How long should a multichannel sequence be?

For a cold list, an intro plus two or three follow ups on the first channel, spread over about two to three weeks, then a short second channel step, is enough. In the campaigns we can show, between 61% (Shelly) and 85% (Lamothermic) of replies arrived by the first follow up, and each later touch brought fewer replies and more refusals. Length should fall as the relationship warms, not rise.

Sequence length is the setting people argue about most and measure least. Vendors default to long ones because more touches look like more effort. The campaigns we run show a different shape: most of the value is early, and the tail costs goodwill. This post sets out the evidence and then gives lengths by situation.

Where do replies actually come from?

Three studies, three shapes.

Lamothermic, LinkedIn, cold, US manufacturing. Of the prospects who replied, the first reply came at the intro for 56%, at follow up 1 for 29%, at follow up 2 for 8% and later for 7%.

Shelly Group, email, warm partner base. Share of all replies by touch: first email 24%, follow up 1 37%, follow up 2 17%, follow ups 3 and 4 together 22%. Follow up 1 was the best single touch.

Sellinger, our own LinkedIn outbound. 80% of conversions came from a reply to the first message. The reply rate at each touch, among people who had not yet replied, fell from 26.5% at the intro to 9.8%, 6.8%, 5.6% and 4.7%.

In every case the first two touches hold most of the value, and the third onward holds a tail. The tail is not worthless. Shelly's touches 3 and 4 produced 22% of replies. But it has a price, below.

What does the tail cost?

Refusals. The share of first replies that were negative:

• Sellinger: 0.9% at the intro, 8.5% at follow up 1, 14.3% at follow up 2, 14.7% from follow up 3.
• Wealth advisor network: 0.4%, 2.8%, 9.6%, 16.0%.
• ShortVibe: 0.7%, 9.1%, 19.2%, 20.0%.

And in the Shelly programme, the only two "please stop" complaints across 1,966 installers and more than 6,700 emails came after touches 4 and 5. Shelly capped the sequence at the first email plus two follow ups.

So the tail brings a few more replies and a rising share of people who would rather not hear from you. Whether that is worth it depends on the relationship.

How long should it be, by situation?

These are starting points and every value can be changed per campaign.

How long in days?

Gaps matter as much as counts.

• LinkedIn: our own gap was five days. With an intro and three follow ups, a sequence spans about 15 to 20 days.
• Email: Shelly's follow ups came a median 8.3 days after the first email, then about ten days later. Sellinger's email defaults start at one to three days between follow ups, which suits inbound or time sensitive offers more than a cold list.
• Replies take time. On our own account the median from first message to first reply was 10 hours, and in the Shelly programme the median from first email to first reply was 8.8 days. Do not judge a touch before people have had time to answer.

What about the second channel?

A second channel should change the door and the framing. Sellinger's defaults move a lead from LinkedIn to email after four follow ups, and from email to WhatsApp after five, with every value editable. We suggest using fewer than the defaults for a cold list. The mechanics are in a multichannel sequence that works. A reply on any channel should stop the others.

What does the arithmetic look like on 500 people?

A worked example using the reply rates from our own LinkedIn account, among people who had not yet replied: 26.5% at the intro, 9.8%, 6.8%, 5.6% and 4.7% at the four follow ups. Start with 500 people messaged, and treat the rates as an illustration of the shape, not a forecast for your list.

The intro gives about 133 replies. Follow ups 1 to 4 together add about 90. Follow up 4 adds 14 replies, and on the refusal numbers above, a share of any first replies at that stage will be refusals. The question is not whether the fourth follow up works, it is whether 14 replies justify the people who would have preferred silence.

What about the exit message?

A sequence needs a clean end. After the last follow up, stop, and do not send a "closing the loop" message that pretends to be a favour. If you want an exit, make it one line that gives the person a reason to come back, such as where to reach you. Then leave the lead alone. A person who stays silent through a full sequence has answered.

Should the second channel come after silence or in parallel?

After silence. A reply on any channel should stop the others, and running two channels at once on a cold lead doubles the pressure without a second chance to be seen. In Sellinger the default moves a lead from LinkedIn to email only after the LinkedIn follow ups are used up, and every value can be changed per campaign.

How do you decide for your own campaign?

Write three numbers on a sheet before you launch: the most touches you will send on one channel, the days between them, and the refusal share at which you will cut a touch. For example: intro plus three follow ups, five days apart, and cut a touch when refusals at that touch exceed positive replies. After four weeks, replace the guesses with your counts. If a touch brings fewer positive replies than refusals, remove it from the next campaign.

How do you tune length for your list?

1. Start with intro plus two follow ups.
2. After four weeks, count first replies by touch for your own list.
3. Count refusals by touch.
4. Add a touch only if its positive replies justify its refusals. Remove one if refusals outrun replies.
5. Test a second channel on a split of at least 100 people per arm.

The goal is the shortest sequence that captures most of the replies, not the longest one you can defend. For how the number depends on the relationship, read how many follow ups.