One trigger, one claim, one ask — written after the agent researches the account, not from a fill-in-the-blank pattern.
Write a cold email to the Head of Growth at a Series A B2B SaaS company that just posted three content marketing roles.
Subject: "three content hires". Body opens on the hiring pattern, connects it to the brief bottleneck that usually follows, makes one claim about cycle time, and asks whether they are hitting that bottleneck yet. Eighty-four words, one question.
A cold email template that hands you a paragraph with brackets to fill has skipped the hard part: knowing why this account should care this week. This cold email template researches the company first, finds a genuine trigger, then writes a short first touch built around it.
Here it is the reasoning behind an email rather than its wording: which trigger justifies reaching out, which outcome claim maps to that trigger, what the single ask is, and what to leave out. The sentences are generated per prospect, which is why the result does not read like a merge field.
One specific observation about their situation, one concrete claim about the outcome you produce, and one low-friction ask. Under 120 words. Everything else — company background, feature lists, a second question — measurably reduces replies.
Enough to prove you looked, which is usually one sentence tied to something real: a hiring pattern, a launch, a page they published, a change in their category. Personalizing the greeting and nothing else is worse than not personalizing at all, because it advertises the automation.
Three to five plain words that describe the email honestly. Curiosity gaps and fake reply threads get opens and lose trust in the first line, which costs you the reply you were optimizing for.
Indirectly and significantly. Short, relevant, link-light emails to accounts that actually fit generate engagement, and engagement is what inbox providers score. Volume without relevance is the fastest route to the spam folder.
The agent researches the account from public sources, checks fit against the ICP definition in workspace Knowledge, and pulls your positioning, proof points, and voice so the claim is one you can defend. Drafts export to your sequencer or CRM, and the follow-up sequence template takes over from touch two.
A company and a role is enough; it takes your positioning and ICP from Knowledge.
It looks for a real trigger and will tell you when there is not one worth using.
Subject line, body under 120 words, and the trigger it relied on so you can verify it.
Push to your sequencer, then generate the follow-up arc from the same research.
Researches the account before writing rather than filling in brackets
Tells you when there is no honest trigger, instead of inventing one
Claims come from your real proof points, so you can defend them on a call
Records what was referenced so follow-ups add new angles
Yes, at low volume with high relevance. Inbox providers score engagement, so a small number of well-researched emails to genuinely fitting accounts outperforms mass sending — and mass sending increasingly damages the domain you need later.
Lines that state something specific and verifiable about their situation, then connect it to a consequence they recognize. Openers about yourself, or generic compliments about their company, are read as noise.
Under 120 words, ideally under 90. It should be readable on a phone without scrolling, because that is where it will be read.
Do not open with your company, stack two asks, attach anything, use a fake reply subject line, or send to an account that fails your own qualification criteria. That last one causes more damage than any copy mistake.
They read like a person who did the research, because the research actually happened and the wording is generated per account rather than templated with variables.