AI prompts help LinkedIn outreach get replies when you use them to research the prospect and vary the message, not to write one generic template you paste to everyone. The messages that fail are the ones that read like they could have been sent to anyone; the ones that get replies reference something specific about the person or their company in the first line. Here’s how to prompt for that specificity without spending twenty minutes per message.
Why Do Most LinkedIn Outreach Messages Get Ignored?
A connection request note on LinkedIn is capped at 200 characters for free accounts and 300 for Premium and Sales Navigator accounts, which is barely two sentences. Most people waste that space on a generic pitch — “I help companies like yours grow revenue” — instead of a specific reason for connecting. A recruiter, founder, or buyer scanning a pending request in three seconds can tell the difference between a templated line and one that references their actual role or a recent post.
Regular LinkedIn messages to first-degree connections go up to 8,000 characters, and InMail messages allow up to 1,900 characters in the body with a 200-character subject line — plenty of room, but more room doesn’t mean more words. The best-performing outreach messages tend to be short precisely because they respect the reader’s time.
How Do You Prompt for a Specific, Non-Generic Opening Line?
Give the model the prospect’s job title, company, and one real detail — a recent post, a job change, a company announcement, a shared connection or group — and ask it to write an opening line that could only apply to that person. A prompt like “write a one-sentence LinkedIn connection note referencing that [name] recently became [new role] at [company], without mentioning what I do or sell” forces the model to lead with relevance instead of a pitch.
Feed the model the actual text of the prospect’s recent post or headline rather than describing it yourself. Pasting the real source material produces a far more specific line than summarizing it from memory, and it avoids the model inventing details that aren’t true.
How Should You Prompt for the Follow-Up Message?
Once someone accepts a connection request, the follow-up message is where the actual ask happens — but it should still open with something other than your pitch. Ask the model to draft a message that references the reason you connected, states one specific, low-commitment ask (a 15-minute call, a quick reaction to an idea, a resource worth sharing), and ends with a question rather than a call-to-action button-style line. Messages that end in a genuine question get replies more often than messages that end in “let me know if you’re interested,” because a question is easier to answer than a vague open door.
Ask for three variations of the same follow-up in one prompt, each with a different opening hook, so you can A/B the angle across similar prospects instead of guessing which one will land. For prompting techniques that carry over directly from cold email to LinkedIn outreach, see our guide to AI prompts for cold email.
How Do You Keep AI-Written Outreach From Sounding Like AI Wrote It?
Explicitly instruct the model to avoid the tells: no “I hope this message finds you well,” no em-dash-heavy sentence rhythm, no exclamation points on every line, no phrase that could be swapped into a message to a completely different person without editing. Ask it to write at the reading level and sentence length of a real LinkedIn message, not a marketing email — short sentences, contractions, and a specific, checkable claim instead of a vague compliment like “I’m impressed by your work.”
Always read the draft back before sending. A prompt can get you 90% of the way to a good message, but the last pass — checking that the specific detail is actually accurate and that the tone matches how you’d really talk to this person — is worth doing by hand every time.
Frequently Asked Questions
How long should a LinkedIn connection request note be?
Free accounts are limited to 200 characters and Premium or Sales Navigator accounts to 300, so aim for one clear sentence rather than trying to fill the space. A short, specific note outperforms a longer generic one almost every time.
Should I mention what I’m selling in the first message?
Generally no. The connection request and the first follow-up work better when they lead with a genuine, specific reason for reaching out rather than a pitch. Save the actual ask for after you’ve established why you’re relevant to that particular person.
How many follow-up messages is too many?
There’s no universal number, but if someone hasn’t responded after two follow-ups spaced several days apart, a third message rarely changes the outcome and can come across as pressure. It’s usually better to add value in the last attempt — a genuinely useful resource or insight — than to simply repeat the ask.
For more on adapting outreach copy to a sales context, see our guide on AI copywriting rules for sales. For the current, official character limits referenced above, LeadLoft’s breakdown of LinkedIn’s messaging limits is a useful reference to check before you plan an outreach campaign.
