Quick answer
Two to three LinkedIn messages after the first one. Expandi's 70,130-campaign study finds the first follow-up adds almost nothing, the second adds about 4 percentage points of reply rate, and a third adds roughly 1 more. Space them 3 to 5 days apart, then move to another channel if nothing lands.
How many times should you follow up on LinkedIn?
I run cold email and LinkedIn as one sequence for B2B clients, and this exact question, how many times to follow up before giving up, comes up on almost every account I take on. The honest answer sat in a dataset I hadn't gone looking for until this year: Expandi's State of LinkedIn Outreach H1 2026 report, built from 70,130 real campaigns run on its own platform, breaks reply rate down by exactly how many follow-up messages a sequence sent. The first follow-up changes almost nothing. The second is where the lift actually happens. A third adds a little more before the curve goes flat.
That's a specific, counterintuitive shape, and it means two or three messages after your first one is the right target for most LinkedIn sequences, not the "keep going until they block you" instinct and not the one-and-done habit either.
The shape of the data: first adds nothing, second adds the most
Expandi's own report states the finding in exactly these terms: "First follow-up = 0.6% fewer responses than no follow-up. Second follow-up = 4.05% more responses. Third+ follow-ups = diminishing returns (up to 1% boost)." Read that first line again. Sending one follow-up after your opening message doesn't just plateau, it comes in very slightly behind sending no follow-up at all, close enough to flat that it isn't worth treating as a real effect either direction. The second message is where reply rate actually moves, by a full 4 percentage points. A third message is still worth sending, since it adds up to another point, but the slope has clearly flattened by then.
| Touch | Reply rate change (Expandi, 70,130 campaigns) |
|---|---|
| First follow-up | 0.6% fewer responses than sending none |
| Second follow-up | +4.05% more responses |
| Third and later follow-ups | Diminishing returns, up to +1% more |
A second, independent dataset lands on almost the same shape. SmartReach's State of LinkedIn Outreach 2026 report, drawn from 1,800+ campaigns, 500,000+ connection requests and 460,000+ messages across 160+ businesses between January and June 2026, describes its own cohort this way: "the first follow-up added ~0, the second added +4 percentage points, and returns faded from the third touch onward." Two vendors, two different platforms, two different periods, the same curve. That's a stronger basis for a number than either study would be alone.
LinkedIn vs cold email, follow-up by follow-up
If you already run cold email, this shape probably looks wrong, and that's because it's the opposite of what email does. Saleshandy's current cold email data, 30.31 million emails across 40,006 sequences sent between July and October 2026, puts the first follow-up at 29% of all positive replies on its own, the single highest-value email in the whole sequence after the opener. Reply share then drops step by step: 15.3% on the second follow-up, 7.3% on the third.
| Touch | Cold email (Saleshandy) | LinkedIn (Expandi) |
|---|---|---|
| First follow-up | 29% of all positive replies, the peak | 0.6% fewer responses than no follow-up |
| Second follow-up | 15.3% of all positive replies | +4.05% more responses, the peak |
| Third follow-up | 7.3% of all positive replies | Up to +1% more |
Email front-loads its value onto the first follow-up and tapers from there. LinkedIn does the opposite: the first follow-up is close to dead weight, and the second one carries the sequence. Running both channels the same way, stopping after one follow-up on each, quietly shortchanges LinkedIn specifically.
Why the two channels run backwards from each other
Neither report states a mechanism, so this part is my own reading of it, not a cited finding. A cold email follow-up is still a cold, unsolicited message landing in an inbox the recipient didn't ask to receive anything in, so a prompt, specific second nudge while the first email is still fresh reads as organized rather than pushy. A LinkedIn follow-up is different because it's a message to someone who, in most sequences, has already accepted a connection request, there's a real relationship step already behind it. A single follow-up straight after that still reads as a stranger continuing to sell. A second message, especially one that adds something rather than repeating the ask, is what starts to read as a person rather than a sequence.
A three-message sequence built on this shape
Given the data, here's the structure I'd build a LinkedIn sequence around: a first message once the connection is accepted, that states one specific reason you reached out. A second message, the one doing the real work per both studies, that adds a new piece of information rather than repeating the first, a relevant resource, a specific observation about their company, or a direct question tied to a problem you can actually solve. A third, shorter message that lowers the bar for a reply, a yes or no question works better here than another paragraph. Stop there. Nothing in either dataset argues for a fourth.
What not to do. Don't copy the first message's wording into the second one with a "following up" line added on top. Both studies measure a follow-up that exists, not what's actually inside it, and a repeated message is the fastest way to turn your one useful follow-up into the flat one.
