Picture two ways to reach the same busy executive. The first is a post in their LinkedIn feed, competing with their old colleague's promotion, a viral hot take, and forty other things scrolling past. The second is a message that arrives in their LinkedIn inbox, sitting right next to a note from their boss and a reply from a recruiter, with their name at the top and a couple of buttons to tap. Same person. Wildly different odds of being seen.
That second one is a LinkedIn Conversation Ad, and it's the format almost nobody outside of dedicated B2B advertisers is using well. It gets opened at rates a feed ad never sees - often a third to half of the people it reaches, sometimes more. It can start a real back-and-forth. And it can walk your offer straight into the room with a decision-maker who would have scrolled right past you everywhere else.
It is also the single easiest way to light money on fire on LinkedIn. So before we get into how to run one, let's be honest about when you shouldn't.
What a Conversation Ad Actually Is
A Conversation Ad is a sponsored message delivered to someone's LinkedIn inbox. It opens with a short note from a sender, and instead of a wall of text with one link, it gives the reader buttons to choose from. Tap one, and it opens the next message. Tap another, and it goes somewhere else. It's a choose-your-own-adventure built out of a few short messages and reply options, ending in a form or a link when the person is ready to act.
Two things make it different from a feed ad. First, placement: it lands in the message inbox, a place people actually check, next to conversations that matter to them. That's why open rates are so high. Second, interactivity: a feed ad shouts one thing at everyone, but a Conversation Ad can branch - one path for the person ready to book, another for the person who needs more convincing, a third for the person who just wants to look before they leap. It's less like a billboard and more like a scripted first conversation.
A simple version looks like this. The opening message names a problem your buyer has and offers a short call. Underneath it sit three buttons: "Yes, let's talk," "Tell me more first," and "Not right now." Tap the first and a form opens to book the time. Tap the second and a follow-up message answers the obvious hesitation, then asks again. Tap the third and, instead of a dead end, they get a link to a case study so the door stays open. You've built one ad that meets three different readers where they actually are - and you wrote all of it in advance.
First, the Honest Part: When It's Worth It
Here's what the case studies breathlessly selling this format leave out. You pay per send. Every message that goes out costs you whether the person opens it or not, and it's a premium placement - more expensive than putting an ad in the feed. That single fact should shape every decision you make.
It means this is a scalpel, not a megaphone. You do not blast a Conversation Ad at a broad, cheap audience and hope. LinkedIn also caps how often any one person can receive a sponsored message - the exact window shifts with how active they are - so your reachable inventory is limited by design. You literally cannot message everyone on demand.
So the format earns its cost in a specific situation: you have a tight, high-value audience - a real list of dream accounts or a precisely defined role - and a deal worth enough that a single booked conversation pays for a lot of sends. Think B2B services, software, anything where one new client is worth thousands. If that's you, keep reading. If your budget is small, your audience is broad, or your price point is low, a feed ad is the smarter place for your money, and running Conversation Ads will just drain it faster. The honest answer to "should I use this format?" is "usually no, and here's the narrow case where it's a yes."
It Lives or Dies on Who It's From
If the fit is right, the first lever - and the biggest one - is the sender. This is the name and face at the top of the message, and it decides whether the thing gets opened at all.
The instinct is to send it from your company or from a salesperson. Resist it. The message that gets opened comes from a peer - someone whose title looks like the reader's own. A marketer opens a message from another marketer. A head of operations opens one from someone who clearly does operations too. A recognizable name in your industry does even better. What kills it is a sender the reader can't place: a random rep, a generic "sales team," a title that has nothing to do with their world. People decide whether to open a message based on who it's from before they read a single word, exactly like they do with email. Pick the sender who makes your prospect think "this is someone like me," not "this is someone selling to me."
Write It Like a Message, Not an Ad
Once it's opened, you have a few short lines to earn the next tap, and the fastest way to lose is to write like an ad. The winning messages are skimmable on sight: short lines, white space between them, no dense paragraph anyone has to wade through. If it looks like work to read, it doesn't get read.
