Follow-Up Timing: How Many Touches, How Far Apart, When to Stop

A cold outreach sequence should run four to six touches over two to three weeks, with the gaps widening as it goes: two to three days between the first touches, then four, then a week. Stop when a touch adds nothing new, not when you run out of templates. The number of touches matters far less than whether each one carries information the last one did not.

Spacing, Not Volume, Is the Variable

Most sequence advice argues about how many emails to send. That is the less interesting question. Four touches spaced well outperform eight spaced badly, and the reason is what the gap communicates.

A message arriving the day after the last one says the sender is working a list. A message arriving three days later, with something new in it, says someone is thinking about this. The gap is doing part of the persuading, and closing it too tightly undoes the work of the copy.

There is a mechanical reason too. Compressed sending to the same address concentrates your complaint risk, and a recipient who was mildly uninterested on Monday becomes actively annoyed by Thursday if three messages have arrived. Annoyance produces the one reply you genuinely cannot afford, which is a spam complaint (how daily volume and complaint rate interact).

A Cadence That Works

The widening gap is deliberate. Early touches are close together because the first message is easily missed and a quick second attempt is forgivable. Later touches spread out because by then the silence is informative, and pressing harder against a signal you already have is what turns a non-response into a negative one.

The fifth touch is optional and does more than it looks. A message that explicitly stops trying often gets a reply, because it costs the recipient nothing to answer and it removes the pressure that was suppressing the reply.

Every Touch Has to Add Something

This is the rule that makes the rest work, and the one most sequences break.

If touch three says the same thing as touch one with "just following up" in front, it is not a follow-up, it is a repetition with an implied complaint attached. The recipient learns that your messages contain nothing new, and that lesson applies to every future message including the good ones.

What counts as new:

• A different angle on the same problem. The first email framed it as a cost; the second frames it as a risk. Same issue, different door.

• A concrete comparable. A situation close enough to theirs to be recognisable, and what actually changed.

• Something genuinely useful, unconditionally. A relevant observation they can use whether or not they ever reply.

• A smaller ask. Reducing the request is new information: it tells them you noticed the first one was too big.

• A trigger event. Something changed at their company since you last wrote. The strongest of the five, and the least available on demand.

What does not count: checking in, bumping, circling back, wondering whether they saw the last one, or asking if now is a better time. Each of those asks the recipient to do the work of restarting a conversation you are supposed to be carrying.

When to Stop

Three stopping conditions, and the first is the one people ignore:

Stop when you have nothing new to say. Not when you reach touch six. If you run out of angles at touch three, the sequence ends at three. Padding the remainder with nudges is worse than a shorter sequence, because it spends credibility to buy nothing.

Stop at any negative signal. An explicit no, an opt-out, or a reply indicating wrong person or wrong time. Continuing past this is where complaint rates come from, and a complaint costs the domain far more than one prospect is worth.

Stop at four to six touches regardless. Beyond that the marginal reply is small and the marginal irritation is not.

Stopping cleanly matters more than it sounds. A final message that says you will stop, and means it, preserves the option of a conversation in two quarters. A sequence that simply trails off after nine unanswered emails does not, because by then you are filtered.

Where to Put Follow-Ups: Same Thread or New

Reply within the existing thread for touches two through four, and keep the original subject line.

The thread supplies context the recipient does not have to reconstruct, and it means a reader who opens touch three can see touch one directly above it. A new subject line for every touch presents each message as a fresh interruption from a stranger.

One thing not to do: fake a thread on a first touch. Prefixing an opening message with Re: to imply prior correspondence is the fastest available way to lose a reply you might otherwise have earned, and recipients spot it immediately.

A genuinely new subject is defensible on the final touch, if the angle has changed enough that the old subject misdescribes it.

Timing Within the Week

Less important than the spacing, and worth a small amount of attention.

Mid-week mornings in the recipient's timezone are a reasonable default. Monday competes with the weekend backlog and Friday afternoon competes with leaving. Those are tendencies rather than rules, and the effect is small compared with whether the message is relevant.

The genuinely important timing point is different: send in the recipient's working hours, not yours. An email arriving at 3am local time sits at the bottom of a morning backlog, and if your list spans timezones and your sending does not, you are systematically disadvantaging part of it.

Be sceptical of precise claims about the best hour to send. Anyone publishing an exact figure is describing their own campaigns to their own audience. Test it on your list and trust that instead.

What Multi-Channel Changes

Adding a channel changes the spacing rather than the count. A LinkedIn view or connection between email touches two and three fills the gap without consuming an email, and it makes the next email arrive from a name the recipient has now seen twice.

The rule that keeps this from becoming pursuit: one channel switch per sequence, not one per touch. Email, then a LinkedIn touch, then email again is a sequence. Email, LinkedIn, phone, LinkedIn, email, phone is surveillance, and it reads that way regardless of intent. How to sequence channels properly is its own subject: combining email, LinkedIn and phone in one outbound sequence.

Two Mistakes That Look Like Good Practice

Front-loading the sequence. Four touches in the first week looks energetic and produces the worst version of every problem here: it concentrates complaint risk, it reads as automated, and it exhausts your angles before the recipient has had a chance to notice the first message. The instinct behind it is that speed shows commitment. What it actually shows is that the sequence was built around your calendar rather than theirs.

Pausing the sequence and resuming it later at full intensity. A gap of several weeks resets the recipient's context completely, and picking up at touch four of six means arriving mid-argument at someone who has forgotten touches one to three. If a sequence has genuinely stalled for weeks, the honest options are to close it out cleanly or to start again with a new reason for writing. Resuming from the middle gets the worst of both.

There is a version of the second mistake that operates at the domain level too. A sending pause followed by a return to full volume resets expectations on the receiving side and looks like exactly the pattern providers filter for, which is a separate reason to ramp rather than resume (how daily volume and warmup interact).

FAQ

How many follow-ups should a cold email sequence have? Four to six touches total, including the first email. Stop earlier if you run out of things worth saying.

How many days should I wait between follow-ups? Two to three days for the early touches, widening to four or five, then a week. Compressed sequences raise complaint risk and read as automated.

Should follow-ups go in the same email thread? Yes, for the middle touches. The thread carries context and shows the recipient the earlier message. Never fake a thread on a first touch.

What is the best day to send a cold email? Mid-week mornings in the recipient's timezone are a fine default, and the effect is small. Sending inside their working hours matters considerably more than which day.

Is it worth sending a final "I will stop reaching out" email? Yes. It is often the highest-replying message in the sequence, because it is easy to answer and removes the pressure that was suppressing a response.

What if a prospect replies positively and then goes quiet mid-sequence? Different problem with different causes, and cold-sequence logic does not apply: why prospects go quiet after they said yes.

Want sequences where every touch carries something new and stopping is a rule rather than an accident? That is how Lidgen builds them, with human review on each step. Book a demo.