We raised LinkedIn connection acceptance on one program from 8.4% to 15.2% by changing one thing, which was who we sent requests to. The campaigns, the sequence and the copy stayed the same. We only sent requests to people who are active on LinkedIn, and we dropped everyone else.
We keep our other LinkedIn numbers on our cold email and LinkedIn statistics page. This post covers the activity rule, how we build a campaign, and how we write the connection note.
What is the LinkedIn activity rule?
The rule is a short filter we apply before anyone enters a campaign. A person qualifies if they have 500 or more connections. They also qualify with 200 or more connections plus at least 3 posts in the last 12 months, or at least 2 posts in the last 30 days.
Our reasoning is that someone who rarely opens LinkedIn is less likely to see our request, so it sits unanswered. We think someone who posts and has a large network checks the app more often.
How much did the activity rule change our results?
We ran the same campaigns and the same copy for two periods of two weeks each. Only the list changed between them. In the first period we sent 464 connection requests, and 8.4% were accepted. In the second we sent 520 requests, and 15.2% were accepted.
Replies moved the same way. The reply rate went from 6.3% to 13.3%. We think the rule caused the lift because nothing else changed, though a program this size is a small sample.
When LinkedIn volume looks thin, we do not loosen the rule to get more sends. We check a bigger pool of people, or we add more people who finished the email sequence. That keeps the quality of the list steady while the volume grows.
Why do we check where people work first?
Before a list enters a campaign, we check each profile live to see if the person still works at the target company. On one list, 33 of 212 pending people, or 16%, should never have been queued. They had left the company, the profile belonged to someone else, the link was dead, or they were open to work or on a contract.
A stale list also costs us replies from people saying they no longer work there. We deprioritize profiles with fewer than 100 connections instead of removing them. A small profile may be the only person who sees the message at that company.
How do we structure a LinkedIn campaign?
Each campaign has two variations only. One sends a connection request with a note, and the other sends a blank request. If we want to test a soft note against a direct one, we run that as an A/B test inside the note variation instead of building a separate campaign.
The blank variation needs a first message that does not sound like a connection request. We take the note text and rework it a little. A line like "I've been looking to connect with" becomes "I work with", and a statement at the end becomes a question. Then it runs the same follow-ups as the note variation, moved down by one message.
The note variation never gets a third follow-up. Its last message is a test between an offer and a soft breakup. Every follow-up has to read correctly after any version of the note, since we cannot know which one the person saw.
How do we write the connection note?
We keep a connection request to 270 characters or fewer, since the platform limit is 300. Direct messages stay at 360 characters or fewer. Short messages get read on a phone without a second tap, which is why we stay short.
We write to the one person, using what we know about their company, such as their products, claims or certifications. We do not open with "most teams I talk to". When we open on something we found, we soften it as an observation, such as "looks like you're making" or "saw you're", because stating it bare sounds like an assumption.
Then we end the note when the point ends. We cut lines such as "always good to know more people in your field" or "great to have you in my network", since they add nothing. The close has to grow out of the message itself, so we either offer the thing we described or ask the question it raised.
A light question like "Open to swapping notes?" only works when we have something to swap, like data we pulled on sites like theirs. We never paste one stock ending into every message. Every message in the sequence ends on an offer, and the offer matters more than any phrase in the note.
Common questions
What LinkedIn connection acceptance rate do we see?
We can only share our own numbers. One program went from 8.4% of 464 requests accepted to 15.2% of 520 after we limited the list to active people. That is a small sample.
Does a LinkedIn connection request need a note?
We run both in every campaign. One variation sends a request with a note, and the other sends a blank request followed by a reworked first message.
How long is a LinkedIn connection note?
We keep a connection note to 270 characters or fewer, since the platform limit is 300. Our direct messages stay at 360 characters or fewer.
Who do we leave out of a LinkedIn campaign?
We leave out people who are not active on LinkedIn. A person qualifies with 500 or more connections, or with 200 or more connections plus recent posts.