Then the content. Use their first name. Skip the throat-clearing - nobody needs "Hi, I'm the director of growth at Acme and I wanted to reach out." Nobody cares who you are yet; they care whether you understand their problem. So compare two openers. Weak: "Hi Sarah, I'm the founder of a platform that helps marketing teams with attribution and reporting across channels." Strong: "Hi Sarah - still stitching together your pipeline numbers by hand every month? That's exactly why I built this." The first talks about you. The second names her Monday-morning headache and earns the next line. Lead with their problem, in their words, and connect it to what you do in one clear line. Talk about them, not your logo wall. And end with one obvious next step, not three. There's an old rule from web usability - don't make me think - and it is the whole game here. The reader should understand what you're offering and what to do about it in about five seconds, without effort. Every extra sentence, every clever-but-vague phrase, is another reason to tap away.
Build a Conversation, Not a Dead End
The mistake most people make is treating a Conversation Ad like an email with buttons: one ask, yes or no, done. The whole advantage of the format is the branching, and if you don't build it out, you've paid a premium price for a worse email.
Map the flow on paper first, because LinkedIn's builder makes it easy to lose track. Give people more than a binary. Beyond "yes, book me" and the "not interested" button LinkedIn forces on you, add a middle path - a "tell me more" that opens a second message with a bit more detail and asks again. And give a "no" somewhere soft to land: if they're not ready to book, offer a short case study or a useful resource instead, so even a decline moves them one step closer and the send still earns something.
The best flows also handle the real objection right in the copy. Why doesn't someone book? Because they picture a thirty-minute call with a pushy salesperson. So take that off the table before they can raise it: "Fifteen minutes, with me, our founder - not a sales pitch, just a look at whether this fits." Name the reason they'd say no, and answer it, and a lot more of them say yes.
The Incentive Trick - and Its Catch
A lot of the biggest Conversation Ad wins run on an incentive: a gift card in exchange for taking a demo or a meeting. It works, and the math is less crazy than it sounds. The offer lifts the response rate so much that your cost per booked conversation can actually drop, even after you pay for the card - especially with cold audiences who have no reason to trust you yet.
But here's the catch nobody puts in the headline. People who show up for a gift card are, on average, earlier and less ready to buy than people who raise their hand with no bribe attached. One team that ran this heavily found their incentivized leads converted to real opportunities at a fraction of the rate their unpaid leads did. That doesn't make the tactic wrong - those early conversations still plant a seed, and a good demo can bring someone back months later when they're finally ready. It just means you have to measure the right thing. Track your cost per qualified conversation, not your cost per raw lead, or the cheap-looking number will lie to you. And whatever you offer, capture it with LinkedIn's native lead form rather than sending people off to a landing page - keeping them in the inbox to convert consistently beats making them travel.
A Scalpel, Not a Megaphone
Put it all together and Conversation Ads reward discipline and punish spray-and-pray. Keep the audience tight and high-value. Set your bidding to manual rather than letting LinkedIn spend on autopilot, which is almost always the pricier way to pay. Start with a small test and a rough draft of your flow, learn from how people move through it, and refine before you scale - LinkedIn even shows you a chart of where people drop off, which is the fastest way to find the weak message. And judge the whole thing on pipeline, not on opens. A 50% open rate feels great and pays no bills.
Used wrong - blasted wide, written like an ad, sent from a stranger - this format is an expensive mistake, which is exactly why so many businesses try it once and quietly give up. Used right, on a precise list of the accounts you most want to win, with a message that reads like it came from a real person who understands their problem, there is no other ad on LinkedIn that gets you into the room like this one. That's the whole trade: it asks more of you than a feed ad, and when the fit is right, it puts you somewhere a feed ad can't.
If you're weighing whether your business is one of the narrow cases where this format pays off - the right audience, the right deal size, the right offer - that judgment is a big part of what we do on LinkedIn Ads. We'd rather tell you it's not a fit than sell you sends you'll regret.
