per-appointment math worked in front of you. 3 booked appointments in your first 30 days or you don't pay.
Book your demo →Frequently asked questions
What is a good cost per appointment for cold outreach?
There is no transferable number, which is why the published ones disagree with each other so violently. Cost per appointment is derived from four inputs — your total spend, how many of your contacts are actually reachable, what share of reachable contacts reply, and what share of replies become held meetings — and three of those four are properties of your market rather than of the channel. The number that matters is not a benchmark but a ceiling: what one closed customer is worth to you, multiplied by the share of meetings you close, gives the most you can pay per appointment and still profit. Anything under that ceiling is good, and anything over it is bad, regardless of what someone else publishes.
Is cold SMS cheaper than cold email per booked appointment?
It depends almost entirely on how deep your market is. Cold email carries a large fixed cost in domains, inboxes, warmup and ongoing deliverability work, and a marginal cost per send that rounds to nothing, so its cost per appointment falls as volume rises and only becomes cheap if your list is deep enough to spread that fixed cost across tens of thousands of sends. Cold SMS is the mirror image: almost no fixed cost, but a real per-message cost and a real per-record validation cost, so its cost per appointment is roughly the same on message five hundred as on message fifty thousand. Deep markets favor email on cost. Shallow, owner-operated, local markets usually favor SMS, because the fixed cost never gets amortized.
Why do published cold email cost-per-meeting benchmarks vary so much?
Three reasons. They are computed on different markets, and market depth and reachability move the result more than anything the sender controls. Most are published by companies selling sending tools or sending services, so the flattering inputs get included and the unflattering ones get left out. And nearly all of them count software and data spend while excluding the labor of maintaining deliverability, which for cold email is frequently the largest single line in the true cost. Two honest operators can publish numbers an order of magnitude apart and both be telling the truth about their own campaign.
Should I run cold SMS and cold email at the same time?
Eventually yes, because they reach different people and reinforce each other, but not while you are trying to compare them. A fair comparison needs one list split into two halves that never touch each other, the same offer on both, the same window, and a decision made on held appointments rather than booked ones. Running both channels at the same contacts at the same time tells you what the combination does, which is useful later, but it cannot tell you which channel earned the meeting. Decide the comparison first, then blend.