LinkedIn Connection Request Acceptance Rate: What’s Normal in 2026 (and Why It’s the Wrong Target)

TL;DR: The average LinkedIn connection request acceptance rate is 28.5% in 2026, and 30 to 45% is a healthy range. Below 20% is a targeting problem, not a wording problem. Personalizing the note barely moves acceptance at all. What it moves is the reply rate after the accept, and that’s the number worth chasing.

That last part is where most advice gets it backwards. There’s data behind it.

What counts as a normal LinkedIn connection request acceptance rate

Expandi’s 2026 outreach benchmarks put the platform-wide average at 28.5%, drawn from 13.2 million connection requests sent across 2025 and 2026. Their read on the spread:

• 30 to 45% is a healthy campaign
• Below 20% points at targeting or at the profile people land on, and it’s the one band Expandi calls a problem rather than a variation
• 60% and up is top-tier, and it almost always means you’re sending to people who already recognize your name

(Methodology note: vendor aggregate, measured sender-side across mixed industries. Directional, not gospel. Your own last 200 requests beat anyone’s published benchmark.)

So if you’re sitting at 25%, you’re normal and you can stop rewriting your opener. If you’re at 15%, something upstream of the message is broken.

The finding that should change what you optimize

Belkins ran a LinkedIn outreach study across 14,077 tracked contact records plus 15.1 million outreach touchpoints, and split the requests by whether the sender attached a personalized note. The result:

• With a note: 25.3% accepted, 8.2% replied afterward
• Without a note: 27.6% accepted, 5.3% replied afterward

Read it twice, because it’s the opposite of what every template post promises. The personalized note did not raise acceptance. Blank requests got accepted slightly more often. What the note did was lift replies from 5.3% to 8.2%, a 55% improvement in the only step that turns into an actual conversation.

(Methodology note: Belkins’ own campaign data combined with an Expandi dataset covering 2025 campaigns. Expandi’s write-up of the same study quotes slightly different figures. The numbers above are the ones published on Belkins’ own page, which is the primary source.)

Which means acceptance rate is a terrible thing to optimize on its own. You can raise it tomorrow by sending blank requests to anyone with a pulse, and your pipeline will get worse. The metric that survives contact with reality is accepted and replied. I wrote about what that note should actually say in LinkedIn connection request message.

Your industry sets the floor, not your job title

Expandi’s acceptance rate breakdown has a second finding people skip past. Industry swings acceptance far harder than seniority does. Broadcast media ran 40.1%. Consumer electronics ran 17.5%. Meanwhile every seniority bucket landed inside a three-point band, from 26.3% for junior senders up to 29.4% for C-level.

That’s a 23-point spread on industry against a 3-point spread on who you are. Most of the distance between your acceptance rate and someone else’s screenshot is the market you sell into.

Two things follow. Stop benchmarking against a generic 30%, and benchmark against your own last month instead. And if your rate really is low for your vertical, the fix is who you send to, not what you type.

Why I stopped tracking mine

I used to check acceptance rate weekly, the way you’d check a step count. It went up when I got less picky about who I sent to, which felt like progress and wasn’t. Nobody accepted, nobody answered, and I’d added a hundred strangers to a network I use to find coaching clients.

What changed it was the order of operations. I stopped starting with a list and started with the feed. I’d find people already talking about the problem I coach on, comment with something worth reading, wait a day, then send the request. Acceptance went up, but the part that mattered was that people wrote back, because by then I wasn’t a cold name in a notification.

The cost was time. Finding those people meant most of an hour scrolling before I’d done any real work, and on a heavy client day it just didn’t happen. That’s the specific chore LeadBase automates. It surfaces who engaged with what, so the warm queue is already sitting there. I still write every note myself.

Four things that actually move the number

1. Send while the signal is warm. A request that follows a comment you left on Tuesday behaves nothing like one sent cold. Recency does more work than any phrasing trick. The full version is in how to network on LinkedIn.
2. Fix the profile the request lands on. People accept or ignore based on your headline and banner, viewed in a two-second glance from a notification. That glance is doing more for your acceptance rate than your 200 characters.
3. Withdraw stale pending invites. A pile of ignored requests drags the rate down and tells you nothing useful while it sits there. Clear them monthly. Separately, and this is a different question with a different answer, what actually gets accounts limited is covered in LinkedIn automation safe account rules.
4. Keep the note, but keep it for the reply. The Belkins split says the note earns its keep after the accept, not before it. Counted examples that fit inside the real 200-character limit are in LinkedIn connection message examples.

FAQ

What is a good LinkedIn connection request acceptance rate?
30 to 45% is healthy against a 2026 platform-wide average of 28.5%. Above 60% is top-tier and usually means warm targeting. Below 20% is worth treating as a problem with who you’re sending to, not with how you worded it.

Does personalizing a connection request improve acceptance?
Barely. In Belkins’ study, requests with a note were accepted 25.3% of the time against 27.6% for blank ones. The note’s value shows up later, lifting replies from 5.3% to 8.2%.

Why is my LinkedIn connection request acceptance rate so low?
Usually targeting or industry, not wording. Acceptance ranged from 17.5% to 40.1% across industries in Expandi’s 2026 data while seniority moved it by only three points. Check your rate against your own vertical before you rewrite anything.

How do I calculate my LinkedIn connection request acceptance rate?
Accepted requests divided by requests sent over the same window, usually 30 days. Count withdrawn and still-pending invites as not accepted, otherwise the number flatters you and hides the problem.

What to do next

See how it works. LeadBase surfaces who engaged with what on LinkedIn, so there’s a warm name in the queue before you hit Connect. Take a look.

Go deeper. The step that lifts both acceptance and replies happens before the request. Start with the LinkedIn comment strategy.

Start free. Stop optimizing a rate you can game and go fix the week before the request. Start your free LeadBase trial.