LinkedIn Outreach Copy: 7 Message Frameworks That Get Replies

Why the copy still decides, even with a good list

I'm Hlib Storchak. I build outbound systems for B2B founders and sales teams, and between cold email and LinkedIn I've helped book 2000+ meetings for B2B clients. The question I get most about LinkedIn is some version of "what should I actually say." Not targeting, not tooling, the words in the message. That's fair, because a well-targeted list sent with lazy copy still underperforms a smaller list with copy that reads like a real person wrote it for that specific reader.

The seven frameworks below aren't seven equally good options to pick at random. They're structures for seven different situations, depending on how much genuine context you have on the person you're writing to. Use the wrong one for the situation, HEART on a total stranger with no shared context, or a bare direct ask on someone who's never heard of you, and the structure won't save you. Match the framework to what you actually know, and the copy stops being the hard part.

The one number that should shape every framework below

Belkins' 2026 LinkedIn Outreach Study, built from 11.5 million connection requests sent in 2025, found that requests carrying a personalized note landed a 55% higher reply rate than blank requests: 8.2% versus 5.3%. The blank request actually pulled a marginally higher acceptance rate, 27.6% versus 25.3%, so a note isn't free. It costs a little acceptance and buys a lot more reply. Source: Belkins, LinkedIn Outreach Benchmarks 2026.

Every framework below is really just a different way of writing that personalized note well enough to be worth the small acceptance-rate cost. If you take one thing from this article and skip the rest, take this: a note that says something true and specific about the person you're writing to is doing most of the work, before you've even picked a structure.

The seven frameworks, at a glance

Ranked here by how much they depend on genuine context you already have, from "you know something real about this person" down to "you know nothing except their title." No independent study ranks these seven head to head, so treat the order as a practical guide built from what the personalization and message-length data above actually supports, not a scored benchmark.

1. HEART: hook, empathy, authority, request, thanks

HEART is a five-part structure documented by PhantomBuster: a hook that shows you actually looked at something specific (a recent post, a shared group, a company announcement), a line of empathy that shows you understand a real challenge in their world, a brief note of authority (your role, or results with similar clients, kept short), a small and specific request, and a short, gracious close. Source: PhantomBuster, LinkedIn Outreach Message Frameworks.

Example shape: "Hi [name], saw your post in [group] about [specific challenge], that onboarding gap is one I hear from a lot of teams your size. I've worked through similar setups with a few other B2B teams. Would you be open to connecting to compare notes? Either way, thanks for the read." This is my default structure when I have exactly one real, specific detail to work with and want the note to read as researched rather than templated.

2. The mutual-connection opener

Naming someone you both genuinely know, "[Name] suggested I reach out" or "we're both connected to [name] from [context]", is the single fastest way to move a stranger's read of your message from "cold pitch" to "someone in my network vouched for this." It only works when the connection is real and, ideally, the other person is comfortable being named. Fabricating or exaggerating a mutual link is the fastest way to lose the trust this framework is built on the moment the prospect checks.

3. The trigger-event opener

A trigger event is a public, timely, specific change: a new role, a funding round, a hiring push, a product launch, an award, a conference talk. It gives you a real, current reason to reach out that has nothing to do with your pitch yet, which is exactly why it lowers the reader's guard. "Congrats on the [role/round/launch], curious how that's changing [specific priority]" reads as timely rather than templated, provided the event is recent enough that name-dropping it still feels current, not like you're three weeks behind.

4. The question-first opener

Leading with a genuine, specific question ("How are you handling [specific process] now that [context]?") can outperform a statement because a question invites a reply by design, rather than requiring the reader to generate one from nothing. The catch: it only works if the question is real and answerable in one line. A vague or obviously rhetorical question ("Struggling with pipeline?") reads as bait immediately and gets ignored or reported at a higher rate than almost anything else on this list.

5. Value-first, no ask

This one hands over something useful, a relevant resource, a specific observation about their public content, a genuine compliment tied to something real, with no request attached in the first message at all. It's the slowest framework to convert because there's nothing to reply to except "thanks," but it builds a different kind of trust than the other six, and it's the one I'd reach for with a short list of accounts worth a longer play rather than a large batch running at volume.

6. Peer-proof, "teams like yours"

Referencing similar teams or a believable pattern you've seen ("I've noticed a few [industry] teams around your size dealing with [specific issue]") borrows credibility from pattern-matching rather than a named mutual connection. It's genuinely useful, but it's also the framework most likely to drift into a disguised pitch if you're not careful, since "teams like yours" is one sentence away from "and here's my product." Keep the observation specific and resist tacking on the pitch in the same message.

7. The direct, no-fluff ask

Sometimes the honest move is the short move: state who you are, what you do, and ask directly, in one or two sentences, no hook. This is weakest cold, on a stranger with zero shared context, but it's often the right call once an account already knows your name, from a previous touch, a shared piece of content, or an earlier reply, because at that point the extra framing just adds length without adding trust. Given that shorter messages consistently test better than long ones across the benchmarks cited above, this is also the fallback worth defaulting to any time you're unsure which of the other six actually fits.

Where copy actually moves the number: note vs message

Copy quality doesn't matter equally everywhere in the sequence. Expandi's 2026 benchmark, built from 13.2 million connection requests and 6.7 million outbound messages sent through its platform between May 2025 and April 2026, found connection-note reply rates falling from 3.5% to 2.2% over the year, a real decline, while reply rates on messages sent after the connection is accepted held steady at 10 to 11%. Source: Expandi, LinkedIn Outreach Benchmarks 2026.

That gap is the practical takeaway: the connection note is a filter, not your pitch, so keep it short and specific rather than trying to cram a full framework into it. Save the fuller version of whichever framework fits, HEART, trigger-event, value-first, for the first message after acceptance, where the reply rate is both higher and more stable, and where the reader has already agreed to be in your network.

Mistakes that sink all seven frameworks equally

The mistake I see most often when I take over a client's LinkedIn account isn't a bad framework choice, it's a good framework applied to a fake detail: a "mutual connection" who barely knows either party, a "trigger event" that happened two months ago and no longer reads as current, a "peer-proof" reference to results nobody can verify. Every one of these seven frameworks depends on the specific detail inside it being true. The structure is the easy part. The other common failure is running the same framework, unchanged, across an entire list segment, which turns a genuinely good structure into a template the moment two people in the same company compare notes.

Which one I reach for, and when

For a cold list with decent public signal, I default to HEART or the trigger-event opener, because both force you to find one real detail before you write anything, which is most of the work anyway. When there's an actual mutual connection worth naming, that opener wins outright, nothing else on this list earns trust as fast. For accounts that already know my name from a prior touch, I drop straight to the direct ask, since re-explaining context at that point just adds length. Value-first is the one I keep for a short, high-value account list where I'm willing to trade speed for a stronger first impression. Whichever one I pick, the note itself stays short, and the fuller version of the framework moves to the first message after the connection is accepted, where Expandi's data says the reply rate actually holds up.

FAQ