Subject Lines That Get Opened
A brilliant email that never gets opened has achieved nothing. The subject line is the single highest-leverage six to ten words you will write, because it is the entire basis on which a busy reader decides whether to open your message now, later, or never. AI tools are particularly good at subject lines because the task is short, pattern-based, and easy to generate many variations of quickly, which suits how language models work. This lesson shows you how to get genuinely strong subject lines out of ChatGPT, Claude, Gemini or Copilot, instead of the generic, forgettable ones they produce by default.
What makes a subject line work
Across internal emails, client updates and cold outreach, the subject lines that reliably get opened share a few traits. They are specific rather than vague, “Contract renewal, action needed by 1 August” beats “Following up”. They set correct expectations about what is inside, so the reader is not misled or annoyed. They are short enough to display in full on a phone, roughly under fifty characters. And where relevant, they signal urgency or relevance to the specific reader rather than reading like a mass broadcast. Vague subject lines such as “Quick question” or “Checking in” get opened out of curiosity once and then get filtered as spam-like by an experienced inbox owner the second or third time.
The default AI weakness on subject lines
Left unguided, most AI tools default to safe, slightly bland subject lines, “Update on Project X” or “Following Up on Our Conversation”. These are not wrong, but they are not memorable, and in a crowded inbox, unmemorable loses. The fix is the same principle from earlier lessons: give the AI more of the real content to work with, and ask explicitly for several distinct options rather than accepting the first one.
A strong prompt looks like this: “Give me five subject line options for an email to a prospective client, Jordan, proposing a 30-minute call about reducing their shipping costs by an estimated 15 percent. Mix styles: one direct and factual, one that leads with the number, one that asks a short question, one under 6 words, one slightly more formal for a first contact.” Asking for a mix of styles in one prompt, rather than five similar variations, produces more genuinely useful options to choose between.
Subject line styles and when to use them
| Style | Example | Best for |
|---|---|---|
| Direct and factual | Invoice 2214 overdue, payment needed by Friday | Internal updates, payment chasers, deadlines |
| Number-led | Cut your shipping costs by 15 percent | Cold outreach, sales, proposals |
| Short question | Still interested in the Q3 partnership? | Follow-ups, re-engagement emails |
| Ultra-short | Re: Thursday meeting | Fast internal exchanges, scheduling |
| Personalised reference | Following our chat at the Leeds conference | Warm outreach, referrals, networking follow-ups |
Avoid the traps that hurt deliverability and trust
Some subject line patterns that AI tools sometimes suggest by default should be avoided in professional email. All capital letters or multiple exclamation marks read as spam, both to email filters and to human recipients, and should be removed on sight from any AI draft. Misleading urgency, such as “Final notice” for a routine first email, damages trust the moment the recipient realises it was not really urgent, and professional readers remember who does this. Clickbait-style vagueness borrowed from consumer marketing, “You won’t believe this”, has no place in business correspondence and instantly signals an inexperienced or careless sender. When reviewing AI-generated subject line options, apply a simple honesty test: does this line accurately represent what is inside the email? If not, edit it or ask the AI to try again with an instruction like “make it accurate, no false urgency.”
Testing and refining
If you send similar types of email regularly, sales outreach, client check-ins, internal reports, treat subject lines as something to improve over time rather than get right instantly. Keep a short running note, even a simple text file, of which subject lines you have used and which ones seemed to get faster replies. Feed that pattern back into your prompts: “Here are three subject lines that worked well for me before: [list]. Write five more in a similar style for this new email about [topic].” This turns your own track record into a growing, personalised prompt asset rather than starting from zero every time, an idea we build on further in the lesson on templates and reusable prompts later in the course.
Matching the subject to the body
Finally, always write or at least review the subject line after the body is finalised, not before, since edits to the message often shift its real headline. A quick closing prompt to any AI tool, “Now that the email body is final, suggest a subject line that matches it exactly,” ensures the two are aligned and prevents the common mismatch where a subject line promises one thing and the email delivers another.
With attention now on getting emails opened, the next lesson turns to one of the highest-stakes uses of AI email writing: cold outreach and sales emails, where subject lines, structure and tone all have to work together under real pressure to generate a reply.
