LinkedIn First Message After Connecting: 8 Templates That Start Conversations (2026)

The accept is the start, not the finish. Most outreach campaigns die in exactly this gap: someone clicks „Accept,“ nothing happens, and three weeks later the connection is dead weight in a list. The first message you send right after acceptance is, by a long way, the highest-ROI message in the entire LinkedIn workflow — and the one most people get wrong. This guide gives you the 2026 data, eight templates that fit on a mobile screen, the timing rules, and the testing approach. Manual. No automation.

The data: this is the touchpoint that actually moves in 2026

Expandi published an analysis of 13.2 million LinkedIn interactions through April 2026, and inside it is the data point that should change how you spend your outreach time. The post-connection message reply rate has held flat at 10.4% across the year. Over the exact same window, the connection-request note reply rate fell from 3.5% to 2.2% — a 37% relative decline. The market is wising up to one channel and not the other.

Pair that with the data from my connection request guide: a separate study of 16,492 requests (Botdog) showed that adding a note to your connection request doubles your post-accept reply rate (5.44% → 9.36%). The acceptance rate itself barely moves with or without a note (26.37% vs. 26.42%). So the real return on writing a connection-request note isn’t the accept — it’s priming this message.

For reference, here’s where you want to land on first-degree connection replies:

• 10–15% reply rate: average. Most outreach campaigns live here.
• 20% reply rate: you’re doing the basics right.
• 25–35% reply rate: good. Personalized, well-targeted, no pitch in message 1.
• 40%+ reply rate: top-tier. Multi-touch sequences with real triggers, mobile-friendly writing, no AI boilerplate.

So you’re investing time in the right place. The next question is: what do you actually write?

What kills your first message — the 6 anti-patterns

Before the templates, the failure modes. Each of these tanks your reply rate and most outreach guides ignore them.

1. The thank-you trap

„Thanks for accepting!“ sent on its own reads like an auto-responder. The recipient has nothing to reply to. The top-ranking article for „thanks for connecting on LinkedIn“ right now is literally a list of 12 thank-you templates — and the average reply rate to those is what set the 10.4% benchmark you’re trying to beat. If you say thank you, pair it with one specific question. Or skip it entirely and go straight to the question.

2. Generic personalization

„I see we both work in tech.“ „Looks like we’re in the same industry.“ „Noticed we have a few mutual connections.“ These announce that you glanced at the profile and ran out of things to say. In side-by-side tests, generic personalization performs worse than no personalization — because it raises the recipient’s expectations and immediately fails them. If you can’t reference something specific (5 words from a post they wrote, a project they shipped, an event you both attended), skip personalization and ask a good question instead.

3. Starting with „I“

„I noticed your profile…“ „I wanted to reach out…“ „I think we could…“ The first word of your message is the loudest. Starting with „I“ puts you at the center of a message that should be about them. Open with their name, with an observation about their work, or with the trigger that prompted the message. The word „I“ should appear no earlier than the third sentence, if at all.

4. The immediate pitch

This one is the most common and the most expensive. Pitching in message 1 converts at under 2% across most B2B contexts. The reason is structural: at any moment, only about 3% of any market is actually ready to buy. An immediate pitch alienates the other 97% before you’ve earned a single signal of trust. The accept means they let you into the network, not that they’re ready for a sales conversation. Build the rapport first — at least 2–3 exchanges before anything transactional.

5. The product monologue

Five paragraphs about your platform, your 30 integrations, your „trusted by thousands of companies worldwide.“ Nobody reads it. Around 60% of LinkedIn activity happens on mobile, and on mobile every line past the first three vanishes below „Read more.“ If your message would need a scroll to be read, it’s already failed.

6. AI-generated boilerplate

I’ve tested first-message generation with every major model — Claude, ChatGPT, Gemini. With a generic prompt („write me a friendly first message after a LinkedIn connect“), the output reads identical across all three: corporate, vague, instantly spottable as AI. The only way to get a usable output is to write a prompt so specific that you may as well have written the message yourself. Skip the AI step and use a proven template with one real reference filled in by hand.

The 5 rules of a first message that gets a reply

1. Under 400 characters. LinkedIn’s own data shows messages this short get 22% higher reply rates. The default impulse is to write more to „explain yourself.“ Cut by half. Then cut again.
2. Mobile-first. Short sentences. Single-line paragraphs. No wall of text. If your message fits visibly on a phone screen without scrolling, you’re at the right length.
3. One frictionless question. The message ends with exactly one question — the kind that takes the recipient under 30 seconds to answer. Not „can we hop on a 15-minute call Wednesday?“ That’s a meeting ask. „Is this generally relevant to what you’re working on?“ That’s frictionless.
4. Trigger or profile reference. Five words of real, specific reference — their recent post, a job change, a mutual connection, an event you both attended — lifts reply rates by 32%. Generic openers do the opposite.
5. Make it about them. Don’t start with „I“. Open with their name, an observation about their work, or the trigger.

8 first-message templates that start conversations

Eight templates I and my clients have actually used. Each one fits on a mobile screen, asks one question, and skips the pitch. Token reference:

• {first} — first name.
• {company} — current employer.
• {title} — current role.
• {topic} — a specific 5-word reference (project, post theme, focus area).
• {mutual} — name of a real shared connection.
• {event} — specific event or conference name.
• {post_topic} — the topic of a post they wrote, in 3–5 words.
• {prev_company} — their previous employer (for the recent-job-change template).

1. The relevance question (default)

Hey {first}, happy to connect!

You probably saw from my profile that I'm all about {topic} — is that generally relevant to what you're working on at {company}?

