How to space a LinkedIn message sequence: what each touch should do

Why accepted invites go quiet

The accept and the reply are two different decisions, and most sequences treat them as one. Someone accepted because you looked relevant and accepting cost nothing. Replying costs attention, so it needs a separate reason, and "I sent a follow-up" is not one.

The common failure is a first message that arrives within minutes of the accept, opens with the pitch that the invite note promised was not coming, and asks for a call. It converts a person who just did you a small favour into a person who now regrets it. The second common failure is the opposite: three weeks of nothing, then a message that assumes a context the reader has entirely lost.

The fix is structural rather than stylistic. Decide what each message is for before writing any of them, and let the spacing follow from the jobs rather than from a template's day numbers.

Give each touch exactly one job

Five touches, five jobs. A message trying to do two jobs does neither, because the reader responds to whichever one is most demanding and that is always the ask.

The reframe is the touch most people get wrong. It is not a chase. It is an admission that your hypothesis about their problem may have been off, and a second hypothesis offered plainly: "I assumed this was about pipeline volume. If it is actually about the handover to your AEs, that is a different conversation and probably a more useful one." Silence often means you guessed wrong, not that they are busy.

Space in business days, with widening gaps

Count business days, not calendar days. B2B attention exists Monday to Friday, and a sequence that ignores this drops its most important messages into weekends where they sit unread and then get buried.

1. Day of accept: send nothing. An instant message reveals a process rather than a person.
2. Touch 1, one to two business days after the accept. Recent enough that they remember accepting, delayed enough to look human.
3. Touch 2, three to four business days after touch 1. Long enough that it is clearly a new thought, short enough that context survives.
4. Touch 3, about a week after touch 2. This is the ask, and it deserves distance from the value message so it does not look like the price of it.
5. Touch 4, about a fortnight after touch 3. The reframe works better with real distance, because the gap is itself a signal you are not desperate.
6. Touch 5, two to three weeks after touch 4. Then stop.

Gaps widen for a reason. Early on, context is fragile and needs reinforcing. Later, the constraint is patience rather than memory, and the only thing a short gap communicates is that you have a quota. A sequence that starts tight and ends loose reads as a person who got interested, followed up, and gracefully moved on. A sequence with even spacing reads as software.

Reply behaviour overrides all of it. Any reply ends the sequence and starts a conversation — no exceptions, and no resuming the sequence afterwards. Meanwhile, if they view your profile or engage with something you posted, pulling the next touch forward by a day or two is reasonable, because you have evidence of live attention. That is the only acceptable reason to accelerate.

Widening gaps say you are interested. Even gaps say you are automated.

Write touch 2 so it earns touch 3

The value message is the hinge of the whole sequence, and the test for it is brutal: would this be useful if they never reply? If the answer is no, it is not value, it is a pitch with a gift label.

Things that pass the test: a specific observation about how companies at their stage handle the problem, a short comparison of two approaches with the trade-off named, a number they can compute themselves from their own data with the formula shown, or a plain answer to a question their situation obviously raises. Things that fail: a case study, a link to a landing page, an ebook behind a form, and anything that ends with "happy to walk you through it".

Keep it inside the message. Any link converts a gift into a click-through request, which is an ask, which means touch 2 has become touch 3 and you now have no value message at all. If the useful thing genuinely cannot fit in a message, it is too big for this stage of the relationship.

There is a selfish reason to do this properly. A useful touch 2 is what makes touch 3's ask land as reciprocity rather than as an interruption, and it is also the only part of the sequence that does anything for the people who never reply. Some of them remember you six months later, which is the actual return on the effort.

The ask that gets a yes

Make touch 3 small, specific and easy to decline. Those three properties are what separate an ask from a demand, and the third is what makes the first two credible.

• Small: name a duration and keep it genuinely short, or ask for a reply rather than a meeting
• Specific: propose one concrete thing at a concrete time, rather than asking what works for them
• Scoped: say what the conversation would cover, so they can judge it without asking
• Declinable: give a one-word way out, such as saying that a simple no is a perfectly good answer

One option beats three. Offering a choice of times reads as accommodating but is actually more work, because the reader has to open a calendar to answer. One proposal converts the decision into yes or no, and a no is a useful outcome that an unanswered menu never produces.

If a calendar link belongs anywhere, it belongs here and nowhere earlier. Even then, put it after the proposed time rather than instead of it, so the message still works for someone reading on a phone who will not tap through. Teams that outsource this stage usually hand off at exactly this point, which is what appointment setting covers.

