Outreach Messages That Get Replies
You've found the perfect candidate. Their LinkedIn profile is a dream match for your role. You craft a message, hit send, and... nothing. No reply. No view. Just silence.
Sound familiar? The average InMail response rate is around 10-25%. That means for every 10 messages you send, 7-9 go unanswered. If you're spending 5 minutes per message, that's a lot of wasted time on messages that never land.
Today you're going to fix that. AI doesn't just make outreach faster — it makes it better. Personalized, relevant, and structured in a way that makes candidates actually want to respond.
Why most outreach fails
Before we fix the problem, let's understand it. Most recruiter outreach falls into one of three traps:
The copy-paste trap. "Hi [Name], I came across your profile and was really impressed. I have an exciting opportunity..." Every candidate has received this message 50 times. They can smell a template from the subject line.
The essay trap. Three paragraphs about the company, two paragraphs about the role, one paragraph about the team. The candidate reads the first line, sees a wall of text, and moves on.
The vague trap. "I have a great opportunity that might interest you." What role? What company? What salary? Why should they care? Vague messages get vague results — which is no response at all.
What works instead: Short, personalized, specific, and with a clear reason to respond. That's exactly what AI helps you produce at scale.
The anatomy of a high-response outreach message
After analyzing thousands of successful recruiter messages, a clear structure emerges. The best outreach messages contain these four elements — and nothing else:
1. Personal hook (1 sentence) — Reference something specific from their profile. Not "I was impressed by your experience" but "I noticed you led the migration from monolith to microservices at Deliveroo — that's exactly the challenge we're solving." This proves you've looked at their profile.
2. The opportunity (2-3 sentences) — What's the role, who's the company (or describe it if confidential), and what makes it interesting? Be specific: "Series B fintech, 80 people, building the payments infrastructure for Southeast Asia. They need a VP of Engineering to scale the team from 15 to 40 over the next year."
3. Why them specifically (1 sentence) — Connect their experience to the role. "Your experience scaling engineering teams at a similar stage — and your payments background — makes you a really strong fit."
4. Easy next step (1 sentence) — Make it low-effort to respond. Not "Let's schedule a call" but "Would a 15-minute chat this week be worth it, even just to hear more?" Remove friction.
Total length: 4-6 sentences. That's it. No company history. No exhaustive job description. Just enough to get a reply.
AI prompt templates for different candidate types
Here are three ready-to-use prompts. Copy these and customize with your details:
For passive senior candidates (directors, VPs, C-suite):
"I'm recruiting for a [title] role at [company description — stage, size, what they do]. Here is the candidate's LinkedIn profile summary: [paste key details]. Write a short LinkedIn InMail (4-6 sentences max) that: 1) Opens with something specific from their profile that connects to this role, 2) Describes the opportunity briefly, 3) Explains why they'd be a great fit specifically, 4) Ends with a low-pressure ask for a 15-minute chat. Keep the tone professional but human — not corporate or salesy."
For tech candidates (engineers, data scientists, product managers):
"I'm recruiting for a [title] role at [company description]. Tech stack: [list]. Here is the candidate's profile: [paste details]. Write a concise outreach message that: 1) References a specific project or technology from their background, 2) Mentions the technical challenge (not just the job title), 3) Includes one compelling detail about the engineering culture or team, 4) Asks if they'd like to hear more. Keep it under 5 sentences. Engineers prefer brevity."
For career changers or adjacent candidates:
"I'm reaching out to someone whose background is in [their current field] for a role in [target field]. Their profile: [paste details]. Write a message that: 1) Acknowledges their current expertise, 2) Connects transferable skills to the new role, 3) Explains why their different background is actually an advantage, 4) Keeps the tone encouraging and curious. Under 5 sentences."
The 30-second personalisation check
AI gives you a strong 90% draft. Your job is the final 10%. Before sending any AI-generated outreach:
Check the personal hook. Does it reference something genuinely specific about this candidate? Not just their job title, but a project, an achievement, or a career move that's relevant to the role. If the hook could apply to anyone with a similar title, make it more specific.
Check the tone. Does it sound like you? If you're naturally casual, soften any stiff language. If you're more formal, tighten it up. Candidates respond to humans, not robots.
Verify the facts. Make sure AI hasn't assumed anything about the candidate that isn't in their profile. Don't claim they "led a team of 50" if their profile says "managed a team."
Check the ask. Is the next step genuinely easy? "Would a 15-minute chat be worth it?" is better than "Are you available for a call at 2pm on Thursday?" Remove friction.
Scaling outreach without losing quality
Here's the workflow that lets you send 20-30 personalized messages per day in under an hour:
1. Batch your sourcing. Identify 20-30 candidates in one sitting. Copy their key profile details into a simple list.
2. Run them through AI. For each candidate, paste their details into your prompt template. AI generates the message in 10-15 seconds.
3. Quick review and send. Scan each message, tweak the hook if needed, and send. With practice, each message takes 30-60 seconds to review and send.
The maths: 30 messages × 1 minute each = 30 minutes. At a 20% response rate, that's 6 conversations started. Compare that to manually writing each message at 5 minutes each — 2.5 hours for the same result.