ChatGPT · Workflow · 3 min read · Last verified 17 Sept 2026
Cold emails that get replies
Research the specific recipient first and open with something only true of them, then keep the ask to one sentence, most cold email fails because it's generic and asks for too much too soon. AI is useful here for the research step and for ruthlessly cutting length, not for inventing a generic template.
Free PDF
Cold emails that get replies, as a printable PDF
Enter your email and the download opens straight away. You also get occasional free workshops, tools and prompts, unsubscribe any time.
What it is good at
- Fast at researching a company or person's recent public work to personalise the opening line.
- Good at ruthlessly cutting length, cold emails need to be short.
- Can generate several subject line options to A/B test.
- Useful for drafting a short, non-pushy follow-up sequence.
Where it falls short
- It can't verify facts about a person beyond what's publicly findable, always sanity-check the personalised detail.
- Left unchecked it drifts toward generic template language that reads as obviously automated.
- It doesn't know your relationship history with a contact unless you tell it.
How to use it for creative work
- 01
Research the specific person or company
Ask the model to find one or two recent, specific, public facts about the recipient, a launch, a post, a hire, that could open the email. Generic 'I love your work' openers get deleted.
- 02
State the one ask clearly
Decide the single thing you want, a 15-minute call, a reply, a portfolio look, and tell the model to build the whole email around getting only that one response, not multiple asks.
- 03
Draft short, then cut further
Ask for a first draft under 100 words, then ask it to cut that by a further 30% keeping the specific detail and the ask intact. Cold email length is a major factor in reply rate.
- 04
Test subject lines
Ask for five subject line variants: specific and curious, direct and short, and a question format. Test a couple of these across sends rather than guessing which works.
- 05
Plan a short follow-up, not a nag
Ask for one polite follow-up for a week later that adds new value or information rather than just asking again, since a bare 'just following up' rarely improves reply rates.
Ready-to-use prompts
Research [person/company name] and find one or two recent, specific, publicly verifiable facts I could reference to open a cold email, a launch, a post, an award, a hire. Give me the source for each.
Write a cold email to [name/role] at [company]. Opening hook: [paste researched fact]. The single ask is: [state one specific ask]. Under 100 words. No generic flattery, no multiple asks.
Cut this email by roughly a third while keeping the specific detail and the single ask intact. Remove any sentence that isn't doing real work: [paste draft]
Give me five subject line options for this email: one specific-and-curious, one direct-and-short, one framed as a question, and two more of your choosing. Under 6 words each. Email: [paste]
Write a short follow-up to send one week after this original email if there's no reply: [paste original]. Add one small new piece of value or information rather than just 'checking in'. Under 50 words.
Questions people ask
- How long should a cold email actually be?
- Under 100 words is a reasonable target for most cold outreach, long emails to strangers get skimmed or ignored.
- Is personalisation worth the extra research time?
- Yes, a specific, verifiable detail about the recipient consistently outperforms generic flattery, and the research step is exactly what AI speeds up.
- How many follow-ups is too many?
- Two to three spaced-out follow-ups with new value each time is a common approach; repeated bare 'just checking in' messages tend to hurt rather than help.
Your next step
Go from free tool to real results
Get free workshops, tools and prompts
Occasional updates from Fashion AI School: free live workshops, new AI tools and new prompts. No spam, unsubscribe any time.
