What should your first outbound sequence look like?

The short answer

Fifty to a hundred and fifty contacts, one segment, four steps over two weeks, sent from a separate warmed domain, and you read every reply yourself. The goal of the first sequence is not pipeline. It is finding out whether your message is wrong before you have spent your domain reputation discovering it at scale.

Start much smaller than feels useful

The instinct at launch is volume, because the maths looks obvious: if one percent reply, then ten times more contacts means ten times more replies. That maths holds only if the message works, and at launch you do not yet know whether it does.

A small first send costs you a week if it fails. A large one costs you the domain, and domain recovery takes months. So the first sequence is deliberately small enough that a person can read every single reply, including the annoyed ones, which are the most informative.

Pick one segment, not a market

The first sequence goes to one segment, defined tightly enough that a single message is true for everyone in it. Not "B2B SaaS companies". Something like "heads of partnerships at trading platforms with between 50 and 300 staff who already run an affiliate programme".

Tight targeting is what lets you write a first line that could not have been sent to anyone else, and that first line is most of the reply rate.

The four steps

Step	Day	What it does
1. The observation	0	Something specific and true about their business that led you to write. One sentence on what you do, one clear ask. Under 90 words.
2. The proof	3	A short, concrete example of the same problem solved elsewhere, with a number if you have one you are allowed to share. Reply in the same thread.
3. The angle change	7	A different reason they might care, in case the first was wrong. This step exists because your first guess about their priority is often wrong.
4. The close	14	Short, no pressure, and an explicit offer to stop. "If this is not a priority, say so and I will stop." People answer this one.

Four steps is not a compromise. Steps five through nine mostly convert people who were going to reply anyway, while adding unsubscribes and complaints from people who were not.

What each message must not do

• No merge field theatre. A first line that says "I saw {Company} is doing great things in {Industry}" tells the reader they are in a list of ten thousand.
• No asking for thirty minutes in step one. The first ask should be cheaper than a meeting: a yes or no to whether the problem is real for them.
• No attachments, no images, no tracking pixels on the first send. All three hurt deliverability, and open rates have been unreliable since Apple Mail Privacy Protection anyway.
• No sending to a list you have not suppressed. Existing customers receiving a cold pitch is the fastest way to lose the trust of your own sales team.

Stopping is more important than sending

The mechanics that matter most are the ones that stop the sequence. Any human reply stops it immediately and routes the record to an owner with the thread attached. An out-of-office is not a reply and should pause rather than stop. A hard bounce suppresses that address everywhere, not just in the tool that sent it.

Get this wrong and someone replies "not now, try me in Q3" and receives step three two days later. That is a lost deal and a story they tell other people.

What to measure on the first sequence

Not open rate. It is unreliable and it does not tell you anything you can act on. Measure reply rate including negative replies, meetings booked, and the reasons in the negative replies. The reasons are the actual product of a first sequence, because they tell you whether your targeting or your message is the thing that is wrong.

Then, and only then, scale

When one segment produces a reply rate you are happy with and the negative replies stop telling you that you targeted the wrong people, widen the list and add the second segment with its own opener. Not before.

Common questions

How many contacts should the first outbound sequence target?

Between 50 and 150. Small enough that you can read every reply yourself, and small enough that a mistake in targeting or copy costs you a week rather than your domain reputation.

How many steps should an outbound sequence have?

Four to five over about two weeks. More steps do not produce more replies; they produce more unsubscribes and more spam complaints once the first three have failed.

Should we send from the company domain or a separate one?

A separate sending domain, warmed over two to three weeks before the first send. Never send cold volume from the domain your invoices and customer email go out on.

How long before an outbound sequence produces meetings?

For a well-targeted first sequence at launch, expect the first replies within days and the first meetings within two to three weeks. Anything faster usually means the list was warm rather than cold.

Where this connects