How many days between messages
SmartReach's report states a figure for this directly: "the best-performing gap between touches was 3 to 5 days." That's close to how I'd space it anyway, close enough that the first message is still a live memory, far enough apart that it doesn't read as automated. Weekends count against you here the same way they do in cold email, a message sent Friday afternoon effectively waits until Monday to be read, so the real gap often runs closer to the long end of that range once a weekend sits inside it.
What goes in the second message, the one doing the work
Since the second message is the one both studies credit with the actual lift, it's worth more care than "just checking in." The version that tends to earn a reply adds something the first message didn't have: a specific reason their company, role or a recent move of theirs makes the original point relevant, a short, concrete resource, or a narrower version of the ask that's easier to say yes to than the first one was. The common thread across all three is that the message gives the recipient a new reason to respond, not just a second chance to respond to the same reason.
Does the right number change by who you're messaging?
The follow-up count studies here don't break results down by who's being messaged, but a separate finding is worth layering on top. I've written in more depth about how LinkedIn acceptance rate changes by seniority, where a 2026 study found C-suite recipients accepting connection requests at roughly half the rate of individual contributors. That gap doesn't change how many follow-ups to send, but it does change what to expect from them: a senior list running the exact same three-message sequence as a junior one will land fewer total replies at every step, which isn't a sign the sequence itself needs fixing.
When to stop and move to another channel
Three messages, then stop on that channel specifically. In my own practice, email and LinkedIn run as one sequence rather than two separately-owned campaigns, so when a LinkedIn thread goes quiet after the third message, the move isn't a fourth LinkedIn message, it's letting the email side of the sequence carry the relationship instead, or vice versa if LinkedIn was the second channel. What not to do is covered in more depth in what to do when a LinkedIn prospect accepts and goes quiet, which is the right next read if your third message has already gone unanswered.
How much to trust two vendor datasets
Both Expandi and SmartReach sell LinkedIn outreach software, the same situation as most benchmarks in this space, so their own customers' behavior shapes what each dataset actually measures. That's a real limitation, campaigns run through either platform aren't a random sample of all LinkedIn outreach everywhere. What makes the finding worth citing anyway is that it isn't each vendor's own marketing claim, it's a specific, unflattering-to-nobody shape (the first follow-up barely working) that shows up independently in two platforms' own data at different times of the year. If one study had shown this and the other hadn't, I'd treat it as platform-specific noise rather than a real pattern.
The one-and-done mistake this data argues against
The most common version of this mistake isn't sending zero follow-ups, it's sending exactly one and stopping, which both reports suggest is close to the worst place to quit given the actual shape of the curve. A single follow-up gets you statistically nothing beyond the connection itself, and then the sequence ends right before the step that would have moved the number. If you're only going to send one follow-up message on LinkedIn, per this data, you'd be better off sending none and saving the effort for a second message done well, though sending both is still the better option over either alone.
Key takeaways
- Send two to three LinkedIn follow-ups after your first message, not one and not an open-ended number.
- The first follow-up barely moves reply rate, per Expandi's 70,130-campaign data: 0.6% fewer responses than sending none.
- The second follow-up is where the real lift is, +4.05% per Expandi, +4 percentage points per SmartReach's independent 500,000-request dataset.
- Cold email runs the opposite curve: Saleshandy's current data puts 29% of all positive replies on email's first follow-up alone.
- Space LinkedIn messages 3 to 5 days apart, per SmartReach, and stop after the third if nothing lands.
FAQ
How many times should you follow up on LinkedIn?
Two to three messages after your first one. Expandi's 70,130-campaign study finds the first follow-up adds almost nothing to reply rate, the second adds roughly 4 percentage points, and a third adds about 1 more before returns flatten out.
Does the first LinkedIn follow-up actually matter?
Barely, on its own. Expandi's own data states it as 0.6% fewer responses than sending no follow-up at all, essentially flat. It still matters as the step that leads to the second message, which is where the real lift shows up.
How many days should you wait between LinkedIn follow-up messages?
Around 3 to 5 days, per SmartReach's 2026 State of LinkedIn Outreach report, built on 1,800+ campaigns and 500,000+ connection requests. Faster than that reads as pushy, slower and the first message has usually been forgotten.
Is LinkedIn's follow-up pattern the same as cold email's?
No, it runs backwards. Saleshandy's current cold email data, 30.31 million emails across 40,006 sequences, puts 29% of all positive replies on the first follow-up alone, the single highest-value email in the sequence. On LinkedIn, that same first follow-up is close to flat, and the second message is what moves the number.
When should you stop following up on LinkedIn?
After your third message, since neither dataset behind this article finds a fourth message worth the research to confirm it helps. Stop there and let the other channel in your sequence, usually email, carry the relationship instead of sending a fourth LinkedIn message into a conversation that has already said what it's going to say.