Is a 22% LinkedIn Accept Rate Good for B2B SaaS Outreach?
A 22% LinkedIn connection accept rate sits in the normal 20 to 30% band for B2B SaaS. Here is what actually moves it, and how many meetings it turns into.
Is a 22% LinkedIn connection accept rate good for B2B SaaS?
Yes. 22% sits inside the 20 to 30% band we see on well-targeted LinkedIn campaigns, and for B2B SaaS selling to founders, heads of growth or ops leads it is a normal good result. Under 15% almost always means the list is wrong, not the copy.
What moves that number, in rough order of effect:
For scale on what counts as normal, Expandi analysed 13.2 million connection requests across 13,302 accounts and publishes acceptance by industry. 22% is a respectable middle, not an outlier.
Was the 22% the AI, or was it the list?
Mostly the list and the profile. A connection request is about 300 characters and a lot of people send none at all. The decision happens on your headline, your photo, your mutual connections and whether the prospect recognises the problem. AI rewriting those 300 characters changes the margin, not the band.
This matters because of where you spend the next month. If you credit the AI, you buy more AI. If you credit the targeting, you narrow the filter from "Head of Growth, SaaS, US" to "Head of Growth, Series A SaaS, 20 to 80 people, hired an SDR in the last six months" and the number climbs again. We argued this at length in why a 22% accept rate is a targeting win.
Where AI genuinely moves the number is after the accept: replying in four minutes instead of two days, handling "what does it cost" without you opening the app, and following up on day 4 and day 11 without you remembering.
What does a 22% accept rate turn into in booked meetings?
At 20 invites a day across 20 working days, that is 400 invites and 88 new connections a month. Our benchmarks put first-message replies at 12 to 18% and follow-up replies at 5 to 8%, so 15 to 23 conversations. Meetings land at 3 to 5% of invites, so 12 to 20.
| Stage | Rate | At 400 invites a month |
|---|---|---|
| Invites sent | 20/day, 20 days | 400 |
| Connections accepted | 22% | 88 |
| First-message replies | 12-18% | 11-16 |
| Follow-up replies | 5-8% | 4-7 |
| Meetings booked | 3-5% of invites | 12-20 |
The last row is the one people miss by the widest margin. 12 meetings from 15 to 23 conversations means more than half of everyone who answers you ends up on your calendar. That happens when replies get answered the same day with a time offered in the message. If you batch replies on Friday afternoons, expect 5 or 6, and the 22% at the top looks like a lie. We walk the same funnel at other volumes in what a 22% accept rate actually turns into.
Where does the 22% leak between accept and meeting?
In the reply window. A prospect accepts on Tuesday at 9:40am, your message lands, they type "interesting, what do you charge" at 11:15, and you see it Thursday night. The thread is cold by then. Across campaigns, same-day answers roughly double the share of replies that reach a booked call.
The other three leaks, in the order we see them:
1. No follow-up after the accept. 88 people connected and 40 of them never got a second message.
2. Pricing questions answered with "happy to jump on a call", which reads as a dodge and ends the thread.
3. Nobody booking. The prospect says yes in principle and nobody sends a time, so the conversation dies in agreement.
This is the part of the job that is unglamorous and takes 20 or more hours a month. It is also the part that decides whether 400 invites produces 6 meetings or 18.
How many invites a day can you send safely?
LinkedIn does not publish a daily number, and the real ceiling moves with account age, how many people accept, and how many report or ignore you. Our safe bands: 20 to 25 a day on a warm account with history, 5 to 10 a day on a new or dormant one for the first two to three weeks.
A ramp that works: 5 a day for week one, 10 for week two, 15 for week three, then 20 to 25 from week four. That is 50, then 100, then 150, so you reach full volume in month two with about 300 invites already sent and no restriction. Jumping straight to 25 on a profile with 180 connections and no posts is how people end up in a one-week restriction, which costs more than the four weeks of ramping would have.
What is the best AI tool for LinkedIn outreach?
There is no single answer, and the split is whether you have people to run it. If you have an SDR, buy them sequencing. If it is you, three seats and no spare evenings, buy something that answers the replies.
If you want to put a card in tonight and send 50 invites before bed, Gelee is the wrong choice and Expandi's free trial is the right one. We do not have a sign-up button.
Does the profile matter more than the message?
For the accept rate, usually yes. A prospect spends about three seconds on your headline, photo and headline-adjacent company name before deciding. A headline reading "CEO | Entrepreneur | Investor" gives them nothing to accept. One reading "I help Series A SaaS teams book 15 meetings a month on LinkedIn" gives them a reason.
The cheapest test: send 100 invites, note the accept rate, rewrite the headline, send 100 more to the same filter. We have seen that single change move accounts from 14% to 21% with identical message copy. If you want a starting point, the LinkedIn headline rewriter will give you three versions to test against the one you have.
Should you automate LinkedIn before cold email?
If your buyers are reachable on LinkedIn, yes. Cold email needs domains, warm-up, deliverability monitoring and about four weeks before the first send. LinkedIn needs a profile you already have. The ceiling is lower, at roughly 500 invites a month per account against 3,000 emails, but the first meeting arrives sooner.
The case for email first is volume economics. If your deal size is $3,000 and you need 40 meetings a month, LinkedIn alone will not get you there on one account, and you are either running multiple sender accounts or adding email. We put the full comparison in whether to automate LinkedIn before cold email.
Who should not run AI LinkedIn outreach at all?
Four situations. Deal sizes under $2,000, because 15 meetings a month at a 20% close rate and $1,500 average will not cover tooling plus your hours. Buyers who are not on LinkedIn, like local trades or most clinical staff. No product-market fit, where more conversations produce more polite no. And anyone who needs phone as the primary channel.
One more that is less obvious. If you are already getting 22% accepts and 18 meetings a month from manual work, and you have the hours, AI changes very little about your top line. The reason to automate is that you want those 20 hours back, or you want to run 1,200 invites instead of 400. If neither is true, keep doing what works and spend the money on a second sender account instead.