A Multichannel Sales Sequence, Step by Step, With the Full Cadence

A multichannel sales sequence is defined by its branches and its exits, not by its touch count. "Eight touches over 21 days" describes nothing, the same eight touches produce a booked meeting or a spam complaint depending on what happens between them.

So here is one written properly: every touch, what it is for, what it branches on, and when it stops. Adapt the words; the structure is the part worth copying.

Before Touch One

Three things have to be true or the sequence is wasted effort.

The list is built from a trigger, not a filter. Something happened, a job change, a funding round, a first Form 5500 filing, departmental headcount growth. A demographic list converts at a fraction of a trigger list regardless of how good the copy is.

The email addresses are verified immediately before send, not at import. B2B data decays at roughly 2 to 3% a month, and bounce rate is the fastest way to damage a sending domain.

The sending infrastructure is warmed. New mailboxes need three to four weeks before they carry campaign volume, and that cannot be compressed.

The Sequence

Days 0–5: LinkedIn opens

• Day 0, Profile view. Costs nothing, counts against nothing, occasionally produces a connection on its own.
• Day 1, Connection request. No note for volume work; a short specific note on named accounts.
• Day 3, BRANCH on acceptance. Accepted: a message with one observation and one question, no pitch. Not accepted: wait.
• Day 5, BRANCH. Not accepted: close the LinkedIn path permanently. Do not withdraw and re-invite. It is one of the behaviours most likely to get an account restricted.

Days 7–14: email carries it

• Day 7, Email one. The trigger and the consequence, in under 90 words. No link, no attachment, no calendar. One question.
• Day 11, Email two, replying in the same thread so the history is visible. One new piece of information, a number, a benchmark, a mechanism. Not "just following up."
• Day 14, Email three. The artifact: something genuinely useful they can take without talking to you. This is the touch that most often converts the people who were interested but not urgent.

Days 18–21: close the loop

• Day 18, If a LinkedIn connection was accepted but silent, one message there. Different channel, different framing, one line.
• Day 21, Break-up, on whichever channel is open. Explicit permission not to reply, and a named future trigger: "if renewals get painful in the spring, I am easy to find."

Eight touches, 21 days, two channels. The break-up reliably produces replies and is the touch most people cut.

Exit Conditions

This is the part missing from most published cadences, and it is what separates a sequence from a nuisance.

• Any reply on any channel stops everything immediately. Not "pauses the next email", stops the whole sequence, both channels.
• A negative reply exits and suppresses the contact and, usually, the account.
• An out-of-office pauses and resumes after the return date rather than burning touches into an empty inbox.
• A hard bounce exits, removes the contact, and flags the list for re-verification.
• Someone else at the same account replying exits this contact. Two reps sequencing two people at one company is how you lose the account.

The first of these is the one that breaks when LinkedIn and email live in separate tools: the prospect replies on LinkedIn and the email tool, which cannot see it, sends a cold follow-up four days later. That costs real deals, and it is common.

What to Expect From 300 Contacts

On a well-targeted trigger-based list: roughly 100 connection acceptances, about 25 LinkedIn conversations, and from the 200 who did not accept, 2 to 4% positive email replies, another 4 to 8. Call it 30 conversations, perhaps 8 to 12 meetings.

If your numbers are much better than that, your list is warmer than you think. If they are much worse, the problem is almost always the list rather than the copy, and rewriting the emails is the most common wrong response.

Instrument Four Things

• Positive reply rate by touch, so you learn which touch is actually carrying the sequence. It is usually touch one or the break-up, rarely the middle.
• Time from reply to first response. Anything over a day is losing deals that the sequence already won.
• Bounce rate per sending domain, so a problem localizes before it spreads.
• Meetings per 100 contacts, which is the only number that lets you compare two sequences honestly.