Everyone's Doing LinkedIn DMs Now. Cold Email Still Wins More Often.

You've had this argument in your own Slack. Someone posts a LinkedIn DM screenshot with a 12% reply rate and says cold email is basically over. Someone else points out their cold email program still closes half the pipeline. Nobody has real data, everybody has a strong opinion, and the conversation goes nowhere. Salesforge just ran the actual survey. 425 B2B operators, asked directly which channel performs best. The result isn't the clean 'LinkedIn won' story the DM crowd wants, and it isn't the 'cold email is still king' story the old guard wants either.

Cold email vs LinkedIn: the question every SDR team is asking

Here's what's actually driving the argument, if you strip away the vibes. LinkedIn DMs feel like they're working better right now because when they work, they work obviously: a real reply from a real profile, in a channel that feels more personal than an inbox full of cold pitches. Cold email's failures are louder too, more bounces, more spam folder complaints, more of the deliverability grinding we've written about all year. So the felt experience skews toward 'LinkedIn is winning,' even before anyone runs a real comparison.

Salesforge's survey is the first real attempt this year I've seen to put a number on the actual question instead of the vibe. 425 B2B operators, asked which channel performs best for them, plus 28 qualitative comments giving color on why. It's a real, dated, named survey, published July 27, 2026, not another vendor blog post repeating the same 2024 stat everyone's tired of.

What 425 operators actually said

LinkedIn DMs took the plurality: 37% of operators named it their best-performing channel. Cold calling came in at 31%. Cold email at 26%. Everything else, Telegram, inbound, in-person, split the remaining 6%. If you stop reading right there, the headline writes itself: LinkedIn wins, cold email is fading, update your channel mix accordingly.

But stopping there misses the qualitative note attached to the LinkedIn number, which the survey itself flags: it's the 'highest performer per message sent, hardest to scale past one person.' That's not a footnote. That's the entire ballgame.

The number the vote doesn't show

A best-performing-channel vote answers 'which channel gives you the best odds per message.' It doesn't answer 'which channel can actually produce the volume of pipeline your number requires.' Those are different questions, and the same survey has the data to answer the second one, buried under the headline vote.

Run the arithmetic on what that actually means for a team of five reps needing real monthly pipeline, not a single founder sending a hundred thoughtful DMs a week. LinkedIn's cap is per profile, and a company can't just spin up twenty more LinkedIn profiles the way it can spin up twenty more sending domains, without tripping every platform-abuse detection LinkedIn runs. Cold email's reply rate per message is the lowest of the three, but it's the only channel on this list built to scale linearly with headcount and infrastructure instead of being bottlenecked by a single human's daily attention and a platform's per-profile limits.

“The channel that wins the vote isn't always the channel that can produce the volume your pipeline number actually needs.”

Why the 'per message' framing wins the internet argument

It's worth being honest about why LinkedIn's per-message number travels so much further on social media than cold email's does. A 12% LinkedIn reply rate makes for a great screenshot: a real person's real reply, in a channel where fewer people are sending, so the response feels earned and personal. A 2% cold email reply rate makes for a terrible screenshot, even when the absolute number of replies it produces, at ten times the daily volume, ends up higher. Nobody posts 'I sent 250 emails today and got 8 replies' the way they post 'this one LinkedIn message got a reply in twenty minutes.' The format of how these results get shared online systematically favors the channel with the better per-message story, regardless of which channel actually filled more of the pipeline that quarter.

That's not a criticism of anyone posting their LinkedIn wins. It's a reason to be suspicious of channel-mix decisions made by scrolling a feed instead of running the actual math for your own team's size and quota. A five-person SDR team that shifts its entire outbound budget toward LinkedIn DMs because of a viral screenshot is about to discover, the hard way, that five reps can't run five times the LinkedIn volume the way they can run five times the cold email volume, because the platform's per-profile cap doesn't move just because you hired more people.

Cold email vs LinkedIn: so which one do you actually run

It depends entirely on what you're solving for, and the honest answer requires admitting that 'best channel' isn't one question. If you're a founder or a single AE running your own outbound with no team to scale, LinkedIn's higher per-message reply rate is the better bet, because you're already capped at one person's daily output either way, and you might as well spend that limited daily volume on the channel with better odds per message. That's exactly the profile of operator who's loudest about LinkedIn working, and they're not wrong, for their situation.

If you're building a program that needs to scale past what one person can send by hand, cold calling and cold email lead generation are the two channels built for that, and cold email specifically scales the most cheaply, since adding sending capacity is a domain and infrastructure decision, not a hiring decision the way scaling cold calling is. That's not a knock on cold calling, which posted a genuinely strong 31% vote share and a healthy connect rate range, it's a cost-structure observation: a rep costs a full salary, an extra sending domain costs a fraction of that.

There's a deliverability angle here too, one that rarely makes it into the LinkedIn-versus-email debate. Cold email's reply rate ceiling isn't fixed the way LinkedIn's per-profile message cap is. It moves with how well you manage sender reputation, list hygiene, and deliverability fundamentals. A team running disciplined infrastructure can push its reply rate meaningfully above the 1-5% range this survey reports as typical, while a LinkedIn profile capped at 20-25 messages a day has a hard ceiling that better copy alone can't lift much higher. That's an underrated argument for cold email specifically: it's the channel with the most room to improve through operational discipline, not just the channel with the lowest floor.

The real answer: both, on purpose

The most useful comment buried in Salesforge's 28 qualitative responses wasn't a vote for any single channel. It was operators describing a layered approach, cold email for volume and top-of-funnel coverage, LinkedIn DMs for warming and multi-touch on accounts that matter most, cold calling for the moment a prospect shows real signal. That's not a cop-out answer. It's the same logic behind hub-and-spoke content architecture applied to outbound: different channels reaching the same buying committee at different moments, not one channel declared the permanent winner. A prospect who ignores three cold emails might still pick up the phone on the fourth touch, or reply to a LinkedIn note referencing the exact page they visited after the email landed. The channels aren't really competing for the same interaction. They're covering different moments in the same buyer's path, and treating the survey as a horse race misses that entirely.

Stop asking whether cold email is dead. It isn't, the data says 26% of operators still call it their best channel, right behind two others that have their own real constraints. Ask instead which channel matches the volume you actually need this quarter, and build the mix, rather than betting the whole program on whichever channel had the loudest LinkedIn post about it this month. The teams still winning a year from now won't be the ones that picked a side in this argument. They'll be the ones running all three, deliberately, for the specific job each one is actually good at, and reviewing the mix quarterly against real numbers instead of whatever channel felt hottest in last week's Slack thread.

See where you are cited today

A free snapshot audit of your rankings and AI citations before we ever talk.