01Section 1 of 9
Start here: why cold email is changing
The old cold email playbook is losing ground for a simple reason.
Most cold email programs use the same intent signals to chase the same small group of people, then send those people three or four follow-ups in a week. The inbox gets crowded while most of the market never hears from them.
That model worked better when fewer teams had the same data. Now the same accounts sit on everyone’s lists, and the same people get slightly different versions of the same email.
When performance drops, the usual response is more follow-ups, more personalization, or more AI-generated copy. That creates more activity around the same narrow audience without fixing the real problem.
Most of the qualified market still never sees the offer.
We’ve been running a different model.
We map the full market, send one useful email to each contact every 30 to 60 days, test creative in batches, and connect every campaign to website and revenue events.
I think of the email as an ad placement and the list as an audience we control. The next send is another creative cycle instead of a follow-up asking, “thoughts?”
| Common short-sequence program (most teams) | How we run our “private ad network”-style cold email |
|---|---|
| Targets the small slice showing visible intent today | Reaches the full qualified market, which already includes the in-market buyers |
| Sends a multi-step sequence over one or two weeks | Sends one email per contact, then waits 30 to 60 days |
| Writes a few messages/angles | Tests many creatives and lets conversion data pick the winners |
| Adds more personalization or AI when results are weak | Tests the offer first, then the copy, then the audience |
| Uses reply rate as the scoreboard | Uses emails per signup, meeting, customer, and dollar of revenue |
| Treats each send like another manual follow-up | Treats each send like a measurable ad impression |
Those differences change what the channel is built to do.
A short sequence tries to get a response right now. This model reaches the market consistently, learns which combinations create business outcomes, and puts more volume behind what works.
It also changes the scoreboard.
One of the clearest examples came from RB2B.
Their free-forever offer generated one signup per 52 emails. A 7-day trial needed 643 emails for one signup.
That’s roughly a 12.4x difference in signup efficiency.
If reply rate had been the deciding metric, we might have put more volume behind the worse campaign, even though the other one drove more signups (the thing RB2B actually cared about).
Replies still matter as an operational signal. I just don’t want them deciding where the next million emails go unless they predict signups, meetings, customers, or revenue.
Where we go from here
The rest of this playbook explains how we build the system behind that measurement, even if you don’t have access to Halo (I’ll explain what that is later).
We start with deliverability, then test the offer, creative, audience, and market coverage before connecting everything to downstream attribution.
Let’s get started.