An outbound sequence is a planned series of contacts with one person, across one or more channels, that runs until they answer or until you decide to stop. It exists because almost nobody replies to the first message, and because following up by memory does not survive a busy week. The sequence is the memory. This guide covers how to structure one so that each step earns its place, rather than repeating the first email with “just bumping this up.”
Every step needs something new to say
The most useful rule for designing a sequence is that each step should give the person a new reason to reply. The first message states why you are writing to them specifically and what you think their problem is. Later steps can bring a different angle on the problem, a concrete example of how others in their position deal with it, a short resource that is useful on its own, a question that is easy to answer, or a different channel that makes you more real. A step that only reminds them of the previous one is noise, and it trains the reader to ignore your name.
This rule also answers the question of length better than any benchmark: a sequence can be as long as the number of genuinely different things you have to say to that person. We look at that in more detail in How many touches should an outbound sequence have?
What each channel is good for
Email carries the argument. It is where you can explain the problem in a few sentences and where a reply costs the reader nothing. It is also the channel that can run on its own once written.
LinkedIn makes you a person rather than an address. A connection request, a comment on something they posted or a short message after they have seen your name in their inbox all make the next email more likely to be read. Keep it human and manual: automated LinkedIn activity breaks the platform’s terms and reads like what it is.
Phone calls are the fastest way to find out whether the problem is real. A call that reaches someone gives you more information in two minutes than a whole email thread. A voicemail is short, gives one reason to call back and points to the email you sent.
SMS and WhatsApp are personal channels. They work when the relationship allows it, often after a first conversation or to confirm a meeting, and they feel intrusive when used as a cold opener in many markets. Use them later in a sequence, sparingly, and only where you have a legitimate reason to hold the number.
A structure to start from
A reasonable first version for a B2B motion alternates channels so that each one supports the next:
- Email that states why this person, what problem and one question.
- LinkedIn: view the profile or connect with no pitch.
- Call, referring to the email; leave a short voicemail if nobody answers.
- Email with a different angle: an example, a resource, or the cost of the problem.
- LinkedIn message, if connected, that is short and conversational.
- Call at a different time of day.
- Closing email that says you will stop writing, and leaves the door open.
This is a starting structure, not a recommendation of exactly seven steps. Adjust it to the motion: a founder-to-founder motion may lean on LinkedIn, a motion selling to field teams may lean on the phone.
Spacing the steps
Leave enough time between steps that each one arrives as a separate event rather than a pile-up, and keep the whole sequence short enough that the person still remembers the first message by the end. In practice that means gaps of a few working days early on, widening towards the end. Avoid stacking two channels on the same day unless the second clearly refers to the first (“I’ve just sent you an email about…”).
Send at the times the recipient works, in their timezone. A sequence that fires at your 9am into someone else’s 3am makes the message look automated, because it is.
Automatic steps and manual steps
Some steps can go out on their own: emails, and in some cases SMS. Others should not: LinkedIn activity and calls need a person. The practical problem is making sure the manual steps actually happen on the day they are due. A sequence tool that sends the emails but leaves the calls to memory produces sequences that are, in practice, email-only. Put manual steps in a daily queue, handle the queue first thing, and treat an unhandled queue as a missed send.
When to stop
A sequence ends when one of these happens:
- They reply, positive or negative. Take them out of the sequence immediately; nothing looks worse than an automated follow-up to someone who just said yes, or no.
- They ask you to stop. Remove them from every sequence, not only this one, and record it.
- You run out of new things to say. Send a closing message and stop.
- The information changes. They left the company, the company does not fit the ICP after all. Stop and fix the list.
After a sequence ends without an answer, the person is not lost; they are not interested now. Depending on the motion, it can be reasonable to come back months later with something new. Coming back with the same sequence is not.
Check the rules where your prospects are
Cold outreach is regulated differently by country and by channel. In the EU, GDPR applies to the personal data you use, and national rules govern unsolicited electronic messages; B2B email is treated differently from SMS and WhatsApp, and from messages to consumers. Check what applies to your market, always offer a way out, and keep a record of opt-outs.