Skip to content
ColdReach

B2B cold email templates that earn replies and how AI writes the first line

A template is a starting structure, not a finished email. Below are six complete B2B cold email templates with subject lines and bodies, each built for a different situation. After them we show how per-prospect research changes the first line, using fictional prospects described in general terms, so you can see what separates a template from an email someone wants to answer.

Hands typing on a laptop at a home office desk next to a notebook with a handwritten prospect list

What makes a B2B cold email template work

Every template here follows the same rules. It is short enough to read on a phone. It names one problem and one outcome. It asks for one small next step. It sounds like a person, not a brochure. Curly braces mark the parts you replace for each prospect, and the first line is always the part that deserves the most care.

Keep subject lines plain and honest. A subject that pretends to be a reply, a calendar invite or an internal message is deceptive, it breaks the spirit of CAN-SPAM and it destroys trust the moment the email opens. Our CAN-SPAM cold email checklist covers the rest of the rules.

The observed trigger template

Use it when something visible changed at the prospect's company: a new product page, a new market, a hiring push.

Subject: {company} and {trigger}

Hi {first name}, I noticed {specific change on their site or public profile}. Teams that make that move usually run into {problem} within the first few months. We help {type of company} {outcome} without {common cost}. Would it be useful if I sent over a short breakdown of how that works for a team your size?

The peer problem template

Use it when you know a problem common to companies like theirs.

Subject: {problem} at {type of company}

Hi {first name}, most {role} leaders I talk to at {type of company} say {problem} takes more of their week than it should. We built {product} to {outcome}, and it fits alongside the tools you already use. Is this on your list for this quarter, or is the timing wrong?

The specific idea template

Use it when you can offer one concrete suggestion based on research.

Subject: An idea for {company}'s {area}

Hi {first name}, I was reading {page on their site} and had one idea. {One sentence describing the idea and why it fits them}. This is the kind of thing we help {type of company} put in place. Want me to send the two-step version?

The right person check template

Use it when you are not sure who owns the problem.

Subject: Who owns {area} at {company}

Hi {first name}, I am trying to reach whoever looks after {area} at {company}. We help teams like yours {outcome}. If that is you, I would value ten minutes. If it is someone else, would you point me in the right direction?

The short follow-up template

Use it as step two, three to four business days after the first email, in the same thread.

Subject: Re: {original subject}

Hi {first name}, a quick follow-up on my note about {problem}. One thing I left out: {a new, useful detail such as how setup works or what changes in the first week}. If it is not a priority right now, a one-word reply is enough and I will close the loop.

The polite close template

Use it as the last step of a sequence.

Subject: Closing the loop, {first name}

Hi {first name}, I have not heard back, so I will assume {problem} is not a focus right now and stop here. If that changes, this thread is the fastest way to reach me. Thanks for your time.

How AI personalization changes the first line

The templates above share one weak spot: the first line. Writing it well means opening the prospect's website, reading what they do and finding a detail worth mentioning. For five prospects that is fine. For two hundred a week it takes a rep hours, so the line becomes generic, and generic first lines get ignored.

An AI SDR closes that gap. It reads each prospect's website and public company profile, summarizes what they sell and to whom, and drafts a first line and body in your voice. The rep reviews, edits and approves. Here is how that plays out with three fictional prospects.

A 40-person logistics software company

Generic first line: "I hope this email finds you well and that things are going great at your company."

Researched first line: "Your new route planning page leads with fewer empty miles for regional carriers, which tells me your buyers are fleet managers watching fuel costs."

The second version proves the sender looked, and it sets up an offer aimed at the people this company sells to.

A regional accounting firm with three offices

Generic first line: "I came across your firm and wanted to reach out."

Researched first line: "I saw your team now offers outsourced CFO services to manufacturers, alongside tax and audit work."

That detail points the conversation at a new service line, where a firm is most open to new tools and partners.

A 15-person design agency

Generic first line: "Love what you are doing over there."

Researched first line: "Your case study page focuses on product launches for hardware startups, which is a narrow and smart niche."

Praise is cheap. A specific observation shows attention and gives the prospect a reason to keep reading.

Rules for reviewing AI drafts before they send

AI drafting is fast, so review has to be disciplined. Before approving a draft, check these points.

  • Every fact in the first line appears on the prospect's site or public profile.
  • Nothing is invented: no fake mutual connections, no made-up results, no claims you cannot back.
  • The tone matches your brand and the setting you chose: direct, friendly or consultative.
  • The ask is one small step, not a thirty-minute call request in the first email.
  • The subject line is honest and matches the body.
  • Your signature carries your physical postal address and the unsubscribe link is present.

In ColdReach a human approves AI emails before the first send by default, which keeps the speed of drafting without handing judgment to a model.

Turning templates into a tested sequence

Pick two templates that fit your offer, write a strong first line for each prospect and run them side by side. Test one thing at a time, starting with the subject line, then the first line, then the body. Judge winners by replies on your own data rather than by anything you read online. Keep the sequence to three or four email steps and consider adding a manual LinkedIn task between them, as described in our guide to LinkedIn outreach without automation.

Use the templates, then let the research scale

Templates give you structure. Research gives you a reason to be read. Browse more structures in our cold email templates library, or try the cold email generator: enter your website, pick who you want to reach and see three researched emails written for fictional prospects in the demo.

Write my emails