By Across 2,511 messages Linkdify sent on behalf of users in August and September, the median reply arrived about 14 hours after the message went out. Most of the replies that were ever going to come, came within the first day. That one number quietly breaks the follow-up cadence almost every founder copies into their first LinkedIn campaign.
You know the one. Seven touches over three weeks. Bump on day 3, "just floating this to the top" on day 7, a case study on day 10, a breakup message on day 21. It is the default in every sales playbook, and it was built for a channel where it actually works. That channel is email.
The seven-touch cadence is an email artifact
Email follow-ups work because email is lossy. Your first message lands under 200 others on a Tuesday morning. It gets filtered to Promotions. It gets skimmed on a phone and forgotten. Touch five converts in email because touches one through four were often never seen at all. Persistence is a workaround for a bad delivery channel.
LinkedIn messaging is a different shape. The inbox is short. Most people get a push notification on their phone. Your message sits at the top of a list of maybe a dozen threads, under your name and your face. If the person opened the app at all, they saw it. The 14-hour median says exactly that: people see the message quickly, and they decide quickly.
So when you send touch five on LinkedIn, you are not resurfacing something that got buried. You are reminding someone of a decision they already made, and asking them to make it again in your favor. That is a worse ask than the first one.
Silence after two days is an answer
When a prospect goes quiet on LinkedIn, they fall into one of three buckets. They are not interested. They are interested but not right now. Or they genuinely did not see it, which on LinkedIn mostly means they are a low-activity user who opens the app twice a month.
Only the third bucket is helped by a nudge, and it is the smallest. The first bucket is not coming back no matter how good your case study is. The second bucket is the interesting one, and a "bumping this" message does nothing for them, because it repeats the ask they already declined to answer. What moves them is a different ask. Smaller, or about a problem they actually have this quarter.
I used to read a non-reply as "didn't land yet." I now read it as "landed, and the answer was no for now." That reframe changed how I spend sends.
Want LinkedIn outreach that runs itself?
Linkdify finds your leads, writes personalized messages, and sends them at safe, human-paced limits.
What I changed in how Linkdify runs follow-ups
The builder still lets you stack as many follow-ups as you want, because that is what people ask for. Watching the data, the long tails were mostly dead sends. They burned daily message capacity and, worse, they trained prospects to see the sender as noise. On a channel where your account's reputation is the whole asset, that is expensive.
Two things changed. Follow-ups now read the thread before writing the next message, so a second touch never repeats the first one. And the cadence I recommend to every new user is short. One message after the connection is accepted. One follow-up about two business days later that changes the ask. One last note a week or more after that, or nothing. Three touches, then the lead is done, and the daily budget goes to someone who has not heard from you yet.
When you send matters more than how often
If the reply window is a day, the hour you send in is a real lever. A message that lands at 11pm in the prospect's time zone sits under whatever arrives before they wake up. A message that lands mid-morning on a workday gets the top slot while they are already in work mode.
I want to be honest about what I can and cannot claim here. We added per-campaign business-hour windows so sends only go out during working hours, and anecdotally it helps. I do not yet have a clean, published number on reply rate by send hour, so I am not going to invent one. Keep sends inside working hours because it is cheap and obviously sensible, and I will publish the number when I have it.
The arithmetic that makes this obvious
Here is why the short cadence wins even before you argue about message quality. Across 8,400 connection requests, the average acceptance rate was 25.2%, roughly one in four. Across the messages that followed, the reply rate was 14.6%. Your constraint on LinkedIn is never persistence. It is the daily cap on how many new people you can reach without tripping limits.
Every follow-up you send to a dead thread is a send you did not use on a new person, and new people convert at a known rate. Touch six on a prospect who has ignored you for three weeks converts at something I cannot measure because it is so close to zero. Trade those sends for fresh connection requests and the math takes care of itself.
One more thing from the same dataset, because it is the same lesson. Requests sent with no note were accepted 48.6% of the time. Requests with a note, 8.6%. The extra effort was not neutral. It hurt. More touching is not more selling, on either side of the connection.
The full benchmark is published at linkdify.io/reports/linkedin-outbound-benchmarks if you want to check my numbers.
Want LinkedIn outreach that runs itself?
Linkdify finds your leads, writes personalized messages, and sends them at safe, human-paced limits.
The takeaway
If someone has not replied in two days on LinkedIn, treat that as their answer. Send one follow-up that asks something smaller, then move on. Build your cadence for the people who have not heard from you yet, because on this channel they are the only ones still deciding.