The playbook
Most cold email advice is taste dressed as fact. This is the opposite: the rules our own sending actually supports, measured across 44 client accounts, with what is evidence marked as evidence and what is craft marked as craft. Read the chapters, or run your email through the tools.
The four principles
Open in the reader's world. The first sentence is a fact about them, checkable from their own site. You appear in sentence two, once the email has earned it.
A number that describes you is credibility. A number that predicts the reader's future is noise. State what you have done; never promise what they will get.
Shorter wins. It is the only copy rule that survives scrutiny across the whole corpus. Under 40 words on the first touch, under five sentences everywhere.
One small ask, and never a meeting on the first touch. Ask for a reply, not a slice of someone's calendar.
The chapters
Three to five words, one checkable fact, no fake Re:. The subject earns a glance, nothing more.
READ ↗The open question beat the meeting request roughly two to one across 44 accounts. Never ask for the call first.
READ ↗Four touches, each carrying a new fact. Under 40 words on the first; the length rule is the one the data is strictest about.
READ ↗A follow-up earns its place with information, not a reminder. Three, then the honest walk-away.
READ ↗Your ideal customer profile is a countable list you can research, not a demographic you can only blast.
READ ↗Separate domains, warmed mailboxes, SPF/DKIM/DMARC. The best email lands in spam on the wrong plumbing.
READ ↗Questions
Four things, in order of how well the evidence supports them: keep it short (under 40 words on the first touch is the one rule that holds across the whole corpus), open with a checkable fact about the reader, ask a small question rather than for a meeting, and follow up only when you have something new to say. Everything else is craft, and we label it as craft.
From the cold email we run for clients: a model trained on more than 4,000,000 sends, with the specific effect sizes measured across 44 client accounts on variants with enough volume to mean something. We publish the rankings and relative differences and keep the absolute reply rates private, and we separate what is measured from what is house judgement.
Yes, if it is done as a system rather than a blast. The teams that win treat deliverability as infrastructure, research each account, keep emails short, and measure on replies and meetings rather than opens. The teams that lose buy a list, send a template, and wonder why. This playbook is the first kind.
FREE · WRITTEN · SAME WORKING DAY