How we write a cold email angle

On the homepage we say we write 10–15 campaigns for every market. People ask what that actually means, since the product stays the same. The answer: each campaign runs a different angle, and an angle is a testable guess about why this buyer, this quarter, will answer.

Here's the anatomy, and how we write them.

The three parts of an angle

Every angle that has ever worked for us reduces to the same three parts:

1. The pain hypothesis. One specific problem we believe this segment has right now. This quarter's problem, at this company size, in this function.
2. The proof. One reason to believe we've solved it before. A number, a named pattern, a result. One only; stacking proof reads like a brochure.
3. The ask. One question small enough to answer from a phone in a checkout line.

If any part is missing, it's a template blast wearing an angle's clothes.

The angle types we race

The homepage's Build stage shows three angle cards. Here's what each type looks like written out.

The pain-point angle. Leads with the problem, in the buyer's words. The test: could the reader tell we know their job? "Most Series B marketing teams we talk to have pipeline meetings that argue about attribution and end without a number anyone trusts" earns more replies than any sentence containing "streamline."

The funding-signal angle. Leads with a public trigger. A raise, a hiring spree, a new exec, a launch. The trigger buys the first sentence; the pain hypothesis still has to carry the second.

The case-study angle. Leads with a result for a company the reader would recognize as a peer. Same segment, same size, same problem. A logo from another industry impresses nobody; a peer's number keeps them reading.

We write variants across all three types, plus whatever the market's own language suggests, then race them A/B/C/D on real sends. The dashboard picks the winner.

What kills an angle before it sends

• Cleverness. If the first line needs to be admired before it's understood, it's dead. Clear beats clever every time we've measured it.
• A borrowed pain. Pains copied off a competitor's website. Real pains come from win/loss calls. The market can tell.
• The wrong owner. A perfect angle about pipeline math, sent to someone who owns brand. Angle and list are one decision, never two.
• Two ideas. An angle carries one hypothesis. The second idea goes in the next campaign, where it can be measured on its own.

How many angles is enough

Ten to fifteen per market, tested over the first weeks. Fewer, and a flat result teaches you almost nothing, since one guess failing proves only that one guess failed. More, and the volume per angle drops below what the reply math needs to reach significance.

The honest part: most angles lose. A typical race crowns one clear winner, keeps one backup, and kills the rest. That's the point. Every dead angle is a wrong guess you stopped paying for after a few hundred sends, and the winner gets your entire market.

If you want to see the first five angles we'd race on your market, that's literally what the free GTM plan is: your TAM mapped, five angles written in your buyers' vocabulary, five signal plays. Built by hand, free, delivered within a week.

Draft in hand? The free cold email grader scores it the way our internal reviews do, and the subject line analyzer handles the first five words.