How to Combine LinkedIn and Email in One Outreach Sequence
Running LinkedIn and email side by side is not multichannel. Multichannel means the email step knows what happened on LinkedIn, and that is the part most stacks cannot do.
Most teams running "multichannel" are running two channels in parallel with no connection between them. The LinkedIn tool sends invitations on its schedule and the email tool sends on its own, and neither knows what the other did. That is two sequences aimed at one person, and the prospect experiences it exactly that way.
A real LinkedIn and email outreach sequence has branch logic: what email sends depends on what happened on LinkedIn. This is how to design one.
Which Channel Opens
LinkedIn opens, almost always. A connection request is a smaller ask than an email, it carries a face and a headline the prospect can evaluate, and acceptance tells you something useful before you have spent an email on them.
The exception is when you have a strong trigger and a verified address and the prospect is a senior title that historically ignores invitations. Then lead with email, because you are spending your one good piece of timing on the channel they actually read.
What you should not do is fire both on day one. The prospect gets a connection request and a cold email from the same stranger within an hour, which reads as pursuit rather than outreach.
The Sequence, With Branches
Fourteen days, six touches, two decision points.
- Day 0, LinkedIn profile view. Silent, free, occasionally sufficient on its own.
- Day 1, LinkedIn connection request.
- Day 3, BRANCH. Accepted: LinkedIn message, no pitch, one question. Not accepted: nothing yet, keep waiting.
- Day 5, BRANCH. Still not accepted: close the LinkedIn path and start email. Accepted but silent: leave it alone. They will see the message.
- Day 7, Email one. The specific thing: a trigger, a number, a reason it is them.
- Day 11, Email two, replying in the same thread. One new piece of information, not a nudge.
- Day 14, Close the loop, on whichever channel is open. Explicit permission to not reply.
The day-five branch is the one that matters. If the invitation has not been accepted, the LinkedIn path is finished. You cannot send a second invitation, and withdrawing then re-inviting is among the behaviours most likely to get an account restricted.
What Breaks When the Tools Are Separate
Three specific failures, all of which we have watched happen:
The email fires while the invitation is pending. The prospect gets a cold email from someone whose connection request is sitting unanswered in their notifications. It reads as pressure, and it is the most common version of this problem.
The prospect replies on LinkedIn and the email sequence keeps running. Someone who said "sure, send me details" receives a cold follow-up four days later asking if they saw the last message. This one costs real deals.
Nobody can measure it. The LinkedIn tool reports acceptance, the email tool reports replies, and no system can tell you whether the sequence works, because the sequence does not exist anywhere as a single object.
Inside grobot both channels are one sequence in one workspace. The email branch waits on the LinkedIn outcome rather than on a clock, a reply on either channel stops everything, and the replies themselves land in one inbox where Ezra can draft the routine ones.
Message Discipline Across Two Channels
Do not send the same message twice in different formats. If your LinkedIn message and your first email make the same point, the second one confirms you are running a sequence.
Give each channel a different job. LinkedIn is for the relationship: an observation, a question, no ask. Email is for the specific thing worth their attention: the trigger, the number, the mechanism.
And reference the other channel once, lightly, in email one: "I sent a connection request last week, probably got buried." It acknowledges reality instead of pretending each channel is a separate person.
What the Numbers Look Like
Against a well-targeted list of 300, plan for roughly 100 invitation acceptances, perhaps 25 LinkedIn conversations, and from the 200 who did not accept, an email sequence returning 2 to 4% positive replies, another 4 to 8 conversations.
Call it 30 conversations from 300 contacts over three weeks, per seat. That is what one properly wired multichannel sequence produces, and it is meaningfully more than either channel alone because the email arm works the people LinkedIn could not reach rather than the same people twice.
Frequently asked questions
Should LinkedIn or email come first in a sequence?
LinkedIn first in most cases; a connection request is a smaller ask, and acceptance tells you something before you spend an email. Lead with email only when you have a strong trigger, a verified address, and a senior title that historically ignores invitations.
Can I send a LinkedIn request and a cold email on the same day?
You can, and you should not. A connection request and a cold email from the same stranger within an hour reads as pursuit. Let LinkedIn run for about five days before email takes over.
What happens if the connection request is never accepted?
The LinkedIn path is finished at about day five. You cannot send a second invitation, and withdrawing then re-inviting is one of the fastest ways to get an account restricted. Branch to email instead.
Why do multichannel sequences fail in separate tools?
Because the email step fires on a timer rather than on the LinkedIn outcome. The three standard failures are emailing while an invitation is still pending, continuing to send after someone replied on LinkedIn, and having no single place where the sequence can be measured.
The handoff is the whole problem.
grobot runs LinkedIn and email as one sequence, so the email branch waits on the connection outcome instead of a timer. One inbox for the replies, one pipeline for the results.
See the cadence builder →Running outreach for a book of clients? See how benefits agencies run a whole book on one record.
