
What's a good LinkedIn connection acceptance rate?
Benchmarks from 13M requests, what counts as low, and a two-week way to find out why your LinkedIn invitations aren't accepted.
Short answer
Across 13.2 million LinkedIn connection requests sent between May 2025 and April 2026, the average acceptance rate was 28.5%. Treat 25–35% as normal for cold B2B outreach, 35% and above as good, and anything under 20% as a sign that something specific is broken. It's usually the list, not the note.
Two cautions before you compare yourself to that number. First, measure after at least a week: most people who will accept do it within seven days, so a campaign that is three days old always looks worse than it is. Second, the benchmark is an average over everyone, including recruiters and VC partners, whose requests get accepted far more often than a software seller's.
- average acceptance across 13.2M requests
- 28.5%
- of connection notes get a direct reply
- 3.0%
- reply rate on messages sent after accepting
- 10.4%
What the benchmarks actually say
The biggest public dataset is from Expandi, an outreach tool, which pooled 13,218,869 requests from 13,302 accounts on its platform (Expandi benchmarks). Acceptance held between 28% and 30% in eleven of the twelve months, with one dip to 25% in November 2025. Belkins, an agency, puts the 2025 average at 26% (Belkins study). Botdog measured 37% on its own 16,492 invitations (Botdog), which tells you more about Botdog's users than about LinkedIn.
All three are vendors. They have an interest in outreach looking like it works. But the large samples agree with each other within a few points, so the "high twenties" figure is solid enough to plan around.
The industry you sell in moves the number more than anything you write:
Acceptance rate by sender's industry
Recruiters and investors get accepted because people expect something useful from them: a job, money. A software seller gets accepted at roughly the average. Seniority of the sender barely matters: C-level senders got 29.4%, managers 27.3%, junior people 26.3%. A fancy title won't save a bad list.
How to read your own number
| Your acceptance rate | What it usually means |
|---|---|
| Under 15% | The audience is wrong or the profile looks like a bot. Stop and fix before sending more |
| 15–25% | Below the average. Check targeting first, then the profile |
| 25–35% | Normal for cold B2B. Work on what happens after acceptance |
| Over 35% | Good. You're probably reaching people who already know your name or your company |
These bands are my reading of the benchmarks above, not an official scale. Use them as a starting point and compare against your own history: a campaign that drops from 30% to 18% means more than one that has sat at 22% for months.
Count it the boring way: accepted divided by sent, for requests sent at least seven days ago. Botdog found that 21% of acceptances arrive within an hour, 63% within a day and 88% within a week. If you include yesterday's requests in the denominator, you're measuring your patience, not your outreach.
Why people aren't accepting: five causes, most likely first
1. The list is wrong. This is the cause most of the time. People accept requests from someone whose job makes sense next to theirs. If you sell payroll software and your list mixes HR directors with students who once worked in HR, the students drag the rate down and won't buy anyway. Look at the people who did accept: what do they have in common? Rebuild the list around that.
2. The profile doesn't explain you. The recipient sees your photo, name and headline before deciding. A headline like "Helping companies grow" gives them nothing. One like "Payroll for 50–500 person companies in Germany" tells the right person to accept and the wrong person to skip, which is what you want. No photo, an empty About section, or 40 connections reads as a fake account.
3. The note is a pitch. On average a note doesn't raise acceptance at all. Belkins found 25.3% with a personalized note against 27.6% without. A note that opens with your product lowers it further. If you add one, it should give a reason to connect that is about them, not you. We covered that test in detail in should you add a note to a LinkedIn connection request.
4. Old requests are piling up. Pending invitations that nobody will ever accept still count against you. A stack of several hundred is one of the signals LinkedIn looks at when deciding whether an account is spamming. Withdraw anything older than three weeks. LinkedIn notes that after you withdraw, you can't invite the same person again for up to three weeks (LinkedIn Help), so only withdraw people you're willing to leave alone for a while.
5. You're sending too many, too fast. High volume tends to go with sloppy lists, and the rate falls. It also brings warnings. If you're near LinkedIn's weekly cap, see how many connection requests per week is safe.
A two-week fix
Raising a low acceptance rate
- Pause and measureStop new requests for a few days. Calculate acceptance only on requests older than a week.
- Study who said yesExport accepted connections. Note their titles, company sizes and countries. That is your real audience.
- Cut the listDrop everyone who doesn't match that pattern, even if it halves the list.
- Fix the headlineSay who you work with and what you do for them in one line. Add a real photo.
- Withdraw stale invitesClear requests pending longer than three weeks.
- Restart smallSend 10–15 a day to the cleaned list for two weeks, then compare against the old rate.
Change one thing at a time if you can. If you rewrite the note, the headline and the list in the same week and the rate goes up, you won't know which change did it, and you'll repeat the useless ones next time.
What not to do
Do
- Send to people whose role fits your offer
- Write a headline a stranger can understand
- Leave the note blank or make it about them
- Withdraw requests older than three weeks
Don't
- Buy a broad list to push volume
- Pitch in the connection note
- Resend straight after withdrawing
- Judge a campaign on day two
Acceptance is the cheap part
A connection that never replies is worth nothing. In Expandi's data, only 10.4% of messages sent after acceptance got an answer. So a campaign with 30% acceptance turns 100 requests into about three conversations. Pushing acceptance from 28% to 35% adds less than one. Getting the first message right after acceptance moves the result more than any change to the request.
That first message is where most people fail. They accept, the seller waits a week, then sends a pitch. Write within a day or two, while they remember why they clicked, and say something about their work instead. Our guide on what to say after a LinkedIn connection is accepted has examples.
This is the part Nitrino takes off your hands. It reads each person's LinkedIn profile and company site before writing, writes the first message for that one person, and paces invitations by the account's health: a new account starts at five a day and the ramp stops if LinkedIn pushes back. Replies land in one inbox with email, Telegram and WhatsApp, and Nitrino answers them in the account owner's voice until there's a meeting.
The short version
- The average acceptance rate is around 28%. 25–35% is normal for cold B2B, over 35% is good.
- Measure only requests older than a week. Nearly 9 in 10 acceptances arrive within seven days.
- A low rate is usually the list, then the profile. The note rarely helps acceptance.
- Withdraw pending requests after three weeks, and accept you can't resend for a while.
- Spend your effort on the first message after acceptance. That's where conversations are won.