When to stop, and how to close

Stop after touch 5. Not because five is a magic number, but because by the fifth message you have asked in two different framings and received nothing, and the sixth message will change the reader's impression of you from persistent to unwelcome. The cost of that is not one lost deal, it is the door.

The close should do three things: state that you are stopping, say it without resentment, and leave a specific route back. Something like: "I will leave it there — clearly not the right time, and I would rather not clutter your inbox. If the handover to your AEs becomes a problem later, message me and I will pick it up from here." No deadline, no last chance, no passive-aggressive breakup line about assuming they are not interested.

Closes get replies. Not often, but reliably enough that the message earns its place, and the replies they get are honest ones: wrong time, wrong person, already solved it, or ask me in March. All four of those are more valuable than another unanswered follow-up, and "ask me in March" is a genuine pipeline entry.

What to do with the silent majority

Most people will not reply, and that is normal rather than a sign the sequence is broken. The question is what happens to them afterwards, and the default answer at most companies is nothing, which wastes the connection you paid an invite for.

• Set a re-entry window — a season rather than a fortnight — and only re-approach with a genuinely new reason: a role change, a funding round, a product launch, or something they posted
• Let content do the work in the meantime. They are connected, so what the sending profile posts reaches them without a message
• Move the ones who matter most to a different channel rather than repeating on the same one, since a second channel is a new signal and a sixth LinkedIn message is not
• Mark them clearly in your records so a future campaign does not treat them as cold and open with a relevance signal they have already seen
• Drop the ones who replied asking to be left alone, permanently and across every list

A re-approach with a real reason is a fresh first touch, not touch six. It gets the context job, the spacing resets, and it usually performs better than the original sequence did, because the reason is stronger than the one you invented for the first attempt. This is also the argument for keeping your list and contacted-log clean enough to support it, which we cover under lead generation.

Measure per touch, not per sequence

A sequence-level reply rate hides everything you need to know. Measure each touch separately: replies attributable to that touch divided by the people who actually received it. If 60 people received touch 4 and 3 replied to it, that is a 5 per cent reply rate for touch 4 specifically, and you can compare it against touch 3 on the same cohort.

Two patterns are worth acting on. If a touch produces essentially nothing across several cohorts, cut it and close the gap; a shorter sequence that never wastes a message is better than a longer one with a dead touch in the middle. And if touch 1 gets replies while touch 3 gets none, the problem is the ask rather than the targeting, because people are engaging and then stopping at the point where you request something.

Watch the negative signal too. Track how many people disconnect, mute or report across each touch. That number rising is the clearest evidence a sequence is too long or too pushy, and it costs more than a missed reply because it affects the sending profile's standing rather than just this campaign. If you would rather have the whole sequence designed, sent and answered by a team that watches those signals, that is what managed outreach is for.

Key takeaways

• Five touches, five distinct jobs: context, value, ask, reframe, close. A message doing two jobs does neither.
• Space in business days with widening gaps: two, then four, then a week, then a fortnight, then stop.
• Touch 2 must be useful even if they never reply. If it contains a link or a condition, it is an ask in disguise.
• Make the ask one specific proposal with an easy way to say no, rather than a menu of times.
• Measure replies per touch rather than per sequence, and cut any touch that produces nothing across several cohorts.

Frequently asked questions

How many LinkedIn messages should a sequence have?

Five is a sensible ceiling for a cold sequence, because by the fifth you have asked twice in two different framings. What matters more than the count is that each message has a distinct job; three well-differentiated messages outperform seven that all say the same thing in new words.

Should I message immediately after someone accepts?

Wait a business day or two. An instant message reveals that a process is running, which undermines the note that persuaded them to accept. One to two business days is recent enough that they remember accepting and slow enough to read as a person who noticed.

What if someone replies halfway through the sequence?

End the sequence entirely and handle it as a conversation. Do not resume the remaining touches afterwards, even if the conversation stalls, because a scheduled message arriving after a real exchange is the clearest possible sign that nobody was paying attention.

Can I re-approach someone who ignored a whole sequence?

Yes, after a real gap and only with a new reason such as a role change, a funding round or something they published. Treat it as a fresh first touch with its own context message rather than as a continuation. Without a new reason it is just touch six arriving late.

Does the same cadence work for recruiting outreach?

The structure holds but the spacing tightens, because candidates make decisions against a live process rather than a budget cycle. The value touch also changes character: for candidates it is usually clarity about the role and the process, not an insight about their market.