Almost every cold email agency site ends the same way: "pricing depends on scope, book a call." You sit through thirty minutes of discovery before anyone says a number, and the number arrives in a PDF two days later.
I publish mine. But the more useful thing isn't my number — it's understanding what you're actually buying, because the line items are where the money goes missing.
The three shapes this work comes in
Outbound gets sold in three ways, and they are not interchangeable.
- An audit. Someone looks at what you already have — domains, lists, copy, sequences — and tells you why it isn't landing. One-off, usually a week. Mine is $499.
- A build. Someone sets the whole engine up: infrastructure, warmup, data, enrichment, copy, sequences. One-off, four to six weeks. Mine is $3,499.
- A retainer. Someone builds it and then runs it — writing new angles, rotating domains, reading replies, and fixing what breaks. Monthly. Mine is $1,999/mo.
Most agencies only sell the third one, because it's the only one with recurring revenue. That's fine, but it means you get pushed toward a retainer even when an audit would have told you what was wrong for a fraction of the cost.
What actually costs money
Here's the part nobody itemises. Sending cold email has hard costs that exist whether or not you hire anyone:
- Domains. You need several, and you don't send from your main one. At a few dollars each per year this is the cheapest part of the stack and the one people most often get wrong.
- Inboxes. This is the real recurring cost. Every sending inbox is a seat, and you need more than you think, because each one has a daily ceiling you don't want to approach.
- Warmup. Weeks of it before volume. It is time, not money, but it's the reason a build takes four to six weeks and not four days.
- Data. Verified contacts, and enrichment on top. This scales directly with list size and it is the line item that quietly doubles.
- Sending tool. A sequencer, priced per seat or per contact.
When an agency quotes you a flat monthly fee, ask which of those five they're covering and which you'll be invoiced for separately. The gap between "all in" and "our fee, plus your tooling" is often larger than the fee itself.
The question that tells you the most
Ask what happens in month two.
A build is finite work — it ends. A retainer should be paying for something ongoing, and if the honest answer is "we keep the campaign running," ask what running means. Rotating domains, writing new angles per segment, and reading what came back are real work. Watching a dashboard is not.
If nobody can describe month two in concrete terms, you're paying a subscription for a thing that was finished in week six.
What I'd actually recommend
If you have never run outbound: start with a build, not a retainer. You need the engine to exist before paying anyone to run it, and you'll learn enough in those six weeks to judge whether the retainer is worth it.
If you have run outbound and it stopped working: start with an audit. It is almost always one of four things — the domains are burnt, the list is too broad, the copy is templated, or the sending volume outran the warmup. An audit finds which, and then you know whether you need a rebuild or a small fix.
If outbound is already working and you just don't have time to run it: the retainer is the right shape, and it's the one case where it genuinely is.
On the cheap end
There are people who will build you an outbound engine for a few hundred dollars. Some of them are good. The failure mode isn't that they take your money and vanish — it's that they set up something that sends, which looks identical to something that works for about three weeks, right up until the domains go quiet and nobody can tell you why.
The difference is entirely in whether the person has watched enough campaigns go wrong to recognise the shape of it early. That is the thing you are paying for, and it's genuinely hard to assess from a pricing page — including mine.