LinkedIn outreach examples that get replies

Take any opener that works and ask the only question that matters: what did the sender have to know, and choose to notice, to write this? A strong first line like a reference to a specific post the prospect wrote last week is not good because of its phrasing — it is good because it proves the sender looked at *this person* and picked a detail no template could contain. The lesson is not “mention their post.” The lesson is “find the one true, specific thing that signals you are not running a query,” which will be different for every prospect.

This is why lifting openers verbatim fails. When you copy “I loved your recent post about X,” you inherit the words but not the act of noticing, and prospects have a finely-tuned sense for the difference — the second an opener could have been sent to anyone, it reads as automation. The teardown skill is learning to see the *category* of the hook (recent activity, a career move, a shared connection, a specific line in their bio) and then go find your own instance of that category in your prospect’s actual profile.

The reason this matters more every year is that the good templates get around. A phrasing that worked brilliantly for one seller gets shared, copied, and burned into thousands of inboxes until prospects recognize it on sight. Reasoning does not decay that way, because reasoning produces a fresh, person-specific message every time. This is exactly why TopClozer drafts from the profile rather than from a shared template library — Claude reproduces the *act of noticing* per prospect, so what lands is the substance a great example demonstrates, not a phrase the recipient has seen ten times this month.

The second thing a teardown should isolate is the ask, because it is where most messages that got the read still lose the reply. Weak asks are expensive: “can we hop on a 30-minute call to explore synergies” demands time, commitment, and a decision from someone who owes you nothing. Strong asks are cheap to say yes to — a single specific question, a low-stakes yes/no, an offer that requires almost nothing from the prospect. When you study a reply-worthy example, measure the ask by how little it costs the recipient to respond, and you will notice the good ones consistently shrink the next step rather than inflate it.

Follow-ups are the other half of the teardown, and they reveal the most about whether a sender understands the game. A weak follow-up says “just bumping this” or “circling back,” which adds nothing and quietly signals that the sender is working a list. A strong follow-up brings a new reason to reply — a relevant resource, a fresh angle on the problem, a proof point — so each touch stands on its own even if the prospect never saw the last one. Line up a good sequence and you will see every touch carries independent value; line up a bad one and you will see the same ask re-sent with rising impatience.

Once you can see these patterns, the natural next question is how to produce them consistently, and this is where structure plus AI beats both raw templates and pure hand-craft. Use the teardown to internalize the *structure* — specific opener, tiny ask, value-additive follow-ups — and let Claude write each instance from the prospect’s real profile in your voice. You keep the reasoning and the tone; the model handles the per-person execution. That is how you adapt rather than copy: the framework is yours, the words are freshly generated for each human, and nothing you send is a phrase the recipient has already learned to ignore.