The riskiest month in an outbound programme is the one right after it starts working.
Replies arrive. Two calls get booked in a week. Somebody senior finally looks at the dashboard, and the instinct is immediate and wrong: do more, add LinkedIn, double the list, get a second tool in here. Four weeks later the numbers are worse than they were before anybody was paying attention, and nobody can say which change did it, because everything changed at once.
So the short version of how to add a second channel is this. You add capacity to the thing that works, you change nothing else in the same month, and the second channel starts by touching people you have already chosen rather than people you have not.
Here is the order I use.
1. Write down what "working" means, before you add anything
A number you agreed to in advance.
For a cold email programme that means replies and booked calls over a full 30 days, from a list you can describe in one sentence. Opens do not count, and neither does a good week. If you cannot state the channel's current reply rate and its calls-per-month from memory, the channel is still early, and adding a second one will produce two things you cannot read instead of one.
Worth saying plainly: this whole post assumes the first channel is producing. If it is not, the question in front of you is whether cold outbound fits your business at all, which is a different conversation and a cheaper one to have now.
One client of mine, an IT company in Germany, reached three to five calls a day on a channel that had been running long enough to be boring, and had 84 calls booked in a month before anybody proposed changing anything. That is the state you add to.
2. Freeze the working channel for the whole experiment
Same copy. Same list criteria. Same volume per mailbox. Same sequence length.
This is the step people skip, and it is the one doing all the work. If you change the email copy in the same month you switch LinkedIn on, you have no way to attribute the movement, and the answer you take away from the quarter will be wrong in a way that costs you the next quarter too. Fix the funnel first, then scale it. Do not do both at once and call it momentum.
3. Point the second channel at the same list first
The instinct is to build a new list for the new channel. Resist it for the first month.
Most of the people on your list have not heard from you enough times yet. Even the best reps in RAIN Group's research, as quoted in a 2026 roundup, need about five touches to book a first meeting. A second channel's first job is to add touches to people you already decided were worth contacting. Doubling the number of strangers in the system can wait.
There is a second reason. Running the new channel over a known list is the only clean test you will ever get of whether the channel adds anything, because the audience is held constant.
4. Give the new channel its own infrastructure and its own start date
A new channel needs the same unglamorous setup the first one needed. New LinkedIn profiles are new accounts and behave like new accounts. New mailboxes need domains, records and warm-up time before they touch a prospect.
Which means the second channel does not go live the week it is decided. Put the setup weeks on the calendar in front of whoever approved it, so that the silence is expected rather than read as failure. Anybody who lived through the first ninety days of the original channel already knows what that month feels like.
5. Decide the order of the touches, and make sure nobody gets two cold openers
Write the sequence out as one line per touch, per person, with days between them. One channel opens. The other follows.
The failure this prevents is small and common: a prospect receiving a connection request on Tuesday and a cold email on Wednesday, each written as a first contact, each pretending the other did not happen. It reads as two different people from the same company who are not talking to each other, which is exactly what it is.
The full touch order, with the days between each touch, has its own post on running both channels as one sequence. This step is only about which channel goes first and what the gap between them is.
6. Run it for 30 days, measured separately
Separate tracking from day one, or you will spend the following month arguing about attribution.
At the end of 30 days you want three numbers for the new channel on its own: people contacted, replies, calls booked. Compare them against the same three for the frozen channel over the same window. Then, and only then, decide whether the second channel earned a permanent place.
Then scale horizontally, and only then
Scaling outbound means adding more accounts to whatever already produces meetings. More LinkedIn profiles if LinkedIn is producing them, more mailboxes if email is. That is the whole mechanism, and it is deliberately unexciting.
Volume comes last. Applied to a channel that has not proved itself, it produces the same result faster and more expensively. Add a channel and enlarge the list in the same fortnight, and the real cost is not the money. It is losing the ability to explain what happened.
If one channel is producing meetings for you right now, the correct next action is to write down its three numbers today, while they are true, and then not touch it for a month.