"View their profile before you connect" is one of those LinkedIn tips everybody repeats and nobody measures. When we added warm-up touches to OnlyLeads, we built the measurement in at the same time, and ran it on our own outreach first. These are the real numbers from three of our sending accounts: over two thousand connection requests and about twelve hundred follow-up messages.
What the touches are
All of them are off until you switch them on for a sending account:
- Pre-send warm-up. The connection request waits two days. On day one we view the person's profile, visibly, so they see who looked. At least twelve hours later we like their most recent post. On day two the request goes out.
- A like while the request is pending, two to fourteen days after it was sent.
- A like while waiting for a reply, three to twenty-one days after a follow-up message that hasn't been answered.
- A profile view on the third acceptance check, about a day and a half after the request.
Likes are careful. We only like original posts from the last thirty days, never reposts, and never anything that mentions politics, religion, layoffs, illness, bereavement or culture-war topics, or that was written by one of your competitors. If there's no suitable post, there's no like. Every touch counts against per-account daily limits: forty profile views and twenty likes.
The results
For each touch we compare people who got it with people who didn't, counting only sends at least a week old so the outcome has had time to happen.
- Pre-send warm-up: helps. 62% of warmed-up connection requests were accepted, against 51% without (143 requests vs 2,085). It went the same way on all three accounts.
- A like while waiting for a reply: helps, a lot, with a caveat. 21% of people we liked replied within two weeks, against 3.4% of people we didn't (150 vs 1,022). Again consistent across accounts. The caveat is real, though: we can only like someone who has posted recently, and people who post are more active on LinkedIn, so some of that gap would be there anyway.
- A like while the request is pending: no effect. 51% vs 52%, and the three accounts disagree with each other. We're keeping it as an option but we wouldn't switch it on for the results.
The number that looks terrible
People who got a profile view on the third acceptance check accepted only 8% of the time, against 57% for everyone else. Taken at face value, that says the view destroys your acceptance rate.
It doesn't. The view only goes to people who haven't accepted yet a day and a half after the request. Everyone who accepted quickly, which is most of the people who were ever going to, is automatically in the other group. The comparison is rigged before the touch happens. We left the number in the product because hiding inconvenient numbers is how you end up believing convenient ones, but it can't tell you whether the touch works.
This is observational data, not an experiment. Accounts switched touches on at different times. Read it as a direction, not a law.
See your own
The same comparison is on your analytics page, per sending account, as soon as each group has at least ten people in it. Every touch is also on the lead's timeline and in a feed of all touches, so you can see exactly what was done in your name. Nothing about it is hidden from you, including the results that don't flatter the feature.