Curious to hear your take.

Why it works: three sentences, no pitch, points back to the profile (which already did the heavy lifting), and ends with a yes/no question that’s easy to answer. This is the workhorse — when in doubt, send this.

2. The post-comment follow-up

Hey {first}, your post on {post_topic} actually changed how I think about it.

The bit about {one_specific_point} hit hardest. Where did you start with that approach?

Why it works: shows you actually read the post (the 5-word specific reference does it), opens with them (not you), ends with a question they’ll enjoy answering.

3. After mutual-connection accept

Hi {first}, glad we're connected — {mutual} mentioned your work on {topic} a while back.

Are you still pushing on that, or has the focus shifted?

Why it works: the named mutual carries 80% of the trust work. Only use if the mutual actually knows the recipient — bluffing here destroys credibility.

4. Same role / peer-to-peer

Hey {first}, fellow {title} here.

Always curious how others are handling {topic}. What's been your biggest unlock the last year?

Why it works: peer framing kills the „they want to sell me something“ reflex. Best when titles genuinely match — recipient sees a peer, not a vendor.

5. Recent job change

Hi {first}, congrats on the move to {company} — {topic} is a fun problem space.

What pulled you over from {prev_company}? Loose question, no agenda.

Why it works: job changes are one of the strongest reply triggers (the +32% recent-activity stat). People love talking about why they moved.

6. After a shared event

Hey {first}, missed catching you at {event} last week.

The {one_session_or_speaker} bit stuck with me — were you in that room too?

Why it works: shared context removes the cold-stranger problem entirely. Don’t fake it — only use if you both genuinely attended.

7. Inbound — they sent the request

Hey {first}, thanks for the connect.

Curious — what made you reach out? Always happy to know what caught your eye, especially if {topic} is relevant on your side.

Why it works: flips the script when they made the first move. Asks one frictionless question that signals openness without putting pressure on them.

8. Re-engaging a dormant connection

Hey {first}, realised we've been connected for a while without ever actually talking.

Working on {topic} these days — anything similar on your side?

Why it works: honest acknowledgment of the gap beats pretending you’re old friends. Low-stakes opener that often unlocks dormant network value.

One rule applies to all eight: do not paste them raw. The {topic}, {post_topic}, and {mutual} slots are where the message lives or dies. Five words of real, specific reference are what separate a template from a generic note.

This is the message that actually paid my bills

I’ve generated multiple six figures through LinkedIn outreach over the years — entirely manual, zero automation. I’ve also helped a number of clients do the same. One pattern keeps showing up in the data, my own and theirs: the highest-ROI message in the entire LinkedIn workflow is this one. The first DM after acceptance. Not the connection request note. Not the InMail. Not the comment on their post. Not the long-form article that drove the impression. This message.

The mechanism is structural. The profile did the heavy lifting before the accept — they decided you were worth a network slot based on what they saw. Once they’ve accepted, the friction is gone. There’s no inbox guard, no spam filter, no „is this a sales person?“ reflex. You have a small open window, usually 24–48 hours, where attention is on you. The first message either opens the conversation or wastes the window. There’s no third outcome.

The mindset shift that matters most: aim for a reply, not a close. The goal of message 1 is exactly one thing — a written reply, however short, in your inbox. Once that’s in hand, you have a conversation thread, you can read tone, you can find the right ask. Without it, you have nothing. Pitching in message 1 trades a 30% chance of starting a conversation for a 2% chance of closing one. That’s a bad trade.

Timing — when to send, when to follow up

• First message: within 24–48 hours of acceptance. Beyond that, the connection feels stale, the recipient has forgotten why they accepted, and the conversation rarely starts.
• Best days: Tuesday gets a 6.90% reply rate, Monday 6.85% — the highest of the week.
• Best time windows: 8–10 AM or 2–4 PM in the recipient’s timezone.
• Follow-up spacing: 3–5 business days between touches.
• Cap at 4 touches over 14 days. Past that, you’re signaling neediness and burning trust.

Each follow-up needs to bring something new — a relevant insight, a useful resource, a different angle question. Sending the same „just checking in“ three times in a row destroys reply rate and trust together. A workable 4-touch sequence:

The „breakup“ message often gets the highest reply rate of the four — recipients who meant to reply but lost the thread frequently respond to the explicit close-out.

How to test what works — the 1-second edge

LinkedIn gives you almost no outreach analytics. So testing means doing it manually — and the cleanest setup is brutally simple: pick one first-message template, send it to a meaningful batch (a week’s worth of 100 new connections in your target audience), and measure the reply rate. When you change something, change exactly one variable.

My default for years has been to send the exact same welcome message to every new accept. That’s the only way I get a clean signal — when the reply rate moves, I know exactly what moved it. Same audience, same timing, same template. Then I tweak one variable (the trigger, the question, the structure) and run another 100 contacts. For the deeper test setup on the connection-request side, see the Connection Request guide — same logic, applied a step earlier.

This is where the autofill workflow earns its place. Filling {first}, {company}, and {topic} by hand 100 times a week is the bottleneck that kills A/B testing in practice. With token autofill it’s one second per message — and the math compounds: a 2% lift in reply rate, on 100 new connections per week, over a quarter, is several additional conversations in your pipeline that wouldn’t otherwise exist.

• Hold everything else constant. Same target list, same time of day, same template structure.
• Batch of 100+ before drawing conclusions. Below that, noise drowns the signal.
• Change one variable at a time. If you tweak the trigger AND the question simultaneously, you can’t tell what moved the number.

FAQ